ProgrammableMarketing/runparagon.com (opens in a new tab)
PDF

Views

  • Plan
  • Roadmap
  • Research

On this page

  • The brief
  • Start here
  • Idea portfolio
  • Effort vs impact
  • This week · 7
  • This quarter · 11
  • Big bets · 3
  • Wildcards · 2
  • How we can help
Generated Nov 10, 2026
  1. Programmable
  2. Playbooks
  3. Paragon
Paragon logo

growth playbook · Nov 10, 2026

runparagon.com (opens in a new tab)

Paragon growth playbook

Diagnosis

Paragon is a two-person Y Combinator (W25) company with a clear, under-served market: distributors where 74% experiment with AI but only 19% have put it into real operations. It says millions of dollars of industrial spend already pass through its agents, and it is hiring its first salesperson and a forward-deployed engineer. The website is not yet working for it.

The sitemap lists only paragon.ai, a domain that now redirects to a different company, so Google is told nothing about runparagon.com. The demo page names "Paragon AI", the other company's name. There are no customer names, the two case studies don't say which platform they used, and there is no pricing, trust or ERP content to answer a buyer's first questions, while Proton shows 26 logos, SOC 2 and comparison pages.

This playbook fixes the site and the name first, then gives the founding salesperson a proof-led outbound motion (a free PO teardown, ERP and vertical pages, an order-entry cost calculator, buying groups), and finally tests a self-serve Outlook add-in and per-order pricing.

Growth Score

5/10

Biggest Opportunity

Order entry is a measurable cost (about 20 minutes and $10–12 per order). With a working site, ERP-specific pages and a free PO teardown, Paragon can turn "millions of dollars through our agents" into proof that sells itself to distributors.

Biggest Risk

A crowded order-entry field (Proton, WizCommerce, Emanate and others) and a generic name shared with several AI companies. Paragon could be outspent or simply not found, while a two-person team sells remotely into a relationship-driven market.

01

Start here

The three moves to make first. Each one is written up in full below.

  1. 1

    Point the sitemap at runparagon.com: today it lists only paragon.ai, which redirects to a different company

    Replace https://paragon.ai with https://www.runparagon.com in the sitemap generator and deploy.

    High impact First result in 1-2 weeks
  2. 2

    Rebuild the two case studies around Surge, Forge and Alloy, and show them on the homepage and demo page

    Ask the most active customer for a quote and permission to cite their ERP and order volume anonymously.

    High impact First result in 2-4 weeks
  3. 3

    Fix the demo page: it says "Paragon AI", another company's name, and asks for nothing that helps a rep prepare

    Change the heading to "Interested in solving your problems with Paragon?" today.

    High impact First result in 2 weeks

02

Your idea portfolio

23 ideas, grouped by when to run them. Open any idea for how to run it: the steps, who to approach, copy you can use, and how to measure it.

  • 8 ideas are marked Unexpected: plays you are unlikely to have tried.

Every idea placed by the effort it takes against the impact it could have. Quick wins sit top left; select a number to open that idea. The filters below narrow the list by status, timing, category and what Programmable can build.

Effort vs impactEach number is an idea. Select one to open it.
Impact →
Quick wins · 3Big bets · 9Fill-ins · 6Reconsider · 5
Effort →

Showing 23 of 23 ideas

This week 7

Quick wins with a clear first action.

  • 01

    All 11 URLs in runparagon.com/sitemap.xml (home, Forge, Surge, Alloy, company, both case studies, book-demo, careers, privacy, terms) are on paragon.ai, not runparagon.com. paragon.ai now 301-redirects to sales.paragon.ai, a different company's call-recording and CRM product. Search engines ignore sitemap entries on another host, so the sitemap tells them nothing about runparagon.com, and the only domain it does advertise sends visitors to someone else. There is also no robots.txt (404) to point crawlers at a sitemap, and every lastmod is 2024-12-19.

    High impactLow effort First result in 1-2 weeks Founder/CTO

    First action

    Replace https://paragon.ai with https://www.runparagon.com in the sitemap generator and deploy.

    How to measure it

    Indexed pages (Search Console)

    Now
    –
    Target
    11 of 11 indexed; branded clicks up 50%
    By
    30 days

    Steps

    1. 1

      Regenerate the sitemap

      Build it from live routes with the runparagon.com base URL and real lastmod dates.

      Founder/CTO 1 hour

    2. 2

      Add robots.txt

      Allow crawling and reference /sitemap.xml.

      Founder/CTO 15 minutes

    3. 3

      Sweep old-domain references

      Canonical, og:url, structured data and image URLs that still say paragon.ai.

      Founder/CTO 2 hours

    4. 4

      Verify the domain

      Google Search Console and Bing Webmaster Tools, domain-level.

      Founder (CEO) 30 minutes

    5. 5

      Submit and request indexing

      Sitemap, then homepage, platform pages and case studies.

      Founder (CEO) 30 minutes

    6. 6

      Check link previews

      Paste the homepage into LinkedIn and Slack and confirm the right title, image and domain.

      Founding Sales 15 minutes

    7. 7

      Monitor

      Coverage report and branded queries each week for a month.

      Founder (CEO) weekly

    First result
    1-2 weeks
    Cost
    $0, a few hours of engineering
    Owner
    Founder/CTO

    Smallest test

    Fix the sitemap and robots file today, submit them, and check the Search Console coverage report in seven days.

    Why it works

    Regenerate the sitemap from the live routes with https://www.runparagon.com URLs and real lastmod dates. Add a robots.txt that references it, verify runparagon.com in Google Search Console and Bing Webmaster Tools, submit the sitemap, and request indexing for the homepage, three platform pages and two case studies. Then search the codebase for any other hard-coded paragon.ai references (canonical tags, Open Graph URLs, structured data, images) and replace them with runparagon.com. In Next.js this is one sitemap and one robots route that build from the site's base URL, so new pages are added automatically.

    For a two-person company with no ad budget, organic and brand search is how a distributor who hears "Paragon" at a trade show or in a cold email checks it out. A correct sitemap and robots file are the minimum for Google to crawl the platform pages quickly, and removing the old domain stops search engines and link previews from pointing at a different business.

    Ready to use

    ChecklistDomain clean-up list

    [ ] sitemap.xml: 11 URLs on https://www.runparagon.com, real lastmod dates
    [ ] robots.txt: Allow: / and Sitemap: https://www.runparagon.com/sitemap.xml
    [ ] Canonical and og:url on every page = runparagon.com
    [ ] No paragon.ai in code, images or structured data
    [ ] Search Console + Bing verified, sitemap submitted
    [ ] LinkedIn/X link preview shows the right title and image

    Risks and how to handle them

    • Pages were already indexed under the wrong host

      Use Search Console's URL inspection to confirm runparagon.com is the chosen canonical.

    • New pages get added by hand and drift again

      Generate the sitemap from the route list at build time.

    Evidence

    • Your websiteSitemap (checked November 10, 2026): all 11 URLs point to https://paragon.ai/ (home, the three platform pages, company, two case studies, book-demo, careers, privacy, terms), none to runparagon.com, all with lastmod 2024-12-19; /robots.txt returns 404.Source
    • Domain checkparagon.ai returns a 301 redirect to sales.paragon.ai, the site of a different company, "Paragon AI, formerly Pilot AI", which sells call recording and CRM updates ("Perfect CRM hygiene on autopilot").Source
    • HiringFounding Sales (Y Combinator job board, San Francisco, $50K–$200K, 0.10%–2.00% equity): Paragon's "first full-time sales hire"; builds pipeline by researching distributors, cold calling and running outbound by email, phone, conferences and referrals; attends trade shows; uses AI tools for lists and outreach; turns what works into playbooks. The company says "millions of dollars in industrial spend already pass through its agents".Source
  • 02

    Paragon's only proof is two unnamed case studies, AI pricing for a packaging distributor and a warehouse layout redesign for a parts supplier. Neither says which Paragon platform was used, gives a timeframe or includes a quote, and the homepage and demo page link to neither. Meanwhile the platform pages make strong claims (99.5% accuracy, 5x faster orders, 15–24% better retention) that no customer story backs up.

    High impactLow effort First result in 2-4 weeks Founder

    First action

    Ask the most active customer for a quote and permission to cite their ERP and order volume anonymously.

    How to measure it

    Homepage-to-demo conversion

    Now
    –
    Target
    +30%
    By
    60 days

    Steps

    1. 1

      Pick the stories

      Two existing and one new Surge story with real volume.

      Founder (CEO) 1 day

    2. 2

      Collect the facts

      ERP, baseline, after numbers, timeframe and a quote.

      Founding Engineer 3 days

    3. 3

      Rewrite

      Same template for all three, tagged by platform.

      Founder (CEO) 2 days

    4. 4

      Add a proof strip

      Three cards on the homepage and demo page.

      Software Engineer, Product 1 day

    5. 5

      Cross-link

      Each platform page links to its story.

      Software Engineer, Product 2 hours

    6. 6

      Use in outreach

      Attach the matching story to each follow-up.

      Founding Sales ongoing

    First result
    2-4 weeks
    Cost
    Staff time only
    Owner
    Founder (CEO)

    Smallest test

    Add the platform name, ERP and timeframe to the two existing stories this week.

    Why it works

    Rewrite each case study in a fixed shape: customer type and size (anonymous is fine), ERP, the platform used, the workflow before and after, the numbers with baseline and timeframe, and one quote from a named role. Add a third story from whichever live customer has the most volume through Surge, since the company says millions of dollars of spend already flow through its agents. Put a three-card proof strip on the homepage and the demo page, and link each platform page to its matching story.

    Industrial buyers are risk-averse and want to see someone like them first. Tying each number to a platform, an ERP and a timeframe makes the claims believable and gives the founding salesperson something to send after every call.

    Risks and how to handle them

    • Customers won't be named

      Use role, region, size and ERP instead of a name.

    Evidence

    • Your websiteCase studies: two pages, both unnamed customers: "a leading national packaging distributor" (AI pricing: 25% faster quotes, 3% higher gross margin) and "a major parts supplier" (warehouse layout redesign: 20% faster picking, 15% more storage); neither names the Paragon platform used, gives a timeframe or a quote.Source
    • Your websiteHomepage (checked November 10, 2026): H1 "The AI Growth Engine For Industrials", "AI agents that automate order processing, customer engagement, and demand forecasting for manufacturers and distributors"; industries HVAC Systems, Pipe, Valve & Fittings, Lumber Supply, Electrical Components and Industrial Packaging; HVAC copy promises to "engage customers around relevant weather events"; a "30-minute demo" CTA; no customer names, logos, numbers or case-study links; the page title says "AI Industrial Automation Platform for Manufacturing & Distribution | Paragon"; footer "Backed by Y Combinator", "© 2026 Serix Inc.", San Francisco.Source
    • Your websiteSurge page: handles purchase orders, RFQs, invoices, bills of lading and work orders, plus a counter-sales AI that "listens" to live conversations and 24/7 order-status bots; claims 99.5% accuracy, 80% less manual document handling and 5x faster order and RFQ processing, "based on customers processing 1M+ quotes annually"; names SAP, Oracle NetSuite, Microsoft Dynamics 365, Infor, Epicor and Sage, with custom builds for on-premise ERPs.Source
    • HiringFounding Sales (Y Combinator job board, San Francisco, $50K–$200K, 0.10%–2.00% equity): Paragon's "first full-time sales hire"; builds pipeline by researching distributors, cold calling and running outbound by email, phone, conferences and referrals; attends trade shows; uses AI tools for lists and outreach; turns what works into playbooks. The company says "millions of dollars in industrial spend already pass through its agents".Source
  • 03

    The book-demo page's heading reads "Interested in solving your problems with Paragon AI?". Paragon AI is the name of the separate call-recording company that paragon.ai now redirects to. The page shows no form fields, testimonials, logos or numbers, only the YC badge, so a distributor arriving from a cold email gets no reason to book and the founders get no context before the call.

    High impactLow effort First result in 2 weeks Founder

    First action

    Change the heading to "Interested in solving your problems with Paragon?" today.

    How to measure it

    Demo bookings per demo-page visit

    Now
    –
    Target
    +25%
    By
    45 days

    Steps

    1. 1

      Fix the name

      Heading and any "Paragon AI" copy.

      Software Engineer, Product 15 minutes

    2. 2

      Add the qualifier

      Five fields including ERP and monthly order volume.

      Software Engineer, Product 1 day

    3. 3

      Embed booking

      Calendar inside the page, 30-minute slots.

      Software Engineer, Product 2 hours

    4. 4

      Add proof

      Three numbers and the case-study cards.

      Founder (CEO) 2 hours

    5. 5

      Route

      Send answers to the founding salesperson's CRM.

      Founding Sales 2 hours

    6. 6

      Track

      Visits, starts and bookings by source.

      Founder (CEO) weekly

    First result
    2 weeks
    Cost
    $0–$20 a month for scheduling
    Owner
    Founder (CEO)

    Smallest test

    Fix the heading and add the ERP dropdown this week.

    Why it works

    Rename the heading to Paragon and make the page a short qualifier: role, company, ERP (SAP, NetSuite, Dynamics 365, Infor, Epicor, Sage, other), monthly POs or RFQs handled, and the platform of interest. Add an embedded calendar so the visitor books in one step, three proof points from the platform pages and case studies, and "what happens on the call" in three lines. Route answers to the founding salesperson with the ERP and volume in the calendar invite.

    The right name stops confusion with another vendor, and asking for the ERP and order volume lets each 30-minute demo open with that buyer's own integration and savings.

    Risks and how to handle them

    • More fields lower bookings

      Keep five fields and make volume a dropdown.

    Evidence

    • Your websiteBook a demo page: heading "Interested in solving your problems with Paragon AI?"; no visible form fields, testimonials, logos or metrics, only the "Backed by Y Combinator" badge.Source
    • Domain checkparagon.ai returns a 301 redirect to sales.paragon.ai, the site of a different company, "Paragon AI, formerly Pilot AI", which sells call recording and CRM updates ("Perfect CRM hygiene on autopilot").Source
    • Your websiteSurge page: handles purchase orders, RFQs, invoices, bills of lading and work orders, plus a counter-sales AI that "listens" to live conversations and 24/7 order-status bots; claims 99.5% accuracy, 80% less manual document handling and 5x faster order and RFQ processing, "based on customers processing 1M+ quotes annually"; names SAP, Oracle NetSuite, Microsoft Dynamics 365, Infor, Epicor and Sage, with custom builds for on-premise ERPs.Source
  • 04

    The homepage H1 says "The AI Growth Engine For Industrials" and the YC profile pitches revenue growth and margin expansion, but the page title that search results show says "AI Industrial Automation Platform for Manufacturing & Distribution". "Industrial automation" usually means PLCs and robotics, which pulls the wrong searchers and hides the order-entry and upsell story.

    Medium impactLow effort First result in 2-4 weeks Founder

    First action

    Rewrite the homepage title and meta description.

    How to measure it

    Non-brand CTR (Search Console)

    Now
    –
    Target
    +20%
    By
    60 days

    Steps

    1. 1

      List the queries

      Ten phrases buyers use, from sales calls.

      Founding Sales 1 hour

    2. 2

      Rewrite home

      Title and description.

      Founder (CEO) 1 hour

    3. 3

      Rewrite platforms

      Forge, Surge, Alloy titles with numbers.

      Founder (CEO) 1 hour

    4. 4

      Rewrite stories

      Result-led case-study titles.

      Founder (CEO) 30 minutes

    5. 5

      Ship

      Next.js metadata per route.

      Software Engineer, Product 1 hour

    6. 6

      Measure

      CTR by page after four weeks.

      Founder (CEO) monthly

    First result
    2-4 weeks
    Cost
    $0
    Owner
    Founder (CEO)

    Smallest test

    Change the homepage title today.

    Why it works

    Rewrite the homepage title and description around the buyer's words: "AI agents for distributors: PO and RFQ entry, upsell and forecasting | Paragon". Give Forge, Surge and Alloy their own titles and descriptions built from their headline numbers, and give each case study a title that states the result.

    Titles decide who clicks. Matching them to how distributors search ("PO automation", "order entry", "RFQ") brings the right visitors and keeps the message consistent from search result to page.

    Risks and how to handle them

    • Dropping a phrase that does bring leads

      Check Search Console queries before changing.

    Builds on

    Evidence

    • Your websiteHomepage (checked November 10, 2026): H1 "The AI Growth Engine For Industrials", "AI agents that automate order processing, customer engagement, and demand forecasting for manufacturers and distributors"; industries HVAC Systems, Pipe, Valve & Fittings, Lumber Supply, Electrical Components and Industrial Packaging; HVAC copy promises to "engage customers around relevant weather events"; a "30-minute demo" CTA; no customer names, logos, numbers or case-study links; the page title says "AI Industrial Automation Platform for Manufacturing & Distribution | Paragon"; footer "Backed by Y Combinator", "© 2026 Serix Inc.", San Francisco.Source
    • Y CombinatorWinter 2025 batch, team of 2, San Francisco; launch post "Free money for distributors and manufacturers": quotes go unfollowed, small accounts get little outreach and reps spend their time on data entry; Surge described as an Outlook add-in that pulls RFQ and PO details from emails, spreadsheets and PDFs for rep review; claims "5 more hours per day" with customers and a 90%+ cut in quote response time. Social profiles use the handle "deployparagon" (LinkedIn, X, YC).Source
  • 05

    The company appears as runparagon.com, "deployparagon" on LinkedIn, X and YC, "Paragon AI" on its own demo page, and Serix Inc. in the footer. Meanwhile a separate Paragon AI sells sales-call software, and Rapidflare launched an agent platform also called "Forge" for electronics distributors in May 2026. A buyer who heard the name at a trade show has several wrong Paragons, and a wrong Forge, to land on.

    Medium impactLow effort First result in 1 month Founder

    First action

    Update the LinkedIn and X bios with the domain and the descriptor.

    How to measure it

    runparagon.com rank for "Paragon AI distributors"

    Now
    –
    Target
    Top 3
    By
    90 days

    Steps

    1. 1

      Choose the descriptor

      One line, used everywhere.

      Founder (CEO) 1 hour

    2. 2

      Update profiles

      LinkedIn, X, YC, Crunchbase-style listings.

      Founder (CEO) 2 hours

    3. 3

      Prefix products

      "Paragon Forge" etc. in site copy and titles.

      Software Engineer, Product 2 hours

    4. 4

      Add Organization schema

      Name, legal name, URL and sameAs profiles.

      Founder/CTO 1 hour

    5. 5

      Standardise signatures

      Domain and descriptor in every email.

      Founding Sales 30 minutes

    6. 6

      Check searches

      Five branded queries each month.

      Founder (CEO) monthly

    First result
    1 month
    Cost
    $0
    Owner
    Founder (CEO)

    Smallest test

    Search "Paragon Surge" before and 30 days after the changes.

    Why it works

    Pick one public descriptor, e.g. "Paragon, AI agents for distributors", and use it in every bio, page title, email signature and YC profile. Make the X and LinkedIn bios and banners say "runparagon.com · AI agents for distributors", and in copy write "Paragon Forge", "Paragon Surge" and "Paragon Alloy" rather than the bare product names. Check that branded searches ("Paragon distributors AI", "Paragon Surge PO") return runparagon.com first.

    A generic name needs a consistent qualifier to be findable. One descriptor repeated everywhere builds the association search engines and buyers need.

    Risks and how to handle them

    • Renaming handles loses followers

      Change bios first and keep the handles.

    Evidence

    • Y CombinatorWinter 2025 batch, team of 2, San Francisco; launch post "Free money for distributors and manufacturers": quotes go unfollowed, small accounts get little outreach and reps spend their time on data entry; Surge described as an Outlook add-in that pulls RFQ and PO details from emails, spreadsheets and PDFs for rep review; claims "5 more hours per day" with customers and a 90%+ cut in quote response time. Social profiles use the handle "deployparagon" (LinkedIn, X, YC).Source
    • Domain checkparagon.ai returns a 301 redirect to sales.paragon.ai, the site of a different company, "Paragon AI, formerly Pilot AI", which sells call recording and CRM updates ("Perfect CRM hygiene on autopilot").Source
    • Competitor researchRapidflare launched an AI agent architecture also named "Forge" for electronics distributors and OEMs on May 15, 2026.Source
    • Your websiteBook a demo page: heading "Interested in solving your problems with Paragon AI?"; no visible form fields, testimonials, logos or metrics, only the "Backed by Y Combinator" badge.Source
  • 06

    The homepage scores well (93 mobile, 99 desktop), but 554 KB of its 864 KB is images, with 366 KB of savings available, and mobile LCP is 2.1 s. Accessibility scores 87, with buttons that have no accessible name, links with no discernible name, low-contrast text and images shown at the wrong aspect ratio. There is not yet enough real-visitor data for a Core Web Vitals verdict.

    Low impactLow effort First result in 1 week Software Engineer

    First action

    Run the PageSpeed report and fix the top three image files.

    How to measure it

    Accessibility score

    Now
    –
    Target
    95+
    By
    2 weeks

    Steps

    1. 1

      Resize images

      next/image with correct sizes.

      Software Engineer, Product 3 hours

    2. 2

      Fix aspect ratios

      Set width and height on flagged images.

      Software Engineer, Product 1 hour

    3. 3

      Label controls

      aria-label on icon buttons and social links.

      Software Engineer, Product 1 hour

    4. 4

      Raise contrast

      Adjust the Tailwind grey tokens.

      Software Engineer, Product 1 hour

    5. 5

      Retest

      PageSpeed mobile and desktop.

      Software Engineer, Product 30 minutes

    6. 6

      Add to review

      Lighthouse check on each deploy preview.

      Founder/CTO 2 hours

    First result
    1 week
    Cost
    $0
    Owner
    Software Engineer, Product

    Smallest test

    Fix the largest image and re-run PageSpeed.

    Why it works

    Serve images through next/image at their display size in AVIF or WebP, fix the stretched images, add aria-labels to icon buttons and social links, and raise the contrast of the grey text. This is polish, not a rescue: the site is already fast, so these fixes are about looking finished on a procurement team's laptop and passing accessibility checks some enterprise buyers run.

    Enterprise buyers judge vendor maturity quickly, and accessibility gaps show up in procurement checklists. The fixes are cheap.

    Risks and how to handle them

    • Design looks different

      Change only contrast tokens that fail.

    Evidence

    • PageSpeedMobile 93/100, desktop 99; mobile LCP 2.1 s; 864 KB page, 554 KB of it images, with 366 KB of image savings; accessibility 87, with buttons without an accessible name, links without a discernible name and low-contrast text; "displays images with incorrect aspect ratio"; not enough real-visitor data for a Core Web Vitals verdict.Source
  • 07

    runparagon.com/careers lists "Applied AI Research Engineer" and "Software Engineer, Product" with no description, location or pay and email-only applications. The roles actually advertised on the YC job board, Founding Sales and Founding Engineer (forward-deployed, $50K–$200K plus 0.10–2.00% equity), aren't on the site at all. For a two-person team, the next two hires are the growth plan.

    Medium impactLow effort First result in 2 weeks Founder/CTO

    First action

    Copy the two YC role descriptions onto the careers page.

    How to measure it

    Applicants passing the founder screen

    Now
    –
    Target
    10 per role
    By
    45 days

    Steps

    1. 1

      Sync the list

      Same roles as the YC board.

      Founder/CTO 1 hour

    2. 2

      Add full text

      Role, pay, equity and interview steps.

      Founder/CTO 2 hours

    3. 3

      One apply path

      Link to the YC board or one company address.

      Founder/CTO 30 minutes

    4. 4

      Add context

      Platforms, ERPs, industries, travel.

      Founder (CEO) 1 hour

    5. 5

      Add JobPosting schema

      So roles appear in job search.

      Software Engineer, Product 2 hours

    6. 6

      Share

      Post each role from the company LinkedIn page.

      Founder (CEO) 30 minutes

    First result
    2 weeks
    Cost
    $0
    Owner
    Founder/CTO

    Smallest test

    Add the Founding Sales role this week and count applicants.

    Why it works

    List all open roles with the full YC text, salary and equity bands, the interview steps and one apply path (the YC board or one runparagon.com address). Add a short "what you'll work on" section with the platforms, the ERPs and the industries, and a line on customer-site travel for the forward-deployed role.

    Candidates check the company site after seeing a YC post. A careers page that matches the posts, with pay shown, converts more of them and reflects well on Paragon with prospects who look too.

    Risks and how to handle them

    • Two lists drift again

      Make the YC board the single source and link to it.

    Evidence

    • Your websiteCareers page: lists "Applied AI Research Engineer" and "Software Engineer, Product" with no descriptions, location or pay, and email-only applications; the Founding Sales and Founding Engineer roles on the YC board are not listed.Source
    • HiringFounding Sales (Y Combinator job board, San Francisco, $50K–$200K, 0.10%–2.00% equity): Paragon's "first full-time sales hire"; builds pipeline by researching distributors, cold calling and running outbound by email, phone, conferences and referrals; attends trade shows; uses AI tools for lists and outreach; turns what works into playbooks. The company says "millions of dollars in industrial spend already pass through its agents".Source
    • HiringFounding Engineer (Y Combinator job board): a forward-deployed engineer who builds agents for "quote intake, order entry, product matching, pricing, inventory checks" and customer and vendor communication, integrates mainframe ERPs, CRMs, email and spreadsheets, travels to customer sites and turns customer-specific fixes into product.Source

This quarter 11

The core plays for the next three months.

  • 08

    Surge names six ERPs plus custom builds for on-premise systems, yet there is a single Surge page and no page per ERP. A distributor's first question is "does it work with our ERP?", and the field already competes there: Proton names Epicor P21, Eclipse, SX.e and NetSuite. Integration cost is one of the biggest adoption barriers, with agency quotes around $500K.

    High impactMedium effort First result in 6-8 weeks Founder/CTO

    First action

    Write the Epicor page first, using notes from the last integration.

    How to measure it

    ERP-page demo bookings

    Now
    –
    Target
    8 a month
    By
    1 quarter

    Steps

    1. 1

      Rank ERPs

      By prospects in the pipeline.

      Founding Sales 1 day

    2. 2

      Document integrations

      Method and setup time per ERP.

      Founding Engineer 1 week

    3. 3

      Write a template

      Documents, flow, setup, week one, example.

      Founder/CTO 2 days

    4. 4

      Build pages

      Six ERPs plus legacy systems.

      Software Engineer, Product 2 weeks

    5. 5

      Link

      From Surge and the demo form.

      Software Engineer, Product 1 day

    6. 6

      Measure

      Traffic and demos per page.

      Founder (CEO) monthly

    First result
    6-8 weeks
    Cost
    Staff time only
    Owner
    Founder/CTO

    Smallest test

    Publish the Epicor page and track its demo bookings for a month.

    Why it works

    Build six pages, one per ERP, each covering which documents Surge reads, how orders land in that ERP (API, import or screen entry), the typical setup time, what the forward-deployed engineer does in week one, and a sample PO-to-sales-order walkthrough. Add a seventh page for legacy and mainframe systems. Link them from Surge and the demo form's ERP dropdown.

    ERP-specific searches come from buyers who have already decided to automate. A page that answers "how long, how hard, with what" for their system removes the main reason not to take the call.

    Risks and how to handle them

    • Vendor trademark use

      Describe compatibility factually and avoid ERP logos unless partnered.

    Builds on

    Evidence

    • Your websiteSurge page: handles purchase orders, RFQs, invoices, bills of lading and work orders, plus a counter-sales AI that "listens" to live conversations and 24/7 order-status bots; claims 99.5% accuracy, 80% less manual document handling and 5x faster order and RFQ processing, "based on customers processing 1M+ quotes annually"; names SAP, Oracle NetSuite, Microsoft Dynamics 365, Infor, Epicor and Sage, with custom builds for on-premise ERPs.Source
    • Competitor researchProton, "the only AI platform built exclusively for distributors": about 26 customer logos, named testimonials, membership of 15 distribution groups and networks, SOC 2 Type II with a Trust Center, comparison pages (/compare), a glossary, podcast, newsletter and case studies; names Epicor P21, Eclipse, SX.e and NetSuite.Source
    • Industry researchB2B eCommerce newsletter (February 20, 2026): 74% of distributors are experimenting with AI but only 19% have implemented it across meaningful operations (American Supply Association figure); barriers are product data scattered across ERPs, spreadsheets and PDFs, pricing logic held by senior reps, understaffed IT and integration quotes of around $500K; Emanate is named as a startup selling quoting and pricing agents.Source
  • 09

    The homepage lists HVAC Systems, Pipe, Valve & Fittings, Lumber Supply, Electrical Components and Industrial Packaging, but only HVAC gets a sentence and none has its own page. Forge, Surge and Alloy do different jobs in each: weather-driven demand in HVAC, counter sales in electrical and lumber, quote speed in packaging, where one case study already lives.

    Medium impactMedium effort First result in 2 months Founder

    First action

    Write the industrial packaging page around the existing packaging case study.

    How to measure it

    Reply-to-demo rate on vertical sequences

    Now
    –
    Target
    +30%
    By
    1 quarter

    Steps

    1. 1

      Interview

      One customer or prospect per industry.

      Founding Sales 2 weeks

    2. 2

      Pick the workflow

      The costliest manual task per industry.

      Founder (CEO) 2 days

    3. 3

      Write pages

      Workflow, platform, sample document, proof.

      Founder (CEO) 2 weeks

    4. 4

      Build

      /industries/<slug> with schema.

      Software Engineer, Product 1 week

    5. 5

      Link outbound

      Each sequence lands on its page.

      Founding Sales 1 day

    6. 6

      Review

      Conversion by industry.

      Founder (CEO) monthly

    First result
    2 months
    Cost
    Staff time only
    Owner
    Founder (CEO)

    Smallest test

    Publish the packaging page and send one sequence to it.

    Why it works

    Publish five industry pages. Each one opens with that industry's costliest manual task, then shows which platform handles it, a sample document (an HVAC submittal, a lumber takeoff, an electrical counter ticket, a packaging RFQ), the relevant claim from the platform page, and the matching case study where one exists. Use them as landing pages for vertical outbound and trade-show follow-up.

    Distributors think in their own industry. A page that uses their documents and language makes a two-person startup look like a specialist.

    Risks and how to handle them

    • Thin pages with no proof

      Launch industries with a customer first; add others as stories arrive.

    Builds on

    Evidence

    • Your websiteHomepage (checked November 10, 2026): H1 "The AI Growth Engine For Industrials", "AI agents that automate order processing, customer engagement, and demand forecasting for manufacturers and distributors"; industries HVAC Systems, Pipe, Valve & Fittings, Lumber Supply, Electrical Components and Industrial Packaging; HVAC copy promises to "engage customers around relevant weather events"; a "30-minute demo" CTA; no customer names, logos, numbers or case-study links; the page title says "AI Industrial Automation Platform for Manufacturing & Distribution | Paragon"; footer "Backed by Y Combinator", "© 2026 Serix Inc.", San Francisco.Source
    • Your websiteCase studies: two pages, both unnamed customers: "a leading national packaging distributor" (AI pricing: 25% faster quotes, 3% higher gross margin) and "a major parts supplier" (warehouse layout redesign: 20% faster picking, 15% more storage); neither names the Paragon platform used, gives a timeframe or a quote.Source
    • Your websiteSurge page: handles purchase orders, RFQs, invoices, bills of lading and work orders, plus a counter-sales AI that "listens" to live conversations and 24/7 order-status bots; claims 99.5% accuracy, 80% less manual document handling and 5x faster order and RFQ processing, "based on customers processing 1M+ quotes annually"; names SAP, Oracle NetSuite, Microsoft Dynamics 365, Infor, Epicor and Sage, with custom builds for on-premise ERPs.Source
  • 10

    WizCommerce's launch puts a public baseline on manual order entry: about 20 minutes and $10–12 per order, and $250,000–$550,000 a year for a midmarket distributor. Surge claims 80% less manual handling and 5x faster processing, but the site has no pricing page and no way for an operations manager to put a dollar figure on that.

    High impactMedium effort First result in 1 month Software Engineer

    First action

    Build the formula in a spreadsheet and test it on two customers' real volumes.

    How to measure it

    Calculator-to-demo rate

    Now
    –
    Target
    15%
    By
    1 quarter

    Steps

    1. 1

      Agree the math

      Inputs, defaults and the 80% assumption.

      Founder (CEO) 1 day

    2. 2

      Check with customers

      Run two live accounts through it.

      Founding Engineer 2 days

    3. 3

      Build

      Client-side calculator on /platforms/surge.

      Software Engineer, Product 1 week

    4. 4

      PDF export

      A one-page summary to forward.

      Software Engineer, Product 2 days

    5. 5

      Prefill demo

      Pass volumes to the demo form.

      Software Engineer, Product 1 day

    6. 6

      Use on calls

      Same sheet in every first meeting.

      Founding Sales ongoing

    First result
    1 month
    Cost
    1-2 weeks of engineering
    Owner
    Software Engineer, Product

    Smallest test

    Put a static three-row example table on the Surge page.

    Why it works

    Add a calculator: monthly POs and RFQs, minutes per document and loaded hourly cost, which outputs hours and dollars spent today and the hours freed at Surge's stated 80% reduction. Show the method and a "results vary" note, as the platform pages do. Offer the result as a PDF the visitor can forward to their CFO, prefill the demo form with the inputs, and give the founding salesperson the same sheet for calls.

    Operations leaders need a number to take to the owner or CFO. A calculator built on a published industry baseline is more credible than a vendor claim on its own.

    Risks and how to handle them

    • Inflated savings hurt trust

      Default to conservative inputs and show the formula.

    Evidence

    • Competitor researchWizCommerce launched Ella AI order entry (June 1, 2026) with a stated baseline of about 20 minutes and $10–12 per manually entered order, and $250,000–$550,000 a year in order entry for a midmarket distributor.Source
    • Your websiteSurge page: handles purchase orders, RFQs, invoices, bills of lading and work orders, plus a counter-sales AI that "listens" to live conversations and 24/7 order-status bots; claims 99.5% accuracy, 80% less manual document handling and 5x faster order and RFQ processing, "based on customers processing 1M+ quotes annually"; names SAP, Oracle NetSuite, Microsoft Dynamics 365, Infor, Epicor and Sage, with custom builds for on-premise ERPs.Source
    • Your websiteSite structure: no blog, pricing page, newsletter, security or trust page, or help centre in the site links or sitemap; the sitemap's last content update is 2024-12-19.Source
  • 11

    The founding salesperson is hired to cold call and email distributors, and the forward-deployed engineer already builds quote-intake and order-entry agents. Press coverage flags remote selling from San Francisco into a relationship-driven market as the main risk. A cold email asking for 30 minutes is easy to ignore; one offering to process the prospect's own POs is not.

    High impactMedium effort First result in 3-4 weeks Founding Sales

    First action

    Write the teardown offer email and a one-page data-handling note.

    How to measure it

    Teardowns that become paid pilots

    Now
    –
    Target
    30%
    By
    1 quarter

    Steps

    1. 1

      Define the offer

      20 documents, 48 hours, what they receive.

      Founder (CEO) 1 day

    2. 2

      Data handling

      NDA template, redaction guidance, deletion policy.

      Founder/CTO 2 days

    3. 3

      Build the report

      Extraction, SKU match, accuracy, minutes saved.

      Founding Engineer 1 week

    4. 4

      Write sequences

      Per industry, leading with the teardown.

      Founding Sales 3 days

    5. 5

      Run

      Cold email and phone to 200 distributors.

      Founding Sales 6 weeks

    6. 6

      Review results

      Walk through the teardown on a call.

      Founding Sales ongoing

    7. 7

      Productise

      Automate the report generator.

      Software Engineer, Product 3 weeks

    First result
    3-4 weeks
    Cost
    Engineering time per teardown, falling as it is automated
    Owner
    Founding Sales

    Smallest test

    Offer five teardowns to warm leads and track pilots.

    Why it works

    Offer prospects a free teardown: they forward 20 recent POs or RFQs (redacted if they like), and within 48 hours Paragon returns the extracted lines matched to their SKUs, an accuracy score, the minutes saved at their volume, and the exceptions a human would still check. The teardown is the first meeting. Use a standard NDA and delete the files afterwards.

    Proof on the buyer's own messy documents beats any demo, it uses the product the team already has, and it filters for prospects with real volume.

    Risks and how to handle them

    • Prospects won't share documents

      Accept redacted copies or sample formats from their own customers.

    • Teardowns eat engineering time

      Cap at five a week until the report is automated.

    Evidence

    • HiringFounding Sales (Y Combinator job board, San Francisco, $50K–$200K, 0.10%–2.00% equity): Paragon's "first full-time sales hire"; builds pipeline by researching distributors, cold calling and running outbound by email, phone, conferences and referrals; attends trade shows; uses AI tools for lists and outreach; turns what works into playbooks. The company says "millions of dollars in industrial spend already pass through its agents".Source
    • HiringFounding Engineer (Y Combinator job board): a forward-deployed engineer who builds agents for "quote intake, order entry, product matching, pricing, inventory checks" and customer and vendor communication, integrates mainframe ERPs, CRMs, email and spreadsheets, travels to customer sites and turns customer-specific fixes into product.Source
    • PressHUGE Magazine (January 6, 2026): names Proton.ai as the closest direct competitor, Benimbl in distributor CRM, and Epicor, Infor and SAP Business One in the ERP layer; no customers or traction disclosed; flags the risk of selling remotely from San Francisco into a relationship-driven market and proposes milestones: ERP integration in days, automated orders on real production orders, then expansion within accounts.Source
  • 12

    Proton lists membership of 15 distribution groups and networks alongside its logos, which is how it reaches many independent distributors at once. Paragon sells into five industries with a two-person team, and press coverage flags relationship-driven buying as the risk.

    High impactMedium effort First result in 1 quarter Founder

    First action

    Ask current customers which buying groups they belong to.

    How to measure it

    Member pilots started

    Now
    –
    Target
    5
    By
    2 quarters

    Steps

    1. 1

      Ask customers

      Which groups they belong to.

      Founding Sales 1 week

    2. 2

      Pick one

      The group in the strongest industry.

      Founder (CEO) 1 day

    3. 3

      Apply

      Vendor or associate membership.

      Founder (CEO) 2 weeks

    4. 4

      Package the offer

      A fixed-fee 60-day member pilot.

      Founder (CEO) 2 days

    5. 5

      Present

      Member webinar with a live PO demo.

      Founding Sales 1 day

    6. 6

      Report back

      Share pilot results with the group.

      Founder (CEO) quarterly

    First result
    1 quarter
    Cost
    Membership fees, typically a few thousand dollars a year
    Owner
    Founder (CEO)

    Smallest test

    Get one customer to introduce Paragon to their group's vendor manager.

    Why it works

    Pick the buying group or association that matters most in one target industry (start with HVAC or electrical). Apply for vendor or associate membership, offer members a fixed-fee 60-day Surge pilot, present at a member meeting or webinar, and publish the results to the group. Add the membership to the homepage proof strip.

    Buying groups pre-vet vendors and gather hundreds of independents, so one relationship replaces hundreds of cold calls and supplies the trust a new name lacks.

    Risks and how to handle them

    • Slow group approvals

      Apply to two groups and use a customer's introduction.

    Evidence

    • Competitor researchProton, "the only AI platform built exclusively for distributors": about 26 customer logos, named testimonials, membership of 15 distribution groups and networks, SOC 2 Type II with a Trust Center, comparison pages (/compare), a glossary, podcast, newsletter and case studies; names Epicor P21, Eclipse, SX.e and NetSuite.Source
    • PressHUGE Magazine (January 6, 2026): names Proton.ai as the closest direct competitor, Benimbl in distributor CRM, and Epicor, Infor and SAP Business One in the ERP layer; no customers or traction disclosed; flags the risk of selling remotely from San Francisco into a relationship-driven market and proposes milestones: ERP integration in days, automated orders on real production orders, then expansion within accounts.Source
    • HiringFounding Sales (Y Combinator job board, San Francisco, $50K–$200K, 0.10%–2.00% equity): Paragon's "first full-time sales hire"; builds pipeline by researching distributors, cold calling and running outbound by email, phone, conferences and referrals; attends trade shows; uses AI tools for lists and outreach; turns what works into playbooks. The company says "millions of dollars in industrial spend already pass through its agents".Source
  • 13

    The Founding Sales role includes attending industry events and trade shows. Paragon has no named customers to put on a banner, but it does have a product that can process a printed or emailed PO in minutes.

    Medium impactMedium effort First result in 1 quarter Founding Sales

    First action

    Choose two shows in the strongest industries and check the cost of a small stand or meeting room.

    How to measure it

    Teardowns from events

    Now
    –
    Target
    15 per show
    By
    1 quarter

    Steps

    1. 1

      Pick shows

      Two, in the strongest industries.

      Founding Sales 2 days

    2. 2

      Build the live demo

      Email-in or photo-in to a mock sales order.

      Founding Engineer 1 week

    3. 3

      Pre-book

      Outreach to attendee lists.

      Founding Sales 2 weeks

    4. 4

      Run the desk

      Process attendee POs live.

      Founding Sales show days

    5. 5

      Follow up

      Results email and teardown offer within 24 hours.

      Founding Sales 1 week

    6. 6

      Compare

      Cost per pilot against cold outbound.

      Founder (CEO) after each show

    First result
    1 quarter
    Cost
    Stand or meeting room plus travel
    Owner
    Founding Sales

    Smallest test

    Walk one show with the demo on a laptop before paying for a stand.

    Why it works

    At two vertical shows this quarter, invite attendees to email or photograph a real PO or RFQ at the stand. Surge processes it live on screen into a mock sales order, and the visitor leaves with the result by email and an offer of a full 20-document teardown. Book follow-ups before the show using attendee lists, and send industry-page links afterwards.

    A live result on the buyer's own document is memorable on a crowded show floor and creates a reason to follow up.

    Risks and how to handle them

    • Messy documents fail live

      Show the human-review step as a feature.

    Builds on

    Evidence

    • HiringFounding Sales (Y Combinator job board, San Francisco, $50K–$200K, 0.10%–2.00% equity): Paragon's "first full-time sales hire"; builds pipeline by researching distributors, cold calling and running outbound by email, phone, conferences and referrals; attends trade shows; uses AI tools for lists and outreach; turns what works into playbooks. The company says "millions of dollars in industrial spend already pass through its agents".Source
    • Y CombinatorWinter 2025 batch, team of 2, San Francisco; launch post "Free money for distributors and manufacturers": quotes go unfollowed, small accounts get little outreach and reps spend their time on data entry; Surge described as an Outlook add-in that pulls RFQ and PO details from emails, spreadsheets and PDFs for rep review; claims "5 more hours per day" with customers and a 90%+ cut in quote response time. Social profiles use the handle "deployparagon" (LinkedIn, X, YC).Source
  • 14

    Surge reads POs, invoices and customer emails and writes into the ERP, and Forge reads email, text, chat and call notes. That is exactly the data an IT lead or owner will ask about. The site has no security, data-handling or trust page, while Proton shows SOC 2 Type II and a Trust Center on its homepage.

    High impactMedium effort First result in 1-2 months Founder/CTO

    First action

    List the questions from the last three prospects' security reviews.

    How to measure it

    Demo-to-pilot cycle

    Now
    –
    Target
    -25%
    By
    2 quarters

    Steps

    1. 1

      Collect questions

      From past security reviews.

      Founding Sales 2 days

    2. 2

      Map data flows

      Per platform.

      Founder/CTO 1 week

    3. 3

      Write the page

      Plain-language answers, no overclaiming.

      Founder/CTO 3 days

    4. 4

      Questionnaire pack

      Standard answers for procurement.

      Founder/CTO 1 week

    5. 5

      Publish and link

      Footer, demo page, ERP pages.

      Software Engineer, Product 1 day

    6. 6

      Plan an audit

      Decide on SOC 2 timing from deal demand.

      Founder (CEO) 1 quarter

    First result
    1-2 months
    Cost
    Staff time; an audit later if deals require it
    Owner
    Founder/CTO

    Smallest test

    Send the draft page to the next prospect in security review.

    Why it works

    Publish /security in plain language: data flows for Surge, Forge and Alloy, hosting and encryption, ERP access scopes (read versus write, human approval before submission), retention and deletion, subprocessors, and the status of any audit underway. Offer a standard security questionnaire pack on request, and link the page from the demo page and footer.

    Security review is where small-vendor deals stall. Answering the questions up front shortens the cycle and shows a two-person team can be trusted with ERP access.

    Risks and how to handle them

    • Claiming controls not yet in place

      State only what is true today and label plans as plans.

    Evidence

    • Your websiteForge page: "Understand Every Customer Conversation"; reads email, text, chat transcripts and call notes, finds buying signals, "identifies local events and trends", runs upsell, cross-sell and win-back campaigns; claims 10–17% higher average order value, 6–12% more wallet share and 15–24% better retention ("results vary"); no integrations named.Source
    • Competitor researchProton, "the only AI platform built exclusively for distributors": about 26 customer logos, named testimonials, membership of 15 distribution groups and networks, SOC 2 Type II with a Trust Center, comparison pages (/compare), a glossary, podcast, newsletter and case studies; names Epicor P21, Eclipse, SX.e and NetSuite.Source
    • Your websiteSite structure: no blog, pricing page, newsletter, security or trust page, or help centre in the site links or sitemap; the sitemap's last content update is 2024-12-19.Source
  • 15

    Buyers comparing AI order entry will find Proton, which already publishes /compare pages, and WizCommerce's Ella, with a public cost baseline. Paragon has no comparison content, so its differences go unstated: one vendor across order entry, engagement and forecasting, forward-deployed engineers, and custom integrations for legacy ERPs.

    Medium impactMedium effort First result in 2 months Founder

    First action

    Ask the founding salesperson which alternatives prospects mention most.

    How to measure it

    Comparison-page demos

    Now
    –
    Target
    5 a month
    By
    1 quarter

    Steps

    1. 1

      Pick alternatives

      From call notes.

      Founding Sales 1 day

    2. 2

      Check facts

      Rivals' current sites, cited and dated.

      Founder (CEO) 2 days

    3. 3

      Write

      Fair tables plus where each fits best.

      Founder (CEO) 1 week

    4. 4

      Review

      Claims and trademarks.

      Founder (CEO) 2 days

    5. 5

      Publish

      /compare/<rival> with the teardown CTA.

      Software Engineer, Product 1 day

    6. 6

      Refresh

      Quarterly fact check.

      Founder (CEO) quarterly

    First result
    2 months
    Cost
    Staff time only
    Owner
    Founder (CEO)

    Smallest test

    Publish the build-vs-buy page first.

    Why it works

    Publish three dated, factual pages: Paragon vs Proton (scope, ERPs, setup model), Paragon vs WizCommerce Ella (order entry plus engagement and forecasting), and Paragon vs a build-it-yourself or agency project (time, the roughly $500K integration quotes the industry cites, maintenance). Each ends with the PO teardown offer.

    Comparison searches come from buyers already shortlisting. Being on the page they read lets Paragon frame the decision, and "build vs buy" answers the IT lead's real alternative.

    Risks and how to handle them

    • Outdated competitor claims

      Date each page and link sources.

    Builds on

    Evidence

    • Competitor researchProton, "the only AI platform built exclusively for distributors": about 26 customer logos, named testimonials, membership of 15 distribution groups and networks, SOC 2 Type II with a Trust Center, comparison pages (/compare), a glossary, podcast, newsletter and case studies; names Epicor P21, Eclipse, SX.e and NetSuite.Source
    • Competitor researchWizCommerce launched Ella AI order entry (June 1, 2026) with a stated baseline of about 20 minutes and $10–12 per manually entered order, and $250,000–$550,000 a year in order entry for a midmarket distributor.Source
    • Industry researchB2B eCommerce newsletter (February 20, 2026): 74% of distributors are experimenting with AI but only 19% have implemented it across meaningful operations (American Supply Association figure); barriers are product data scattered across ERPs, spreadsheets and PDFs, pricing logic held by senior reps, understaffed IT and integration quotes of around $500K; Emanate is named as a startup selling quoting and pricing agents.Source
  • 16

    Press coverage suggests a sequence for Paragon: ERP integration in days, automated orders on real production orders, then expansion within accounts. Forge needs exactly what Surge collects (order history, customer emails, quote follow-ups) to drive its claimed 10–17% higher order value and 15–24% better retention. Once Surge is live, the data for a Forge pitch already exists.

    High impactMedium effort First result in 2 months Founding Engineer

    First action

    Run Forge in shadow mode for one current Surge customer.

    How to measure it

    Multi-platform accounts

    Now
    –
    Target
    40% of customers
    By
    2 quarters

    Steps

    1. 1

      Day-30 report

      Template for Surge results.

      Software Engineer, Product 1 week

    2. 2

      Shadow Forge

      Run on Surge data without sending anything.

      Founding Engineer 1 week per account

    3. 3

      Missed-revenue list

      Unfollowed quotes, dormant accounts, gaps.

      Founding Engineer 2 days

    4. 4

      Offer

      Add-on priced against the list's value.

      Founder (CEO) 1 day

    5. 5

      Alloy at day 120

      Forecasts from the same history.

      Founding Engineer 1 week

    6. 6

      Track

      Net revenue retention by cohort.

      Founder (CEO) monthly

    First result
    2 months
    Cost
    Engineer time per account
    Owner
    Founding Engineer

    Smallest test

    One shadow Forge run and one expansion conversation.

    Why it works

    Make expansion a fixed motion. At day 30, send a Surge results report (orders processed, accuracy, hours saved). At day 45, the forward-deployed engineer runs Forge in shadow mode on the customer's own data and lists the unfollowed quotes, dormant accounts and cross-sell gaps it finds. At day 60, offer Forge as an add-on, priced against the dollars in that list. Repeat at day 120 with Alloy for forecasting.

    Expansion revenue from a happy account is cheaper than a new logo, and a list of the customer's own missed revenue is the strongest upsell there is.

    Risks and how to handle them

    • Customer data used without consent

      Ask for written approval before shadow runs.

    Evidence

    • PressHUGE Magazine (January 6, 2026): names Proton.ai as the closest direct competitor, Benimbl in distributor CRM, and Epicor, Infor and SAP Business One in the ERP layer; no customers or traction disclosed; flags the risk of selling remotely from San Francisco into a relationship-driven market and proposes milestones: ERP integration in days, automated orders on real production orders, then expansion within accounts.Source
    • Your websiteForge page: "Understand Every Customer Conversation"; reads email, text, chat transcripts and call notes, finds buying signals, "identifies local events and trends", runs upsell, cross-sell and win-back campaigns; claims 10–17% higher average order value, 6–12% more wallet share and 15–24% better retention ("results vary"); no integrations named.Source
    • Your websiteAlloy page: demand forecasting, network optimization, project ROI analysis and benchmarking; claims ">60%" higher forecast accuracy, 15% lower network costs and "2x" faster decisions; no integrations named.Source
  • 17

    The forward-deployed engineer travels to customer sites and watches how work is really done. Distribution is a relationship business where branch managers and owners know their peers at other distributors and suppliers. No one is currently set up to turn those visits into introductions or proof.

    Medium impactLow effort First result in 1-2 months Founding Engineer

    First action

    Add the two steps to the next customer-visit checklist.

    How to measure it

    Demos from customer referrals

    Now
    –
    Target
    3 a month
    By
    1 quarter

    Steps

    1. 1

      Write the routine

      Video prompt and referral question.

      Founder (CEO) 1 hour

    2. 2

      Consent form

      Simple release for video use.

      Founder (CEO) 1 day

    3. 3

      Referral offer

      Fee credit per signed referral.

      Founder (CEO) 1 day

    4. 4

      Visit

      Run the routine at each site.

      Founding Engineer per visit

    5. 5

      Hand off

      Log introductions for 48-hour follow-up.

      Founding Sales per lead

    6. 6

      Publish clips

      Best videos on the site and LinkedIn.

      Founder (CEO) monthly

    First result
    1-2 months
    Cost
    Fee credits per signed referral
    Owner
    Founding Engineer

    Smallest test

    Try it on the next two visits.

    Why it works

    Give the engineer a short end-of-visit routine: a 2-minute phone video of the CSR or branch manager describing the before and after (with permission), and one question, "who else do you know who re-keys POs?". Offer a referral thank-you (a month's fee credit) and log every introduction so the founding salesperson follows up within 48 hours.

    The person who just saw the agent remove their data entry is the most credible voice for the next buyer, and the engineer is already in the room.

    Risks and how to handle them

    • Customers feel sold to

      Ask only after a visit that delivered a fix.

    Evidence

    • HiringFounding Engineer (Y Combinator job board): a forward-deployed engineer who builds agents for "quote intake, order entry, product matching, pricing, inventory checks" and customer and vendor communication, integrates mainframe ERPs, CRMs, email and spreadsheets, travels to customer sites and turns customer-specific fixes into product.Source
    • PressHUGE Magazine (January 6, 2026): names Proton.ai as the closest direct competitor, Benimbl in distributor CRM, and Epicor, Infor and SAP Business One in the ERP layer; no customers or traction disclosed; flags the risk of selling remotely from San Francisco into a relationship-driven market and proposes milestones: ERP integration in days, automated orders on real production orders, then expansion within accounts.Source
  • 18

    Paragon's strongest traction claim, that millions of dollars in industrial spend already pass through its agents, appears only in its YC job posts. The homepage shows no customer number at all, and press coverage notes that no traction is disclosed. Surge's page adds that its results are based on customers processing "1M+ quotes annually".

    Medium impactLow effort First result in 2 weeks Founder

    First action

    Pull the cumulative order value and document count from production.

    How to measure it

    Homepage demo rate

    Now
    –
    Target
    +15%
    By
    60 days

    Steps

    1. 1

      Define the numbers

      Order value, documents, hours.

      Founder/CTO 1 day

    2. 2

      Automate

      Monthly job writes the totals.

      Software Engineer, Product 2 days

    3. 3

      Design

      One line under the H1.

      Software Engineer, Product 1 day

    4. 4

      Reuse

      Demo page, signatures, outreach.

      Founding Sales 1 day

    5. 5

      Post monthly

      Short LinkedIn update.

      Founder (CEO) monthly

    6. 6

      Measure

      Demo rate before and after.

      Founder (CEO) monthly

    First result
    2 weeks
    Cost
    $0
    Owner
    Founder (CEO)

    Smallest test

    Add a static line with this month's figure.

    Why it works

    Add a proof line under the homepage H1: dollars of orders processed, documents read and hours returned to reps, updated monthly from production data (rounded and cumulative). Repeat it on the demo page and in email signatures, and use the monthly update as a short LinkedIn post.

    A concrete, growing volume number tells a cautious buyer that real distributors already trust the system with real orders, without naming any of them.

    Risks and how to handle them

    • Small numbers look weak

      Lead with the most impressive unit (dollars or documents).

    Evidence

    • HiringFounding Sales (Y Combinator job board, San Francisco, $50K–$200K, 0.10%–2.00% equity): Paragon's "first full-time sales hire"; builds pipeline by researching distributors, cold calling and running outbound by email, phone, conferences and referrals; attends trade shows; uses AI tools for lists and outreach; turns what works into playbooks. The company says "millions of dollars in industrial spend already pass through its agents".Source
    • Your websiteHomepage (checked November 10, 2026): H1 "The AI Growth Engine For Industrials", "AI agents that automate order processing, customer engagement, and demand forecasting for manufacturers and distributors"; industries HVAC Systems, Pipe, Valve & Fittings, Lumber Supply, Electrical Components and Industrial Packaging; HVAC copy promises to "engage customers around relevant weather events"; a "30-minute demo" CTA; no customer names, logos, numbers or case-study links; the page title says "AI Industrial Automation Platform for Manufacturing & Distribution | Paragon"; footer "Backed by Y Combinator", "© 2026 Serix Inc.", San Francisco.Source
    • PressHUGE Magazine (January 6, 2026): names Proton.ai as the closest direct competitor, Benimbl in distributor CRM, and Epicor, Infor and SAP Business One in the ERP layer; no customers or traction disclosed; flags the risk of selling remotely from San Francisco into a relationship-driven market and proposes milestones: ERP integration in days, automated orders on real production orders, then expansion within accounts.Source

Big bets 3

Larger plays that can change the trajectory.

  • 19

    Paragon's YC launch describes Surge as an Outlook add-in that pulls RFQ and PO details from emails, spreadsheets and PDFs for rep review. That's a product a single inside-sales rep could install without an ERP project. Today every route in is a 30-minute demo, and order entry is a crowded field where WizCommerce, Proton and others compete on speed.

    High impactHigh effort First result in 1 quarter Founder/CTO

    First action

    Count how many current users use only the Outlook extraction step.

    How to measure it

    Free-tier weekly actives

    Now
    –
    Target
    300
    By
    2 quarters

    Steps

    1. 1

      Scope the free tier

      Extraction, SKU match and copy, no ERP.

      Founder/CTO 1 week

    2. 2

      Self-serve onboarding

      Sign-in, price-list upload, first document.

      Software Engineer, Product 4 weeks

    3. 3

      Store listing

      Microsoft AppSource submission.

      Founder/CTO 3 weeks

    4. 4

      Usage limits and upgrade

      Soft cap with an upgrade prompt.

      Software Engineer, Product 2 weeks

    5. 5

      Lead alerts

      Notify sales of active domains.

      Software Engineer, Product 1 week

    6. 6

      Convert

      Outreach to accounts with 3+ active reps.

      Founding Sales ongoing

    7. 7

      Review

      Free-to-paid rate and support load.

      Founder (CEO) monthly

    First result
    1 quarter
    Cost
    1-2 engineer-months plus model and hosting costs
    Owner
    Founder/CTO

    Smallest test

    Give 10 reps at prospects a private add-in link and watch usage.

    Why it works

    Publish a free tier in the Microsoft add-in store: up to 50 documents a month extracted into a structured, copyable order with SKU matching against an uploaded price list, with no ERP connection. Upgrades add ERP write-back, more volume and Forge. Each install is a signal at a named distributor for the founding salesperson to follow up.

    Reps who use it daily become internal champions, and the add-in store is a discovery channel no competitor's demo-only model uses.

    Risks and how to handle them

    • Free usage costs money

      Cap documents per month and per domain.

    • Free tier cannibalises paid

      Keep ERP write-back paid-only.

    Builds on

    Evidence

    • Y CombinatorWinter 2025 batch, team of 2, San Francisco; launch post "Free money for distributors and manufacturers": quotes go unfollowed, small accounts get little outreach and reps spend their time on data entry; Surge described as an Outlook add-in that pulls RFQ and PO details from emails, spreadsheets and PDFs for rep review; claims "5 more hours per day" with customers and a 90%+ cut in quote response time. Social profiles use the handle "deployparagon" (LinkedIn, X, YC).Source
    • Competitor researchWizCommerce launched Ella AI order entry (June 1, 2026) with a stated baseline of about 20 minutes and $10–12 per manually entered order, and $250,000–$550,000 a year in order entry for a midmarket distributor.Source
    • Your websiteBook a demo page: heading "Interested in solving your problems with Paragon AI?"; no visible form fields, testimonials, logos or metrics, only the "Backed by Y Combinator" badge.Source
  • 20

    Industry figures show a wide gap: 74% of distributors experiment with AI but only 19% have put it into meaningful operations. The blockers are data scattered across ERPs, spreadsheets and PDFs, pricing logic held by senior reps, and understaffed IT. Paragon has no blog or newsletter, and Proton runs a newsletter, podcast and glossary.

    High impactHigh effort First result in 2 quarters Founder

    First action

    Draft the 15 survey questions and test them on three customers.

    How to measure it

    Readiness self-scores completed

    Now
    –
    Target
    500
    By
    2 quarters

    Steps

    1. 1

      Draft survey

      15 questions by workflow and ERP.

      Founder (CEO) 1 week

    2. 2

      Recruit respondents

      Customers, buying groups, outbound.

      Founding Sales 6 weeks

    3. 3

      Analyse

      Benchmarks by industry and size.

      Applied AI Research Engineer 2 weeks

    4. 4

      Write the report

      Findings and a first-step guide.

      Founder (CEO) 2 weeks

    5. 5

      Build the self-score

      10 questions with a peer comparison.

      Software Engineer, Product 2 weeks

    6. 6

      Launch

      Trade press, LinkedIn, groups.

      Founder (CEO) 2 weeks

    7. 7

      Follow up

      Low scorers get the teardown offer.

      Founding Sales ongoing

    First result
    2 quarters
    Cost
    Staff time plus survey incentives
    Owner
    Founder (CEO)

    Smallest test

    Launch the self-score alone, with industry figures, before the survey.

    Why it works

    Survey distributors in Paragon's five industries on where manual work sits (order entry, quotes, follow-up, forecasting), what share of POs arrive by email or PDF, and their ERP. Publish an annual report with anonymised benchmarks by industry and size, and a 10-question self-score on the site that places each respondent against peers and recommends a first workflow to automate. Promote it through buying groups, trade press and outbound.

    Original data earns press and links that a two-person team can't buy, and the self-score is a lead form buyers actually want to fill in.

    Risks and how to handle them

    • Too few responses

      Partner with one association to field it.

    Builds on

    Evidence

    • Industry researchB2B eCommerce newsletter (February 20, 2026): 74% of distributors are experimenting with AI but only 19% have implemented it across meaningful operations (American Supply Association figure); barriers are product data scattered across ERPs, spreadsheets and PDFs, pricing logic held by senior reps, understaffed IT and integration quotes of around $500K; Emanate is named as a startup selling quoting and pricing agents.Source
    • Competitor researchProton, "the only AI platform built exclusively for distributors": about 26 customer logos, named testimonials, membership of 15 distribution groups and networks, SOC 2 Type II with a Trust Center, comparison pages (/compare), a glossary, podcast, newsletter and case studies; names Epicor P21, Eclipse, SX.e and NetSuite.Source
    • Your websiteSite structure: no blog, pricing page, newsletter, security or trust page, or help centre in the site links or sitemap; the sitemap's last content update is 2024-12-19.Source
  • 21

    Manual order entry costs a distributor about $10–12 and 20 minutes per order, and a midmarket distributor spends $250,000–$550,000 a year on it. Paragon publishes no pricing, and buyers who struggle to get AI past the pilot stage (only 19% have) want spend tied to results. Surge already claims 99.5% accuracy, so it can stand behind a per-order price.

    High impactMedium effort First result in 1 quarter Founder

    First action

    Model margins per document at current processing costs for three customers.

    How to measure it

    Pilots converting to paid

    Now
    –
    Target
    60%
    By
    2 quarters

    Steps

    1. 1

      Model unit economics

      Cost per document by type.

      Founder/CTO 1 week

    2. 2

      Set the price

      Platform fee plus per-correct-document rate.

      Founder (CEO) 2 days

    3. 3

      Define "correct"

      Entered without human edit.

      Founder/CTO 2 days

    4. 4

      Meter usage

      Billing on processed documents.

      Software Engineer, Product 3 weeks

    5. 5

      Publish pricing

      /pricing with the Surge plan.

      Software Engineer, Product 1 week

    6. 6

      Test

      Offer to the next ten prospects.

      Founding Sales 1 quarter

    First result
    1 quarter
    Cost
    Engineering for metering; pricing risk
    Owner
    Founder (CEO)

    Smallest test

    Offer per-order pricing to three prospects as an option alongside a flat fee.

    Why it works

    Test a usage plan: a small platform fee plus a per-document price well under the manual cost (for example $2–3 per processed PO, the exact figure set from your own margins), charged only for documents that enter the ERP without correction. Publish a simple pricing page with this plan and "contact us" for Forge and Alloy, and let the cost calculator show the net saving.

    Charging per correct order removes the risk of paying for an AI that doesn't work, makes the ROI obvious, and grows revenue with each customer's volume.

    Risks and how to handle them

    • Low-volume customers are unprofitable

      Set a minimum monthly fee.

    • Disputes over what counts as correct

      Use the system's own edit log as the record.

    Builds on

    Evidence

    • Competitor researchWizCommerce launched Ella AI order entry (June 1, 2026) with a stated baseline of about 20 minutes and $10–12 per manually entered order, and $250,000–$550,000 a year in order entry for a midmarket distributor.Source
    • Your websiteSurge page: handles purchase orders, RFQs, invoices, bills of lading and work orders, plus a counter-sales AI that "listens" to live conversations and 24/7 order-status bots; claims 99.5% accuracy, 80% less manual document handling and 5x faster order and RFQ processing, "based on customers processing 1M+ quotes annually"; names SAP, Oracle NetSuite, Microsoft Dynamics 365, Infor, Epicor and Sage, with custom builds for on-premise ERPs.Source
    • Industry researchB2B eCommerce newsletter (February 20, 2026): 74% of distributors are experimenting with AI but only 19% have implemented it across meaningful operations (American Supply Association figure); barriers are product data scattered across ERPs, spreadsheets and PDFs, pricing logic held by senior reps, understaffed IT and integration quotes of around $500K; Emanate is named as a startup selling quoting and pricing agents.Source
    • Your websiteSite structure: no blog, pricing page, newsletter, security or trust page, or help centre in the site links or sitemap; the sitemap's last content update is 2024-12-19.Source

Wildcards 2

Unexpected experiments, each with a cheap way to test it.

  • 22

    Surge includes a counter-sales AI that listens to in-person conversations and handles lookups and data entry so reps can focus on service and upselling. Lumber and electrical, two of Paragon's five industries, run busy branch counters, and most AI order-entry rivals sell inbox-based PO processing. This feature is unusual and hard to picture from text alone.

    Medium impactMedium effort First result in 1 month Founder

    First action

    Ask the customer with the busiest counter for a filming slot.

    How to measure it

    Demos from video viewers

    Now
    –
    Target
    10
    By
    1 quarter

    Steps

    1. 1

      Pick a branch

      Busy counter, willing manager.

      Founding Sales 1 week

    2. 2

      Get consent

      Release from staff and customer on camera.

      Founder (CEO) 2 days

    3. 3

      Film

      One real transaction, phone-quality is fine.

      Founding Engineer 1 day

    4. 4

      Edit

      90-second and 15-second cuts with captions.

      Founder (CEO) 3 days

    5. 5

      Place

      Surge and industry pages, outbound.

      Software Engineer, Product 1 day

    6. 6

      Measure

      Views and demos from viewers.

      Founder (CEO) monthly

    First result
    1 month
    Cost
    Under $1,000
    Owner
    Founder (CEO)

    Smallest test

    Record a staged version in a customer's training room.

    Why it works

    With a willing customer, film a 90-second clip at a branch counter: a contractor asks for parts, the agent pulls stock and pricing and builds the ticket while the rep keeps talking. Use it as the hero video on the Surge and lumber/electrical industry pages, cut 15-second versions for LinkedIn, and use it as the opener in outbound to branch-heavy distributors.

    A feature competitors don't show, seen working in a familiar setting, sets Paragon apart from inbox-only order-entry tools at a glance.

    Risks and how to handle them

    • Customer privacy on camera

      Blur faces and pricing if needed.

    Builds on

    Evidence

    • Your websiteSurge page: handles purchase orders, RFQs, invoices, bills of lading and work orders, plus a counter-sales AI that "listens" to live conversations and 24/7 order-status bots; claims 99.5% accuracy, 80% less manual document handling and 5x faster order and RFQ processing, "based on customers processing 1M+ quotes annually"; names SAP, Oracle NetSuite, Microsoft Dynamics 365, Infor, Epicor and Sage, with custom builds for on-premise ERPs.Source
    • Your websiteHomepage (checked November 10, 2026): H1 "The AI Growth Engine For Industrials", "AI agents that automate order processing, customer engagement, and demand forecasting for manufacturers and distributors"; industries HVAC Systems, Pipe, Valve & Fittings, Lumber Supply, Electrical Components and Industrial Packaging; HVAC copy promises to "engage customers around relevant weather events"; a "30-minute demo" CTA; no customer names, logos, numbers or case-study links; the page title says "AI Industrial Automation Platform for Manufacturing & Distribution | Paragon"; footer "Backed by Y Combinator", "© 2026 Serix Inc.", San Francisco.Source
    • Competitor researchWizCommerce launched Ella AI order entry (June 1, 2026) with a stated baseline of about 20 minutes and $10–12 per manually entered order, and $250,000–$550,000 a year in order entry for a midmarket distributor.Source
  • 23

    The homepage says Paragon helps HVAC distributors "engage customers around relevant weather events", and Forge "identifies local events and trends" to trigger campaigns. That's a concrete, timely use case, but nothing on the site shows it working, and it's the kind of signal an HVAC branch manager would want before a demand spike.

    Medium impactMedium effort First result in 2 months Founding Engineer

    First action

    Draft one sample issue for next week's forecast in two regions.

    How to measure it

    Weather-alert subscribers from HVAC distributors

    Now
    –
    Target
    200
    By
    2 quarters

    Steps

    1. 1

      Pick data

      A public weather-forecast feed by region.

      Founding Engineer 2 days

    2. 2

      Map parts

      Which SKU categories spike, from customer history.

      Applied AI Research Engineer 1 week

    3. 3

      Draft the format

      Forecast, parts, sample outreach.

      Founder (CEO) 2 days

    4. 4

      Sign-up page

      Regions and role, on the HVAC page.

      Software Engineer, Product 3 days

    5. 5

      Send weekly

      Automated with a human check.

      Founding Engineer weekly

    6. 6

      Convert

      Offer Forge pilots to engaged readers.

      Founding Sales ongoing

    First result
    2 months
    Cost
    Engineering time; free or low-cost weather data
    Owner
    Founding Engineer

    Smallest test

    Send one manual issue to 30 HVAC prospects.

    Why it works

    Build a free weekly email for HVAC distributors: forecast heat waves and cold snaps by region, the parts that typically spike in demand, and a sample Forge outreach message their reps could send to contractors. Sign-ups give the distributor's regions; the upgrade is Forge sending those campaigns automatically from their own customer list.

    A useful, recurring signal builds an audience of exactly the buyers Paragon wants and shows Forge's logic before any sales call.

    Risks and how to handle them

    • Seasonal lulls

      Switch to monthly outside peak seasons.

    Builds on

    Evidence

    • Your websiteHomepage (checked November 10, 2026): H1 "The AI Growth Engine For Industrials", "AI agents that automate order processing, customer engagement, and demand forecasting for manufacturers and distributors"; industries HVAC Systems, Pipe, Valve & Fittings, Lumber Supply, Electrical Components and Industrial Packaging; HVAC copy promises to "engage customers around relevant weather events"; a "30-minute demo" CTA; no customer names, logos, numbers or case-study links; the page title says "AI Industrial Automation Platform for Manufacturing & Distribution | Paragon"; footer "Backed by Y Combinator", "© 2026 Serix Inc.", San Francisco.Source
    • Your websiteForge page: "Understand Every Customer Conversation"; reads email, text, chat transcripts and call notes, finds buying signals, "identifies local events and trends", runs upsell, cross-sell and win-back campaigns; claims 10–17% higher average order value, 6–12% more wallet share and 15–24% better retention ("results vary"); no integrations named.Source
    • Your websiteSite structure: no blog, pricing page, newsletter, security or trust page, or help centre in the site links or sitemap; the sitemap's last content update is 2024-12-19.Source

03

How Programmable can help

Where an idea in this playbook is something Programmable builds or runs, it's listed here with what we'd do. Everything else is yours to run.

ProgrammableInfo

Turn the research into a list you can act on

These ideas start with finding the right companies. Info finds them and the right person at each, researches every one and drafts the outreach, then asks before it sends.

  • Build this list in Info for idea #11 (opens programmable.info in a new tab)

New accounts: 100,000 free credits with code BETA100K

Find the companies (opens programmable.info in a new tab)
ProgrammableTalk

Fill the events in this plan

These ideas are events. Talk gets registrants for them and tells you who's coming and what their company does before the day.

Plan an event (opens programmable.talk in a new tab)
More example playbooks and reviews

More growth playbooks

Other companies we have researched, each read from what they are hiring for.

  • Green Thumb Industries growth strategyA US cannabis company that makes and sells branded products (RYTHM, Dogwalkers, incredibles, Beboe, Good Green, Doctor Solomon's, &Shine)…21 growth ideas
  • Nextdoor growth strategyThe neighborhood network: local posts, Ask recommendations, alerts, news from local publishers, For Sale & Free, events and groups, plus…20 growth ideas
  • GlossGenius growth strategyAll-in-one software for appointment businesses: booking websites, scheduling, payments and POS, marketing, payroll, EMR for medspas, and…22 growth ideas
  • EōS Fitness growth strategyA High Value. Low Price. gym chain with large 24/7 gyms, group classes, recovery amenities, personal training, Gainiac AI and EGYM smart…22 growth ideas
  • Fender growth strategyFender Musical Instruments Corporation makes and sells guitars, basses, amplifiers and accessories under Fender, Squier, Gretsch, Jackson…21 growth ideas
  • Oscilar growth strategyAn agentic risk platform for financial institutions: real-time decisioning, data and AI agents across fraud, credit underwriting…22 growth ideas
  • Hologram growth strategyGlobal cellular IoT connectivity: eUICC Hyper SIMs with two independent mobile cores, 550+ carriers in 190+ countries, a dashboard and…23 growth ideas
  • SkyWatch growth strategyA geospatial data platform from Kitchener, Canada: one catalog and price list for satellite, aerial, drone and elevation data from 700+…22 growth ideas

See every playbook and website review

Independent research. Programmable wrote this playbook from public sources, as they stood on November 10, 2026, and cites them throughout. Paragon did not commission, review or endorse it, and Programmable is not affiliated with Paragon. Company and product names are trademarks of their owners, used only to identify them. The ideas are suggestions and opinion, not professional, legal or financial advice, and not statements about how Paragon runs its business.

Something wrong, or do you represent Paragon? Email [email protected] and we will correct it or take it down.

30/60/90 roadmap

What to focus on in each month, and the ideas that carry it.

  1. Days 1 to 307 ideas

    Fix the sitemap and robots file, the demo page, titles, name consistency, page speed and accessibility, case studies and careers.

  2. Days 31 to 6011 ideas

    Equip the founding salesperson: ERP and vertical pages, the cost calculator, PO teardowns, buying groups, trade shows, a trust page, comparison pages, expansion and referral loops, and the spend counter.

  3. Days 61 to 905 ideas

    Test the big bets and wildcards: a self-serve Outlook add-in, an AI readiness report, per-order pricing, a counter-sales video and weather-triggered HVAC outreach.

The research behind the plan

3 research areas, each summarised in a few lines. Open any of them for the full detail.

Customers

What customers say

Customers value fewer hours spent re-keying POs and RFQs. The most cited problem: manual order entry costing about 20 minutes and $10–12 per order.

  • Faster quote responses that win more orders
  • Quotes that go unfollowed and small accounts that get no outreach
  • In their words: “order entry”

Market

Competitive landscape

5 competitors compared: Proton, WizCommerce (Ella), Emanate, Benimbl and 1 more.

  • Gap to exploit: Order and quote automation proven on a buyer's own documents before a contract
  • Proton: Broad platform that requires a data-layer rollout
  • WizCommerce (Ella): Order entry only; wholesale and B2B ecommerce focus

Hiring signals

Paragon is a team of two hiring its first salesperson and first forward-deployed engineer, so the next two hires are the growth plan.

  • Founding Sales: First sales hire: outbound, conferences, referrals and trade shows. Ideas 11, 12, 13.
  • Founding Engineer (forward deployed): Order-entry and quote agents on legacy ERPs at customer sites. Ideas 8, 16, 17.
  • Applied AI Research Engineer: Listed with no description. Ideas 7, 20.

From the team that wrote this playbook

All products
  • ProgrammableAgencyWorking AI systems in days, not quarters (opens programmable.agency in a new tab)
  • ProgrammableInfoAn agent that finds, researches and asks first (opens programmable.info in a new tab)
  • ProgrammableDigitalScreenshots in, an animated product demo out (opens programmable.digital in a new tab)
  • ProgrammableWatchKnow the day a supplier falls out of compliance (opens programmable.watch in a new tab)
  • ProgrammableTalkGet registrants for your events, and know who’s coming (opens programmable.talk in a new tab)
  • ProgrammablePageText what you need, your customer gets a page that works (opens programmable.page in a new tab)
  • ProgrammableCommunityForums with labelled AI members, never empty (opens programmable.community in a new tab)
ProgrammableMarketing
ExamplesLeaderboardsProductsTermsPrivacyCookiesDPALegalContact

© 2026 AI-Powered Growth Intelligence