ProgrammableMarketing/sitefire.ai (opens in a new tab)
PDF

Views

  • Plan
  • Roadmap
  • Research

On this page

  • The brief
  • Start here
  • Idea portfolio
  • Effort vs impact
  • This week · 6
  • This quarter · 12
  • Big bets · 3
  • Wildcards · 2
  • How we can help
Generated Nov 10, 2026 1 view
  1. Programmable
  2. Playbooks
  3. Sitefire
Sitefire logo

growth playbook · Nov 10, 2026

sitefire.ai (opens in a new tab)

Sitefire growth playbook

Diagnosis

Sitefire is a two-person YC W26 company from Munich with a sharp product and unusually strong research: a 354,004-answer citation study, a 13,500-post study of AI-written content, and two case studies with control groups. The gaps sit around those assets.

The three free research tools the study, the sitemap and llms.txt point to all return 404; llms.txt lists one case study of two; posts have no authors or calls to action; one result is described two different ways; and the site has no comparison, industry or German pages despite German-speaking enterprise customers and a near-identical rival price at Peec.

This playbook fixes the dead tools and the agent-facing files first, then builds comparison and industry pages, a trial shaped around a first result, and a recurring citation index.

Growth Score

7/10

Biggest Opportunity

Turn Sitefire's own research into the free first step of the funnel: tools, studies and an index that marketers and AI agents find, use and then upgrade from.

Biggest Risk

Peec AI and Profound sell into the same teams with more funding, more comparison content and similar entry prices, so a two-person team has to win on proof and speed rather than on volume.

01

Start here

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

  1. 1

    Bring back the three free research tools that your sitemap, llms.txt and slop study all link to, and make them the front door to the trial

    Decide which of the three tools should be live, and redeploy or redirect the others today.

    High impact First result in 2 weeks
  2. 2

    End every research post with a "run this on your brand" step into the trial

    Add the call to action to the citation study.

    High impact First result in 2 weeks
  3. 3

    Update llms.txt, the file AI agents read first: add Jerry, the plans and working research links

    Add the Jerry case study and the pricing lines to llms.txt today.

    Medium impact First result in 1 month

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.

  • 7 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 · 7Fill-ins · 8Reconsider · 5
Effort →

Showing 23 of 23 ideas

This week 6

Quick wins with a clear first action.

  • 01

    Sitefire's most original asset is its own research: a 13,500-post study showing that structure alone flags AI-written posts with a score of 98.0 out of 100. The post's main call to action is to play Spot the Slop, and llms.txt lists Spot the slop and Slop Checker as Sitefire's research for AI agents. All three tool pages (/slop-checker, /deslop and /spot-the-slop) return 404 Not Found, and they are still in the sitemap. Readers, journalists and the AI crawlers Sitefire helps customers win all reach a dead end at the moment of highest interest.

    High impactLow effort First result in 2 weeks Founder & CTO

    First action

    Decide which of the three tools should be live, and redeploy or redirect the others today.

    How to measure it

    Tool-sourced trials

    Now
    –
    Target
    40 a month
    By
    60 days

    Steps

    1. 1

      Find the cause

      Check why the three routes return 404 and whether the tools still exist in the codebase.

      Founder & CTO 1 hour

    2. 2

      Restore or redirect

      Redeploy the live tools and 301 any retired route to the closest live one.

      Founder & CTO 1 day

    3. 3

      Clean the files

      Make the sitemap and llms.txt list only working tool URLs.

      Founding Product Engineer 1 hour

    4. 4

      Build the hand-off

      Checker result, then "Fix this with Sitefire", then a trial with that page as the first brief.

      Founding Product Engineer 3 days

    5. 5

      Link it everywhere

      From the study, the blog index, the homepage footer and the docs.

      Founder & CEO 1 day

    6. 6

      Watch every sitemap URL

      A daily check that alerts the team when any listed URL stops returning 200.

      Founder & CTO 2 hours

    7. 7

      Measure

      Tool uses, trial clicks and trials started each week.

      Founder & CEO weekly

    First result
    2 weeks
    Cost
    Engineering time only
    Owner
    Founder & CTO

    Smallest test

    Restore Spot the Slop alone and count plays and trial clicks for two weeks.

    Why it works

    This week, restore the three pages, or 301-redirect any retired tool to the live one. Then make the Slop Checker the free first step: a visitor pastes a URL and gets the structural score, the two features that give the post away, and a "Fix this with Sitefire" button. That button starts the trial with the page loaded as the first improvement brief. Spot the Slop stays as the shareable game. Link all three from the study, the blog index, the homepage footer and llms.txt, and take anything that stays retired out of the sitemap.

    The study gives people a reason to visit, and the checker gives them a reason to sign up. Any marketer worried that AI content is penalized has a question the tool answers in seconds, and the answer leads straight to Sitefire's paid product, which writes and improves articles. Dead URLs in the sitemap and llms.txt also tell crawlers the site is not maintained, which is the opposite of what a GEO vendor needs to show.

    Ready to use

    ChecklistResearch tools relaunch list

    [ ] /slop-checker, /deslop and /spot-the-slop return 200 (or 301 to a live tool)
    [ ] Sitemap lists only live tool URLs, with lastmod
    [ ] llms.txt Research section links working pages
    [ ] Slop study CTA tested end to end
    [ ] "Fix this with Sitefire" passes the URL into the trial
    [ ] Result is cached and shareable by link
    [ ] Daily check on every sitemap URL
    [ ] Trials tagged by source tool

    Risks and how to handle them

    • Each check costs model calls

      Cache results per URL and limit anonymous checks per day.

    • A score reads as an accusation

      Present it as structure advice, not a verdict on who wrote the post.

    Evidence

    • Your website"How to Tell AI Slop From Human Writing" (September 2026): 13,500 blog posts analyzed; structure alone separated AI from human posts with a score of 98.0 out of 100; the main call to action is to play the five-round "Spot the Slop" game; no author is named on the post, although the underlying SlopShape study is credited to Sitefire's CEO, with a public verification repository.Source
    • Your websiteResearch tools: /slop-checker, /deslop and /spot-the-slop all returned "404 Not Found" on repeated checks on November 10, 2026 (also on the www host), although all three are listed in the sitemap.Source
    • Your websitellms.txt (the file written for AI agents): describes Sitefire as a GEO platform; lists a "Research" section with "Spot the slop" and "Slop Checker"; lists one case study (Pointhound, not Jerry); no plan prices and no comparison pages.Source
    • Your websiteSitemap (53 URLs, checked November 10, 2026): 13 blog posts, 2 case studies, 27 docs pages (including CMS guides for WordPress, Webflow, Framer and dotCMS), pricing, contact and legal pages, plus /slop-checker, /deslop and /spot-the-slop with no lastmod. No comparison, industry, use-case or German-language pages. Seven blog posts first published between January and April share one lastmod, 2026-07-29.Source
  • 02

    Sitefire's llms.txt is a summary written for AI agents, the audience Sitefire markets to. It lists only the Pointhound case study, not Jerry's controlled +78% result; it gives no plan prices; and its Research section points to two tool pages that return 404.

    Medium impactLow effort First result in 1 month Founder & CEO

    First action

    Add the Jerry case study and the pricing lines to llms.txt today.

    How to measure it

    Accurate mentions of Sitefire's plans and results

    Now
    –
    Target
    80% of tracked answers
    By
    60 days

    Steps

    1. 1

      Audit the file

      List every missing page, wrong link and missing number.

      Founder & CEO 1 hour

    2. 2

      Add proof

      Both case studies with results and method.

      Founder & CEO 1 hour

    3. 3

      Add plans

      Prices, limits and the 7-day trial.

      Founder & CEO 1 hour

    4. 4

      Fix the research links

      Only working tools and studies.

      Founding Product Engineer 1 hour

    5. 5

      Track Sitefire itself

      Add prompts about GEO tools to Sitefire's own workspace.

      Founder & CEO 1 day

    6. 6

      Review on change

      Update the file with every pricing or case-study change.

      Founder & CEO ongoing

    First result
    1 month
    Cost
    $0
    Owner
    Founder & CEO

    Smallest test

    Add Jerry and pricing, then check tracked answers about Sitefire in a month.

    Why it works

    Rewrite llms.txt as Sitefire's sales page for agents: one line on what Sitefire does differently (actions, not just tracking), both case studies with their numbers and methods, the three plans with prices and limits, the trial terms, working research links, and how an agent can book a demo. Review it every time pricing or a case study changes.

    When a buyer asks an AI assistant to compare GEO tools, the assistant often reads files like this. A complete, current file makes Sitefire easy to describe correctly, and it doubles as a live demo of what Sitefire does for customers.

    Risks and how to handle them

    • The file drifts out of date again

      Generate it from the same source as the pricing and case-study pages.

    Builds on

    Evidence

    • Your websitellms.txt (the file written for AI agents): describes Sitefire as a GEO platform; lists a "Research" section with "Spot the slop" and "Slop Checker"; lists one case study (Pointhound, not Jerry); no plan prices and no comparison pages.Source
    • Your websiteJerry case study: improved a selected set of pages and left a comparable set untouched as a control; three months later "+78%" AI referral traffic, "+56%" AI bot traffic, "+27%" AI citation share and "+9%" AI visibility; improved pages averaged 112% of their earlier level versus 72% for untouched pages. Ends with "Book a demo".Source
    • Your websitePricing: Lite "$249/month" (150 tracked prompts, 8 action briefs and 3 AI-optimized articles a month, MCP access; no CMS integration, Brand Kit, Personas or Agent Analytics); Pro "$499/month" (300 prompts, 18 briefs, 8 articles, CMS integration, Agent Analytics, onboarding call); Enterprise custom. "All plans include a 7-day free trial with monthly billing and monthly cancellation." No annual option and no FAQ.Source
    • Your websiteResearch tools: /slop-checker, /deslop and /spot-the-slop all returned "404 Not Found" on repeated checks on November 10, 2026 (also on the www host), although all three are listed in the sitemap.Source
  • 03

    The August changelog promoted a webinar on how a YC-backed scale-up "grew organic clicks by 78%", but the matching case study and the homepage report +78% AI referral traffic. These are different measures. The webinar itself, live content with Sitefire's best result, has no recording page in the sitemap.

    Medium impactLow effort First result in 2 weeks Founder & CEO

    First action

    Correct the webinar line in the changelog to match the case study.

    How to measure it

    Case-study demo bookings

    Now
    –
    Target
    +30%
    By
    60 days

    Steps

    1. 1

      Agree the wording

      One phrasing of the 78% result and its method.

      Founder & CEO 1 hour

    2. 2

      Fix every mention

      Changelog, homepage, case study and deck.

      Founder & CEO 2 hours

    3. 3

      Edit the recording

      Chapters and a short cut for social posts.

      Founder & CEO 1 day

    4. 4

      Publish the page

      Video, transcript and a demo button, linked from the case study.

      Founding Product Engineer 1 day

    5. 5

      Add to the files

      Sitemap and llms.txt.

      Founding Product Engineer 1 hour

    6. 6

      Measure

      Views, demos and trials from the page.

      Founder & CEO monthly

    First result
    2 weeks
    Cost
    $0
    Owner
    Founder & CEO

    Smallest test

    Fix the wording today and post the transcript on the case-study page.

    Why it works

    Pick the precise wording (AI referral traffic, with the control-group method) and use it in the changelog, the homepage, the case study and the sales deck. Publish the webinar recording as a page under the case study, with chapters, a transcript and a demo button, and add it to the sitemap and llms.txt.

    Sitefire's buyers measure traffic for a living; one result described two ways invites doubt. A recorded walkthrough of a controlled result is the strongest proof a small vendor can give, and a transcript gives AI assistants text to quote.

    Risks and how to handle them

    • The customer has not approved a public recording

      Ask first; if not, publish a founder-narrated version of the case study.

    Evidence

    • Your websiteChangelog (August 26, 2026): Events, chart annotations that add model and search updates automatically plus the customer's own publishes, and a new Actions page; a September 2 webinar promised "How a YC-backed scale-up grew organic clicks by 78% in three months using marketing agents", while the matching case study reports +78% AI referral traffic. No webinar recording is in the sitemap.Source
    • Your websiteJerry case study: improved a selected set of pages and left a comparable set untouched as a control; three months later "+78%" AI referral traffic, "+56%" AI bot traffic, "+27%" AI citation share and "+9%" AI visibility; improved pages averaged 112% of their earlier level versus 72% for untouched pages. Ends with "Book a demo".Source
    • Your websiteHomepage (checked November 10, 2026): H1 "Sitefire - The marketing suite for the agentic web"; modules Agentic Actions, AI Monitoring and Agent Analytics; Spark, "the SEO and GEO agent", in the dashboard and over MCP; drafts pushed to the CMS (Webflow, Framer and more); results "+300% site visits from AI Search" (Pointhound) and "+78% AI referral traffic" (Jerry); testimonials from Jerry, Chamber (YC W26), Wemolo and DWS; CTAs "Start for free" and "talk to sales"; competitive data in "two to three weeks" and content results in "two to three months".Source
  • 04

    None of Sitefire's 13 blog posts shows an author, including the 354,004-answer citation study and the 13,500-post slop study, although the SlopShape study is credited to Sitefire's CEO. Original data with no visible expert behind it is harder for journalists to quote and for AI answers to attribute.

    Medium impactLow effort First result in 1 month Founder & CEO

    First action

    Add the author block to the two research studies.

    How to measure it

    New referring domains to studies

    Now
    –
    Target
    +15
    By
    90 days

    Steps

    1. 1

      Design the block

      Name, role, bio, links and dates.

      Founding Product Engineer 1 day

    2. 2

      Add structured data

      Article and Person markup.

      Founding Product Engineer 2 hours

    3. 3

      Assign authors

      One founder per post.

      Founder & CEO 1 hour

    4. 4

      Add methods boxes

      Dataset, dates and caveats on the two studies.

      Founder & CEO 1 day

    5. 5

      Add a citation line

      How to cite the study.

      Founder & CEO 1 hour

    6. 6

      Pitch the studies

      Send to ten marketing writers with the citation line.

      Founder & CEO 1 week

    First result
    1 month
    Cost
    $0
    Owner
    Founder & CEO

    Smallest test

    Add authors to the two studies and pitch them to five writers.

    Why it works

    Add an author block (name, role, a short bio, links to LinkedIn and the methods repository) and "published" and "updated" dates to every post, plus Article and Person structured data. Give the two studies a short methods box and a suggested citation line.

    Named expertise and clear dates are what editors and AI systems use to judge whether a source is trustworthy, and a suggested citation makes it easy to credit Sitefire by name.

    Risks and how to handle them

    • Founder time

      Write the bios once and reuse them.

    Evidence

    • Your website"How to Tell AI Slop From Human Writing" (September 2026): 13,500 blog posts analyzed; structure alone separated AI from human posts with a score of 98.0 out of 100; the main call to action is to play the five-round "Spot the Slop" game; no author is named on the post, although the underlying SlopShape study is credited to Sitefire's CEO, with a public verification repository.Source
    • Your websiteBlog index: 13 posts from January to October 2026, none showing an author; no newsletter signup, no calls to action and no links to the free tools.Source
    • Your website"Do AI Citations Actually Matter?" (July 2026): 354,004 AI answers; brands were mentioned in 15.1% of answers on days their page was not cited versus 32.9% when it was (2.2x); a citation was worth 9.2x for brands with no existing AI visibility; the post ends with a product description, with no gated download or newsletter signup.Source
  • 05

    Seven blog posts first published between January and April all carry the same sitemap lastmod, 2026-07-29, probably from a site migration. Crawlers use lastmod to decide what to re-fetch, and a date that changes for every page at once is usually ignored.

    Low impactLow effort First result in 2 weeks Founder & CTO

    First action

    Set lastmod from the post's front-matter date instead of the build date.

    How to measure it

    Time to re-crawl after an update

    Now
    –
    Target
    Under 3 days
    By
    60 days

    Steps

    1. 1

      Find the source

      Where the sitemap takes its dates from.

      Founder & CTO 1 hour

    2. 2

      Use real dates

      The content-update date from each post.

      Founder & CTO 2 hours

    3. 3

      Match the page

      Show the same updated date on the post.

      Founding Product Engineer 1 hour

    4. 4

      Date the tools

      Add lastmod to the tool pages.

      Founder & CTO 30 minutes

    5. 5

      Guard the dates

      A build check that fails if every lastmod changes at once.

      Founder & CTO 2 hours

    6. 6

      Watch crawlers

      Use Sitefire's own bot analytics to see re-fetches.

      Founder & CTO monthly

    First result
    2 weeks
    Cost
    $0
    Owner
    Founder & CTO

    Smallest test

    Fix the seven dates and compare crawler visits for a month.

    Why it works

    Generate lastmod from each post's real content-update date, show the same date on the page, and stop rebuilds and migrations from resetting it. Add lastmod to the three tool pages, which have none.

    Accurate dates let search and AI crawlers spend their visits on pages that actually changed, so refreshed studies get picked up sooner.

    Risks and how to handle them

    • Old posts look stale

      Refresh the best ones and date them honestly.

    Evidence

    • Your websiteSitemap (53 URLs, checked November 10, 2026): 13 blog posts, 2 case studies, 27 docs pages (including CMS guides for WordPress, Webflow, Framer and dotCMS), pricing, contact and legal pages, plus /slop-checker, /deslop and /spot-the-slop with no lastmod. No comparison, industry, use-case or German-language pages. Seven blog posts first published between January and April share one lastmod, 2026-07-29.Source
    • Your websiteBlog index: 13 posts from January to October 2026, none showing an author; no newsletter signup, no calls to action and no links to the free tools.Source
  • 06

    The citation study finds that a citation is worth 9.2x more mentions for brands with no AI visibility, which is exactly the reader who should start a trial. Yet the post ends with a product description, and the blog has no calls to action or newsletter. The product produces a first visibility report in under 5 minutes, so the gap between reading and doing is small.

    High impactLow effort First result in 2 weeks Founder & CEO

    First action

    Add the call to action to the citation study.

    How to measure it

    Blog-to-trial rate

    Now
    –
    Target
    2% of readers
    By
    60 days

    Steps

    1. 1

      Write the asks

      One next step per post, tied to its finding.

      Founder & CEO 1 day

    2. 2

      Build the block

      A reusable end-of-post component.

      Founding Product Engineer 1 day

    3. 3

      Pass the domain

      Prefill the domain field in onboarding.

      Founding Product Engineer 1 day

    4. 4

      Add a newsletter

      One field, monthly research digest.

      Founder & CEO 1 day

    5. 5

      Tag the source

      Trials tagged by post.

      Founder & CTO 2 hours

    6. 6

      Review

      Which posts start trials.

      Founder & CEO monthly

    First result
    2 weeks
    Cost
    $0
    Owner
    Founder & CEO

    Smallest test

    Add it to the two studies and compare trials for a month.

    Why it works

    Close each post with a specific next step tied to its finding: "See which of your pages get cited" after the citation study, "Check your last five posts" after the slop study, and so on. Each button opens the trial with the domain field first. Add a one-field newsletter signup for readers not ready to start.

    Readers of original data are at peak curiosity about their own numbers, and a short path from the finding to their own report converts that curiosity.

    Risks and how to handle them

    • Research looks like a sales pitch

      Keep the call to action at the end, after the caveats.

    Evidence

    • Your website"Do AI Citations Actually Matter?" (July 2026): 354,004 AI answers; brands were mentioned in 15.1% of answers on days their page was not cited versus 32.9% when it was (2.2x); a citation was worth 9.2x for brands with no existing AI visibility; the post ends with a product description, with no gated download or newsletter signup.Source
    • Your websiteBlog index: 13 posts from January to October 2026, none showing an author; no newsletter signup, no calls to action and no links to the free tools.Source
    • Your websiteQuick start docs: sign up with a work email, add the domain and key pages, review the generated prompt set and click "Analyze"; the first AI visibility report arrives in "under 5 minutes".Source

This quarter 12

The core plays for the next three months.

  • 07

    Peec AI publishes comparison pages against Ahrefs, Profound and Semrush; Sitefire has none. Sitefire's own YC launch states the difference in one line: "Unlike other GEO tools, sitefire actually tells you how to improve your AI visibility." Buyers comparing GEO tools are reading the rival's version of the comparison.

    High impactMedium effort First result in 1 month Founder & CEO

    First action

    Write the Peec AI page first, since the entry prices are almost identical.

    How to measure it

    Comparison-page conversions

    Now
    –
    Target
    15 a month
    By
    90 days

    Steps

    1. 1

      Pick the rivals

      The tools prospects name most on intro calls.

      Founder & CEO 1 day

    2. 2

      Check their facts

      Current pricing and features from their own pages.

      Founder & CEO 2 days

    3. 3

      Write the pages

      A fair table, where each tool fits, and proof.

      Founder & CEO 1 week

    4. 4

      Build the template

      A reusable comparison layout.

      Founding Product Engineer 2 days

    5. 5

      Publish and list

      Sitemap, llms.txt and footer.

      Founding Product Engineer 1 day

    6. 6

      Track in Sitefire

      Comparison prompts in Sitefire's own workspace.

      Founder & CEO weekly

    7. 7

      Refresh

      Re-check rival facts each quarter.

      Founder & CEO quarterly

    First result
    1 month
    Cost
    Founder time
    Owner
    Founder & CEO

    Smallest test

    Publish the Peec AI page and track conversions for a month.

    Why it works

    Write fair, dated comparison pages: Sitefire vs Peec AI, Sitefire vs Profound, and "monitoring vs action" for SEO suites that added AI tracking. Compare prompts per dollar, briefs and articles included, CMS publishing, bot and referral analytics, users and contract terms, and link the case studies with control groups as proof.

    Comparison searches come from buyers close to a decision, and these pages also shape how AI assistants describe the differences between tools.

    Risks and how to handle them

    • Out-of-date or unfair claims

      Date each page, cite rivals' own pages and correct quickly.

    Evidence

    • Competitor researchPeec AI pricing: Starter $95/mo (50 prompts), Pro $245/mo (150 prompts), Advanced $495/mo (350 prompts), unlimited users, 15% off with annual billing; public comparison pages against Ahrefs, Profound and Semrush; a free "GEO Report" billed as an "Instant brand audit".Source
    • Competitor researchProfound homepage: an "AI Marketer" that drafts AEO, content, PR and social work for human review; Agent Analytics, FactCheck and Sentiment; enterprise features including SOC 2; solutions pages for agencies and for travel, retail and pharma.Source
    • PressLaunch YC post: "Unlike other GEO tools, sitefire actually tells you how to improve your AI visibility"; the founders say they are "already working with Europe's largest brands" (BMW, Xtrackers, DWS) and ask for introductions to marketing and growth teams at Series A+ companies.Source
    • Your websiteSitemap (53 URLs, checked November 10, 2026): 13 blog posts, 2 case studies, 27 docs pages (including CMS guides for WordPress, Webflow, Framer and dotCMS), pricing, contact and legal pages, plus /slop-checker, /deslop and /spot-the-slop with no lastmod. No comparison, industry, use-case or German-language pages. Seven blog posts first published between January and April share one lastmod, 2026-07-29.Source
  • 08

    Sitefire Lite costs $249 a month for 150 tracked prompts; Peec AI Pro costs $245 for the same 150 prompts. On a prompt count, they look identical. What differs is that Lite also includes 8 action briefs and 3 AI-optimized articles a month, but the pricing page lists them as table rows, not as value.

    High impactLow effort First result in 3 weeks Founder & CEO

    First action

    Rewrite the Lite and Pro plan headers around articles and briefs.

    How to measure it

    Pricing-to-trial conversion

    Now
    –
    Target
    +20%
    By
    60 days

    Steps

    1. 1

      Price the outputs

      A fair market cost per article and brief.

      Founder & CEO 1 day

    2. 2

      Rewrite the headers

      Outputs first, limits second.

      Founder & CEO 2 hours

    3. 3

      Add the FAQ

      Trial, cancellation and prompts.

      Founder & CEO 2 hours

    4. 4

      Build it

      Update the pricing page with FAQ markup.

      Founding Product Engineer 1 day

    5. 5

      Test

      Compare trial starts against the old page.

      Founder & CTO 3 weeks

    6. 6

      Use on calls

      The same framing in founder intros.

      Founder & CEO ongoing

    First result
    3 weeks
    Cost
    $0
    Owner
    Founder & CEO

    Smallest test

    Change only the Lite header for two weeks.

    Why it works

    Lead each plan with outputs, not limits: "3 publish-ready articles and 8 page briefs a month", with a note on what an agency or freelancer typically charges for that work. Add a short "tracking vs tracking plus action" line under the plans, and a three-question FAQ (trial, cancellation, what counts as a prompt).

    Buyers compare like with like. When the comparison moves from prompts to finished work, Sitefire's price looks lower, and the FAQ answers trial questions before a founder call.

    Risks and how to handle them

    • Output comparisons sound inflated

      Use a conservative, sourced cost per article.

    Evidence

    • Your websitePricing: Lite "$249/month" (150 tracked prompts, 8 action briefs and 3 AI-optimized articles a month, MCP access; no CMS integration, Brand Kit, Personas or Agent Analytics); Pro "$499/month" (300 prompts, 18 briefs, 8 articles, CMS integration, Agent Analytics, onboarding call); Enterprise custom. "All plans include a 7-day free trial with monthly billing and monthly cancellation." No annual option and no FAQ.Source
    • Competitor researchPeec AI pricing: Starter $95/mo (50 prompts), Pro $245/mo (150 prompts), Advanced $495/mo (350 prompts), unlimited users, 15% off with annual billing; public comparison pages against Ahrefs, Profound and Semrush; a free "GEO Report" billed as an "Instant brand audit".Source
    • PressLaunch YC post: "Unlike other GEO tools, sitefire actually tells you how to improve your AI visibility"; the founders say they are "already working with Europe's largest brands" (BMW, Xtrackers, DWS) and ask for introductions to marketing and growth teams at Series A+ companies.Source
  • 09

    Every Sitefire plan is monthly with monthly cancellation, while the homepage says content results typically appear in two to three months. Customers can cancel before the work they paid for shows results. Peec AI offers 15% off annual billing.

    Medium impactLow effort First result in 1 month Founder & CEO

    First action

    Add an annual toggle to the pricing page.

    How to measure it

    Prepaid plan share

    Now
    –
    Target
    30% of new customers
    By
    90 days

    Steps

    1. 1

      Set the terms

      Annual discount and a quarterly option.

      Founder & CEO 1 day

    2. 2

      Set up billing

      Annual and quarterly prices in the billing system.

      Founder & CTO 1 day

    3. 3

      Add the toggle

      On the pricing page.

      Founding Product Engineer 1 day

    4. 4

      Offer at trial end

      An in-app prompt with the expected timeline.

      Founding Product Engineer 2 days

    5. 5

      Offer to monthly customers

      Email existing customers the switch.

      Founder & CEO 1 day

    6. 6

      Track

      Plan mix and churn by term.

      Founder & CEO monthly

    First result
    1 month
    Cost
    Discount cost
    Owner
    Founder & CEO

    Smallest test

    Offer annual to the next ten trial-ending customers.

    Why it works

    Offer an annual option on Lite and Pro at about two months free, and a quarterly option for teams that cannot commit for a year. Present it after the trial as "lock in the price while your first articles mature", with the expected timeline from the case studies.

    Longer terms match the product's real time to value, cut early churn, and bring cash forward for a two-person team.

    Risks and how to handle them

    • Lower revenue per customer

      Keep the discount near two months and measure churn.

    Builds on

    Evidence

    • Your websitePricing: Lite "$249/month" (150 tracked prompts, 8 action briefs and 3 AI-optimized articles a month, MCP access; no CMS integration, Brand Kit, Personas or Agent Analytics); Pro "$499/month" (300 prompts, 18 briefs, 8 articles, CMS integration, Agent Analytics, onboarding call); Enterprise custom. "All plans include a 7-day free trial with monthly billing and monthly cancellation." No annual option and no FAQ.Source
    • Your websiteHomepage (checked November 10, 2026): H1 "Sitefire - The marketing suite for the agentic web"; modules Agentic Actions, AI Monitoring and Agent Analytics; Spark, "the SEO and GEO agent", in the dashboard and over MCP; drafts pushed to the CMS (Webflow, Framer and more); results "+300% site visits from AI Search" (Pointhound) and "+78% AI referral traffic" (Jerry); testimonials from Jerry, Chamber (YC W26), Wemolo and DWS; CTAs "Start for free" and "talk to sales"; competitive data in "two to three weeks" and content results in "two to three months".Source
    • Competitor researchPeec AI pricing: Starter $95/mo (50 prompts), Pro $245/mo (150 prompts), Advanced $495/mo (350 prompts), unlimited users, 15% off with annual billing; public comparison pages against Ahrefs, Profound and Semrush; a free "GEO Report" billed as an "Instant brand audit".Source
  • 10

    Sitefire already produces a first AI visibility report within minutes of a domain being entered, but only after signup. Peec AI offers a free GEO Report as an "Instant brand audit". The citation study shows the biggest gains go to brands with no AI visibility today, who are also the least likely to start a paid trial blind.

    High impactMedium effort First result in 1 month Founding Product Engineer

    First action

    Define the ten-prompt snapshot from the existing onboarding flow.

    How to measure it

    Snapshot-to-trial rate

    Now
    –
    Target
    10%
    By
    90 days

    Steps

    1. 1

      Scope the snapshot

      Ten prompts, mentions, competitors, sources and one action.

      Founder & CEO 2 days

    2. 2

      Reuse onboarding

      Run the existing prompt generation without an account.

      Founding Product Engineer 1 week

    3. 3

      Limit cost

      Cache by domain and rate-limit.

      Founder & CTO 2 days

    4. 4

      Design the result

      A clear page with a shareable link.

      Founding Product Engineer 1 week

    5. 5

      Email capture

      Full report by email, then a trial offer.

      Founder & CEO 2 days

    6. 6

      Promote

      From the homepage, comparison pages and studies.

      Founder & CEO ongoing

    First result
    1 month
    Cost
    Model usage per snapshot plus engineering time
    Owner
    Founding Product Engineer

    Smallest test

    Run snapshots by hand for 20 inbound leads and compare their trial rate.

    Why it works

    Offer a public snapshot: enter a domain and get, for ten generated prompts, whether the brand is mentioned, which competitors appear instead and the top cited sources, plus one recommended action. Email the full report and offer the trial to act on it. Each snapshot has a shareable URL.

    Seeing a competitor named instead of you is the most persuasive moment in GEO, and a snapshot creates it before any commitment. The shareable link also spreads inside the buyer's team.

    Risks and how to handle them

    • Cost from heavy use

      Require a work email for the full report.

    • Ten prompts give a noisy picture

      Label it a snapshot and explain what the full tracking adds.

    Evidence

    • Your websiteQuick start docs: sign up with a work email, add the domain and key pages, review the generated prompt set and click "Analyze"; the first AI visibility report arrives in "under 5 minutes".Source
    • Competitor researchPeec AI pricing: Starter $95/mo (50 prompts), Pro $245/mo (150 prompts), Advanced $495/mo (350 prompts), unlimited users, 15% off with annual billing; public comparison pages against Ahrefs, Profound and Semrush; a free "GEO Report" billed as an "Instant brand audit".Source
    • Your website"Do AI Citations Actually Matter?" (July 2026): 354,004 AI answers; brands were mentioned in 15.1% of answers on days their page was not cited versus 32.9% when it was (2.2x); a citation was worth 9.2x for brands with no existing AI visibility; the post ends with a product description, with no gated download or newsletter signup.Source
  • 11

    Sitefire is based in Munich, its contact page offers English and German, and its published customers and testimonials include German brands such as BMW, DWS and Wemolo. The site itself has no German pages, so German marketers searching or asking AI assistants in German find only English material.

    Medium impactMedium effort First result in 2 months Founder & CEO

    First action

    Translate the homepage and pricing, and have a native speaker edit them.

    How to measure it

    German-path conversions

    Now
    –
    Target
    10 a month
    By
    90 days

    Steps

    1. 1

      Pick the pages

      Homepage, pricing, case studies and the ruling post.

      Founder & CEO 1 day

    2. 2

      Translate and edit

      Native editing, not raw machine output.

      Founder & CEO 1 week

    3. 3

      Add hreflang

      Between English and German pages.

      Founding Product Engineer 1 day

    4. 4

      German booking

      A German intro call option.

      Founder & CEO 1 hour

    5. 5

      Track German prompts

      In Sitefire's own workspace.

      Founder & CEO 1 day

    6. 6

      Review

      Conversions against the English path.

      Founder & CEO monthly

    First result
    2 months
    Cost
    Editing time
    Owner
    Founder & CEO

    Smallest test

    Publish a German homepage and run German prompts for a month.

    Why it works

    Publish German versions of the homepage, pricing, the two case studies and the Munich AI Overviews ruling post, with hreflang between languages. Add a German demo option that books a German-speaking founder, and track prompts in German for Sitefire's own visibility.

    Enterprise marketing teams in Germany often buy and approve in German, and German GEO content is thinner than English, so Sitefire can be the default German answer.

    Risks and how to handle them

    • Two languages to maintain

      Start with five pages and add more only when they convert.

    Evidence

    • Your websiteContact page: sales is "a 30-minute intro with a founder"; agents and scripts can book through a public API with no key needed; support by email or a shared Slack channel; languages English and German; registered in Munich.Source
    • PressLaunch YC post: "Unlike other GEO tools, sitefire actually tells you how to improve your AI visibility"; the founders say they are "already working with Europe's largest brands" (BMW, Xtrackers, DWS) and ask for introductions to marketing and growth teams at Series A+ companies.Source
    • Your websiteSitemap (53 URLs, checked November 10, 2026): 13 blog posts, 2 case studies, 27 docs pages (including CMS guides for WordPress, Webflow, Framer and dotCMS), pricing, contact and legal pages, plus /slop-checker, /deslop and /spot-the-slop with no lastmod. No comparison, industry, use-case or German-language pages. Seven blog posts first published between January and April share one lastmod, 2026-07-29.Source
    • Your websiteHomepage (checked November 10, 2026): H1 "Sitefire - The marketing suite for the agentic web"; modules Agentic Actions, AI Monitoring and Agent Analytics; Spark, "the SEO and GEO agent", in the dashboard and over MCP; drafts pushed to the CMS (Webflow, Framer and more); results "+300% site visits from AI Search" (Pointhound) and "+78% AI referral traffic" (Jerry); testimonials from Jerry, Chamber (YC W26), Wemolo and DWS; CTAs "Start for free" and "talk to sales"; competitive data in "two to three weeks" and content results in "two to three months".Source
  • 12

    Sitefire's proof clusters by industry: Pointhound in travel, Jerry and a pilot with "one of the largest insurances in the world" in insurance, and DWS and Xtrackers in asset management. The sitemap has no industry or use-case pages, while Profound runs solution pages for travel, retail and pharma.

    Medium impactMedium effort First result in 2 months Founder & CEO

    First action

    Draft the travel page around the Pointhound case study.

    How to measure it

    Industry-page conversions

    Now
    –
    Target
    10 a month
    By
    90 days

    Steps

    1. 1

      Pick three industries

      Where proof already exists.

      Founder & CEO 1 day

    2. 2

      Pull the data

      Common prompts and cited sources per industry from Sitefire's data.

      Founder & CTO 1 week

    3. 3

      Write the pages

      Questions, sources, proof and a sample prompt set.

      Founder & CEO 2 weeks

    4. 4

      Add starter sets

      Industry prompt sets loadable in the trial.

      Founding Product Engineer 1 week

    5. 5

      Publish and list

      Sitemap, llms.txt and navigation.

      Founding Product Engineer 1 day

    6. 6

      Measure

      Trials by industry page.

      Founder & CEO monthly

    First result
    2 months
    Cost
    Founder time
    Owner
    Founder & CEO

    Smallest test

    Publish the travel page and track trials for a month.

    Why it works

    Publish one page per industry with the questions buyers in that industry ask AI assistants, what typically gets cited (comparison sites, regulators, review sites), the matching case study, and a sample prompt set they can load into a trial.

    Marketers trust results from their own industry, and industry-specific pages answer the vertical questions AI assistants get, such as "best GEO tool for insurance".

    Risks and how to handle them

    • Exposing customer data

      Use aggregated data and published results only.

    Evidence

    • Your websitePointhound case study (award flight search): content waves from March 14, 2026; "+300%" site visits from AI Search (GA4), "+294%" AI bot traffic, about 45% of AI bot traffic now from the new content; Visibility Score and Citation Share from 0 to 1.0%.Source
    • Your websiteJerry case study: improved a selected set of pages and left a comparable set untouched as a control; three months later "+78%" AI referral traffic, "+56%" AI bot traffic, "+27%" AI citation share and "+9%" AI visibility; improved pages averaged 112% of their earlier level versus 72% for untouched pages. Ends with "Book a demo".Source
    • HiringFounding Product Engineer role text: will "join customer calls", scope agentic features with customers and build evals, feedback loops and customer-facing controls for agents; Sitefire "closed a pilot with one of the largest insurances in the world", is backed by Y Combinator, CDTM Ventures and Sequoia Scout, and must "scale our analytics layer to handle billions of rows".Source
    • Competitor researchProfound homepage: an "AI Marketer" that drafts AEO, content, PR and social work for human review; Agent Analytics, FactCheck and Sentiment; enterprise features including SOC 2; solutions pages for agencies and for travel, retail and pharma.Source
  • 13

    Sitefire sells to large European brands and runs an insurance pilot, and it connects to customers' CMS and analytics. The site has a privacy policy but no security or trust page, while Profound lists SOC 2 among its enterprise features. Procurement questions that a page could answer instead land on a two-person founding team.

    Medium impactMedium effort First result in 1 quarter Founder & CTO

    First action

    List the questions from the last three enterprise security reviews.

    How to measure it

    Enterprise procurement time

    Now
    –
    Target
    -30%
    By
    2 quarters

    Steps

    1. 1

      Collect questions

      From past security reviews.

      Founder & CEO 1 day

    2. 2

      Map the data

      Hosting, subprocessors and connector access.

      Founder & CTO 3 days

    3. 3

      Write the page

      Plain answers with dates.

      Founder & CTO 2 days

    4. 4

      Legal check

      Against the privacy policy and terms.

      Founder & CEO 1 week

    5. 5

      Answer sheet

      A downloadable questionnaire response.

      Founder & CTO 1 week

    6. 6

      Link from pricing

      Under the Enterprise plan.

      Founding Product Engineer 1 hour

    First result
    1 quarter
    Cost
    Founder time; certification costs only if pursued
    Owner
    Founder & CTO

    Smallest test

    Send the answer sheet to the next enterprise prospect and time the review.

    Why it works

    Publish a /security page: where data is hosted, which subprocessors are used, what access the CMS and analytics connectors need, how customer content is handled with AI models, GDPR roles, and the status of any certification work. Add a downloadable answer sheet for common security questionnaires.

    Enterprise deals stall in procurement, and a clear page shortens that step and signals maturity beyond team size.

    Risks and how to handle them

    • Overstating controls

      State only what is in place today, with dates.

    Evidence

    • Your websiteSite checks (November 10, 2026): no newsletter, no live chat and no trust or security page found; Google Tag Manager is the only third-party script; docs run on Nextra; the domain was registered in February 2026.Source
    • PressLaunch YC post: "Unlike other GEO tools, sitefire actually tells you how to improve your AI visibility"; the founders say they are "already working with Europe's largest brands" (BMW, Xtrackers, DWS) and ask for introductions to marketing and growth teams at Series A+ companies.Source
    • Competitor researchProfound homepage: an "AI Marketer" that drafts AEO, content, PR and social work for human review; Agent Analytics, FactCheck and Sentiment; enterprise features including SOC 2; solutions pages for agencies and for travel, retail and pharma.Source
  • 14

    Sitefire's YC launch asked for introductions to marketing teams at Series A+ companies, and one homepage testimonial already comes from a YC W26 company. YC founders trust each other's tool picks, and the batch network is the cheapest channel a two-person team has.

    Medium impactLow effort First result in 1 month Founder & CEO

    First action

    Write the YC offer and post it to the founders' network.

    How to measure it

    Referred customers

    Now
    –
    Target
    5 a quarter
    By
    2 quarters

    Steps

    1. 1

      Write the offer

      Trial length, setup help and reward.

      Founder & CEO 1 day

    2. 2

      Post it

      In the YC founder network.

      Founder & CEO 1 hour

    3. 3

      Define milestones

      Events that trigger the ask.

      Founding Product Engineer 2 days

    4. 4

      Automate the note

      Result plus a one-click intro request.

      Founding Product Engineer 1 week

    5. 5

      Track referrals

      A referral code per customer.

      Founder & CTO 2 days

    6. 6

      Thank and reward

      Apply credits monthly.

      Founder & CEO monthly

    First result
    1 month
    Cost
    Free months given as rewards
    Owner
    Founder & CEO

    Smallest test

    Ask the five happiest customers for one intro each.

    Why it works

    Create a standing offer for YC companies (for example, an extended trial plus a founder-run setup) and a simple referral reward, such as a free month for each referred customer. When a customer hits a milestone (first citation, a traffic jump), send a short note with their result and ask for one intro.

    Asking right after a visible win gets more yeses, and founder-to-founder intros close faster than cold demos.

    Risks and how to handle them

    • Discounted customers churn

      Tie the reward to the referred customer staying three months.

    Evidence

    • PressY Combinator company page: Winter 2026 batch, founded 2025, team size 2, Munich; one open job.Source
    • PressLaunch YC post: "Unlike other GEO tools, sitefire actually tells you how to improve your AI visibility"; the founders say they are "already working with Europe's largest brands" (BMW, Xtrackers, DWS) and ask for introductions to marketing and growth teams at Series A+ companies.Source
    • Your websiteHomepage (checked November 10, 2026): H1 "Sitefire - The marketing suite for the agentic web"; modules Agentic Actions, AI Monitoring and Agent Analytics; Spark, "the SEO and GEO agent", in the dashboard and over MCP; drafts pushed to the CMS (Webflow, Framer and more); results "+300% site visits from AI Search" (Pointhound) and "+78% AI referral traffic" (Jerry); testimonials from Jerry, Chamber (YC W26), Wemolo and DWS; CTAs "Start for free" and "talk to sales"; competitive data in "two to three weeks" and content results in "two to three months".Source
  • 15

    Sitefire publishes drafts straight to Webflow, Framer, WordPress and dotCMS, with a setup guide for each in its docs, and CMS publishing is a key Pro feature. Those guides live only on sitefire.ai, where a marketer has to know Sitefire already.

    Medium impactMedium effort First result in 2 months Founder & CTO

    First action

    Count trials by CMS and pick the first platform.

    How to measure it

    Directory-sourced trials

    Now
    –
    Target
    10 a month
    By
    2 quarters

    Steps

    1. 1

      Pick the platform

      The CMS most trials use.

      Founder & CEO 1 day

    2. 2

      Check requirements

      Each directory's listing and review rules.

      Founder & CTO 2 days

    3. 3

      Package the integration

      Meet the directory's technical requirements.

      Founding Product Engineer 2 weeks

    4. 4

      Write the listing

      Description, screenshots and trial link.

      Founder & CEO 2 days

    5. 5

      Build the hub

      An integrations page on sitefire.ai.

      Founding Product Engineer 2 days

    6. 6

      Measure

      Installs and trials per listing.

      Founder & CTO monthly

    First result
    2 months
    Cost
    Engineering time
    Owner
    Founder & CTO

    Smallest test

    Publish the integrations hub page and one directory listing.

    Why it works

    Package each integration for its platform's app marketplace, plugin directory or partner listing, with a short description, screenshots and a link into the trial with the CMS preselected. Start with the platforms most customers use, and add an "Integrations" hub page on sitefire.ai that the listings link back to.

    Marketers browse their CMS's directory when they want new capabilities, and a listing is lasting discovery at no ad cost.

    Risks and how to handle them

    • Directory review delays

      Submit to two directories in parallel.

    Evidence

    • Your websiteSitemap (53 URLs, checked November 10, 2026): 13 blog posts, 2 case studies, 27 docs pages (including CMS guides for WordPress, Webflow, Framer and dotCMS), pricing, contact and legal pages, plus /slop-checker, /deslop and /spot-the-slop with no lastmod. No comparison, industry, use-case or German-language pages. Seven blog posts first published between January and April share one lastmod, 2026-07-29.Source
    • Your websitePricing: Lite "$249/month" (150 tracked prompts, 8 action briefs and 3 AI-optimized articles a month, MCP access; no CMS integration, Brand Kit, Personas or Agent Analytics); Pro "$499/month" (300 prompts, 18 briefs, 8 articles, CMS integration, Agent Analytics, onboarding call); Enterprise custom. "All plans include a 7-day free trial with monthly billing and monthly cancellation." No annual option and no FAQ.Source
    • Your websiteHomepage (checked November 10, 2026): H1 "Sitefire - The marketing suite for the agentic web"; modules Agentic Actions, AI Monitoring and Agent Analytics; Spark, "the SEO and GEO agent", in the dashboard and over MCP; drafts pushed to the CMS (Webflow, Framer and more); results "+300% site visits from AI Search" (Pointhound) and "+78% AI referral traffic" (Jerry); testimonials from Jerry, Chamber (YC W26), Wemolo and DWS; CTAs "Start for free" and "talk to sales"; competitive data in "two to three weeks" and content results in "two to three months".Source
  • 16

    Sitefire has an MCP server that lets an AI assistant read actions, briefings and articles and trigger new ones, and its contact page says agents and scripts can book a demo through a public API with no key. Both are rare, but they are only described deep in the docs and the contact page, not where agents and their users discover tools.

    Medium impactLow effort First result in 1 month Founder & CTO

    First action

    Submit the MCP server to two public MCP directories.

    How to measure it

    Agent-sourced signups

    Now
    –
    Target
    10 a month
    By
    90 days

    Steps

    1. 1

      Pick directories

      The main public MCP server lists and catalogues.

      Founder & CTO 1 day

    2. 2

      Write the listing

      What it does, a short demo and setup.

      Founder & CTO 1 day

    3. 3

      Describe the booking API

      In llms.txt and a machine-readable description.

      Founding Product Engineer 1 day

    4. 4

      Homepage section

      Use Sitefire from your assistant.

      Founding Product Engineer 2 days

    5. 5

      Tag sources

      Signups and bookings by MCP and API.

      Founder & CTO 1 day

    6. 6

      Review

      Which listings send users.

      Founder & CTO monthly

    First result
    1 month
    Cost
    $0
    Owner
    Founder & CTO

    Smallest test

    List the server in one directory and count connections for a month.

    Why it works

    List the Sitefire MCP server in the main public MCP server directories and connector catalogues, with a short demo of an assistant pulling this week's actions. Describe the demo-booking API in llms.txt and in a machine-readable description so an agent asked to "book a GEO tool demo" can do it. Add an "Use Sitefire from your assistant" section to the homepage.

    Sitefire's pitch is marketing to the agentic web; being bookable and usable by agents is the most direct proof, and directory listings bring in technical marketers who already work through assistants.

    Risks and how to handle them

    • Spam bookings through the open API

      Require a work email and confirm bookings before the call.

    Builds on

    Evidence

    • Your websiteMCP server docs: gives an AI assistant access to Sitefire actions, tracked topics, briefings, generated articles and opportunities, and can trigger article generation; setup guides for four MCP clients and an optional skills bundle.Source
    • Your websiteContact page: sales is "a 30-minute intro with a founder"; agents and scripts can book through a public API with no key needed; support by email or a shared Slack channel; languages English and German; registered in Munich.Source
    • Your websiteHomepage (checked November 10, 2026): H1 "Sitefire - The marketing suite for the agentic web"; modules Agentic Actions, AI Monitoring and Agent Analytics; Spark, "the SEO and GEO agent", in the dashboard and over MCP; drafts pushed to the CMS (Webflow, Framer and more); results "+300% site visits from AI Search" (Pointhound) and "+78% AI referral traffic" (Jerry); testimonials from Jerry, Chamber (YC W26), Wemolo and DWS; CTAs "Start for free" and "talk to sales"; competitive data in "two to three weeks" and content results in "two to three months".Source
  • 17

    Sitefire's Events feature already marks model updates, search updates and each customer's own publishes on their charts. Customers pay monthly and can cancel monthly, and the Jerry case study shows that results only become clear when actions are compared against untouched pages over months.

    High impactMedium effort First result in 1 quarter Founding Product Engineer

    First action

    Hand-write the report for five customers this month.

    How to measure it

    Monthly churn

    Now
    –
    Target
    -30%
    By
    2 quarters

    Steps

    1. 1

      Write five by hand

      Learn what customers read and forward.

      Founder & CEO 1 week

    2. 2

      Define the template

      Metrics, events, control comparison and next actions.

      Founder & CEO 2 days

    3. 3

      Build the job

      Generate the report from Events and metrics.

      Founding Product Engineer 2 weeks

    4. 4

      Add the control view

      Improved against untouched pages.

      Founding Product Engineer 1 week

    5. 5

      Send and track

      Opens, forwards and replies.

      Founder & CTO monthly

    6. 6

      Compare churn

      Customers who read it against those who don't.

      Founder & CEO quarterly

    First result
    1 quarter
    Cost
    Engineering time
    Owner
    Founding Product Engineer

    Smallest test

    Send hand-written reports to five customers and ask if they forwarded them.

    Why it works

    Send each customer a monthly report: visibility, citations and AI traffic, each change tied to an event ("the article you published on the 4th now gets 12% of AI bot visits"), a comparison of improved against untouched pages, and the three actions for next month. Make it forwardable to the customer's boss.

    Customers renew when they can explain the result internally. A report that ties changes to actions turns Sitefire's data into the proof the marketer needs for their own budget.

    Risks and how to handle them

    • A bad month is visible

      Always pair the numbers with the next actions and the longer trend.

    Evidence

    • Your websiteChangelog (August 26, 2026): Events, chart annotations that add model and search updates automatically plus the customer's own publishes, and a new Actions page; a September 2 webinar promised "How a YC-backed scale-up grew organic clicks by 78% in three months using marketing agents", while the matching case study reports +78% AI referral traffic. No webinar recording is in the sitemap.Source
    • Your websitePricing: Lite "$249/month" (150 tracked prompts, 8 action briefs and 3 AI-optimized articles a month, MCP access; no CMS integration, Brand Kit, Personas or Agent Analytics); Pro "$499/month" (300 prompts, 18 briefs, 8 articles, CMS integration, Agent Analytics, onboarding call); Enterprise custom. "All plans include a 7-day free trial with monthly billing and monthly cancellation." No annual option and no FAQ.Source
    • Your websiteJerry case study: improved a selected set of pages and left a comparable set untouched as a control; three months later "+78%" AI referral traffic, "+56%" AI bot traffic, "+27%" AI citation share and "+9%" AI visibility; improved pages averaged 112% of their earlier level versus 72% for untouched pages. Ends with "Book a demo".Source
  • 18

    Sitefire's trial lasts 7 days, but its own homepage says competitive data takes two to three weeks and content results two to three months. A trial user can see a first report in minutes, yet the trial ends long before they see anything Sitefire did change.

    High impactMedium effort First result in 1 month Founder & CEO

    First action

    Measure what share of trial users publish a brief within 7 days.

    How to measure it

    Trial-to-paid rate

    Now
    –
    Target
    +25%
    By
    90 days

    Steps

    1. 1

      Measure today

      Share of trials that publish a brief, and their paid rate.

      Founder & CTO 1 day

    2. 2

      Map the 7 days

      One step per day toward a published action.

      Founder & CEO 2 days

    3. 3

      Build the checklist

      In-app progress toward the first publish.

      Founding Product Engineer 1 week

    4. 4

      Write the emails

      Day 1, 3, 6 and 7.

      Founder & CEO 2 days

    5. 5

      Schedule the re-check

      An automatic week-four comparison.

      Founding Product Engineer 1 week

    6. 6

      Compare cohorts

      Paid rate before and after the redesign.

      Founder & CTO monthly

    First result
    1 month
    Cost
    Engineering time and model usage for re-checks
    Owner
    Founder & CEO

    Smallest test

    Email the next 20 trial users a day-3 nudge to publish their first brief.

    Why it works

    Redesign the trial around one finished action. Day 1: the baseline report. Days 2-4: the first brief, edited and published. Day 7: a summary with what to expect and when, and a scheduled re-check at week four. Users who publish in the trial get the re-check free even if they have not paid yet, with a payment prompt attached to the result.

    People pay for progress they can see. Publishing one action in the trial creates an investment, and a dated re-check gives a reason to come back once results appear.

    Risks and how to handle them

    • Free re-checks without payment

      Offer them only to users who published in the trial.

    Evidence

    • Your websitePricing: Lite "$249/month" (150 tracked prompts, 8 action briefs and 3 AI-optimized articles a month, MCP access; no CMS integration, Brand Kit, Personas or Agent Analytics); Pro "$499/month" (300 prompts, 18 briefs, 8 articles, CMS integration, Agent Analytics, onboarding call); Enterprise custom. "All plans include a 7-day free trial with monthly billing and monthly cancellation." No annual option and no FAQ.Source
    • Your websiteHomepage (checked November 10, 2026): H1 "Sitefire - The marketing suite for the agentic web"; modules Agentic Actions, AI Monitoring and Agent Analytics; Spark, "the SEO and GEO agent", in the dashboard and over MCP; drafts pushed to the CMS (Webflow, Framer and more); results "+300% site visits from AI Search" (Pointhound) and "+78% AI referral traffic" (Jerry); testimonials from Jerry, Chamber (YC W26), Wemolo and DWS; CTAs "Start for free" and "talk to sales"; competitive data in "two to three weeks" and content results in "two to three months".Source
    • Your websiteQuick start docs: sign up with a work email, add the domain and key pages, review the generated prompt set and click "Analyze"; the first AI visibility report arrives in "under 5 minutes".Source

Big bets 3

Larger plays that can change the trajectory.

  • 19

    Sitefire has already published a 354,004-answer citation study and a 13,500-post content study, and its role text says the analytics layer is scaling toward "billions of rows". Each study is a one-off post. A recurring, dated index (which sources AI answers cite most in each industry, and how that shifts) would become the reference journalists and analysts return to.

    High impactHigh effort First result in 1 quarter Founder & CEO

    First action

    Choose five industries and the quarterly metrics from existing data.

    How to measure it

    Leads per index edition

    Now
    –
    Target
    200
    By
    2 quarters

    Steps

    1. 1

      Define the index

      Industries, metrics and method.

      Founder & CEO 1 week

    2. 2

      Build the pipeline

      Repeatable queries over collected answers.

      Founder & CTO 1 month

    3. 3

      Design the page

      Embeddable charts and a methods section.

      Founding Product Engineer 2 weeks

    4. 4

      Add "your position"

      A brand lookup behind a work email.

      Founding Product Engineer 2 weeks

    5. 5

      Pitch the press

      Exclusive first look to a few marketing writers.

      Founder & CEO 2 weeks

    6. 6

      Publish and repeat

      Same date each quarter.

      Founder & CEO quarterly

    First result
    1 quarter
    Cost
    Founder and engineering time; extra model queries
    Owner
    Founder & CEO

    Smallest test

    Publish one industry's index and measure links and leads.

    Why it works

    Publish a quarterly index: the most cited domains per industry, the shift since last quarter, how often brands are mentioned without being cited, and three practical takeaways, with charts free to embed with a link back. Offer each brand its own position in exchange for a work email, and pitch each edition to marketing press.

    Recurring data earns recurring links and coverage, and it positions a small company as the measurement authority in its category. The "your position" option turns readers into qualified leads.

    Risks and how to handle them

    • Method questioned

      Publish the method, caveats and a reproducibility repository, as with SlopShape.

    • Customer data exposure

      Use only answers from Sitefire's own prompt sets, aggregated.

    Builds on

    Evidence

    • Your website"Do AI Citations Actually Matter?" (July 2026): 354,004 AI answers; brands were mentioned in 15.1% of answers on days their page was not cited versus 32.9% when it was (2.2x); a citation was worth 9.2x for brands with no existing AI visibility; the post ends with a product description, with no gated download or newsletter signup.Source
    • Your website"How to Tell AI Slop From Human Writing" (September 2026): 13,500 blog posts analyzed; structure alone separated AI from human posts with a score of 98.0 out of 100; the main call to action is to play the five-round "Spot the Slop" game; no author is named on the post, although the underlying SlopShape study is credited to Sitefire's CEO, with a public verification repository.Source
    • HiringFounding Product Engineer role text: will "join customer calls", scope agentic features with customers and build evals, feedback loops and customer-facing controls for agents; Sitefire "closed a pilot with one of the largest insurances in the world", is backed by Y Combinator, CDTM Ventures and Sequoia Scout, and must "scale our analytics layer to handle billions of rows".Source
  • 20

    Profound runs a dedicated agency solution, and Peec AI gives every plan unlimited users and multiple projects. Sitefire's pricing page shows no agency or multi-client option, although agencies are the ones writing and publishing GEO content for many brands, exactly what Sitefire's briefs and articles automate.

    High impactHigh effort First result in 1 quarter Founder & CEO

    First action

    Interview five agencies already using GEO tools about how they bill clients.

    How to measure it

    Agency-managed workspaces

    Now
    –
    Target
    50
    By
    2 quarters

    Steps

    1. 1

      Interview agencies

      How they price and report GEO work.

      Founder & CEO 2 weeks

    2. 2

      Design the plan

      Pooled limits and partner pricing.

      Founder & CEO 1 week

    3. 3

      Build workspaces

      Multi-client switching and permissions.

      Founding Product Engineer 1 month

    4. 4

      White-label reports

      Agency branding on monthly reports.

      Founding Product Engineer 2 weeks

    5. 5

      Recruit partners

      Ten design-partner agencies.

      Founder & CEO 1 month

    6. 6

      Partner directory

      A page listing certified agencies.

      Founder & CEO 1 week

    7. 7

      Review

      Workspaces, churn and margin.

      Founder & CEO quarterly

    First result
    1 quarter
    Cost
    Engineering time and partner discounts
    Owner
    Founder & CEO

    Smallest test

    Give two agencies manual multi-workspace access for a month.

    Why it works

    Launch an agency plan: one login, many client workspaces, pooled prompts and articles, white-label monthly reports (from idea 17) and partner pricing. Recruit ten SEO and content agencies as design partners, list them in a partner directory on sitefire.ai, and send them inbound leads too small for direct sales.

    One agency brings many brands, and agencies sell the outcome to clients who would never find a two-person tool on their own.

    Risks and how to handle them

    • Agencies churn clients

      Let workspaces transfer to the brand directly.

    • Price undercuts direct plans

      Price on pooled volume, not a per-brand discount.

    Builds on

    Evidence

    • Competitor researchProfound homepage: an "AI Marketer" that drafts AEO, content, PR and social work for human review; Agent Analytics, FactCheck and Sentiment; enterprise features including SOC 2; solutions pages for agencies and for travel, retail and pharma.Source
    • Competitor researchPeec AI pricing: Starter $95/mo (50 prompts), Pro $245/mo (150 prompts), Advanced $495/mo (350 prompts), unlimited users, 15% off with annual billing; public comparison pages against Ahrefs, Profound and Semrush; a free "GEO Report" billed as an "Instant brand audit".Source
    • Your websitePricing: Lite "$249/month" (150 tracked prompts, 8 action briefs and 3 AI-optimized articles a month, MCP access; no CMS integration, Brand Kit, Personas or Agent Analytics); Pro "$499/month" (300 prompts, 18 briefs, 8 articles, CMS integration, Agent Analytics, onboarding call); Enterprise custom. "All plans include a 7-day free trial with monthly billing and monthly cancellation." No annual option and no FAQ.Source
  • 21

    Sitefire's strongest enterprise signals sit in regulated industries: a pilot with "one of the largest insurances in the world", Jerry's controlled +78% result in insurance shopping, and DWS and Xtrackers in asset management. These buyers have large budgets and specific needs: accurate claims in AI answers, compliance review before publishing, and an audit trail.

    High impactHigh effort First result in 2 quarters Founder & CEO

    First action

    List what the insurer pilot needed that standard Enterprise did not provide.

    How to measure it

    Regulated-industry Enterprise deals

    Now
    –
    Target
    3 signed
    By
    2 quarters

    Steps

    1. 1

      Learn from the pilot

      Requirements and blockers from the insurer.

      Founder & CEO 1 week

    2. 2

      Build prompt libraries

      Insurance and asset-management prompt sets.

      Founder & CEO 2 weeks

    3. 3

      Add an approval step

      Compliance sign-off before CMS publishing.

      Founding Product Engineer 1 month

    4. 4

      Add an audit log

      Who changed and approved what.

      Founder & CTO 2 weeks

    5. 5

      Write the offer

      A one-page package and pricing.

      Founder & CEO 1 week

    6. 6

      Sell it

      To ten target firms through warm intros.

      Founder & CEO 1 quarter

    First result
    2 quarters
    Cost
    Engineering time for approvals and audit log
    Owner
    Founder & CEO

    Smallest test

    Offer the approval step to the current pilot and measure adoption.

    Why it works

    Package an Enterprise offer for insurance and financial services: industry prompt libraries, sentiment and accuracy tracking on every prompt, a compliance approval step before anything is published to the CMS, an audit log of edits, and quarterly business reviews. Publish one anonymized insurer result once the pilot allows it.

    Regulated buyers pay more for control than for features, and a ready-made package shortens a long enterprise sale for a team that cannot afford custom work on every deal.

    Risks and how to handle them

    • Long sales cycles drain a small team

      Use the security page and answer sheet from idea 13 to shorten procurement.

    Builds on

    Evidence

    • HiringFounding Product Engineer role text: will "join customer calls", scope agentic features with customers and build evals, feedback loops and customer-facing controls for agents; Sitefire "closed a pilot with one of the largest insurances in the world", is backed by Y Combinator, CDTM Ventures and Sequoia Scout, and must "scale our analytics layer to handle billions of rows".Source
    • Your websiteJerry case study: improved a selected set of pages and left a comparable set untouched as a control; three months later "+78%" AI referral traffic, "+56%" AI bot traffic, "+27%" AI citation share and "+9%" AI visibility; improved pages averaged 112% of their earlier level versus 72% for untouched pages. Ends with "Book a demo".Source
    • PressLaunch YC post: "Unlike other GEO tools, sitefire actually tells you how to improve your AI visibility"; the founders say they are "already working with Europe's largest brands" (BMW, Xtrackers, DWS) and ask for introductions to marketing and growth teams at Series A+ companies.Source

Wildcards 2

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

  • 22

    Real visitors pass Core Web Vitals on sitefire.ai today (p75 LCP 1.18 s), so this is not a crisis. But Google's mobile lab test scores 61, with the dashboard screenshot as the 12.3-second largest element, 1.76 MB of images on a 2.9 MB page, and field LCP trending 52% slower.

    Low impactLow effort First result in 1 week Founding Product Engineer

    First action

    Re-encode the hero dashboard image with a mobile size.

    How to measure it

    Mobile lab score

    Now
    –
    Target
    85+
    By
    30 days

    Steps

    1. 1

      Re-encode the hero

      AVIF or WebP with mobile and desktop sizes.

      Founding Product Engineer 2 hours

    2. 2

      Prioritize it

      Priority loading for the hero only.

      Founding Product Engineer 1 hour

    3. 3

      Lazy-load the rest

      Screenshots below the fold.

      Founding Product Engineer 1 hour

    4. 4

      Trim JavaScript

      Remove unused code on the homepage.

      Founder & CTO 1 day

    5. 5

      Set a budget

      A page-weight limit checked on each deploy.

      Founder & CTO 2 hours

    6. 6

      Re-test

      PageSpeed and the field trend.

      Founding Product Engineer monthly

    First result
    1 week
    Cost
    $0
    Owner
    Founding Product Engineer

    Smallest test

    Swap the hero image and re-run the mobile test.

    Why it works

    Serve the dashboard screenshot as a properly sized AVIF or WebP with a mobile variant, mark it as the priority image, lazy-load the screenshots below it, and trim unused JavaScript on the homepage. Re-run PageSpeed and watch the field LCP trend.

    Keeping the field number healthy protects the experience on slower phones and avoids a GEO vendor's homepage scoring poorly when prospects run their own audits.

    Risks and how to handle them

    • Blurry product screenshots

      Check image quality on a retina display before shipping.

    Evidence

    • PageSpeedMobile test (November 10, 2026): score 61 (desktop 96); lab LCP 12.3 s with the dashboard screenshot as the LCP element; 2.9 MB page, 1.76 MB of images, about 696 KB of image savings. Real visitors pass Core Web Vitals (p75 LCP 1.18 s, INP 116 ms, CLS 0.01), though field LCP is trending 52% slower.Source
  • 23

    Sitefire writes 3 to 8 AI-optimized articles a month for each customer, and its own SlopShape research shows that structure alone gives AI-written posts away with a 98.0 score, even after rewording. Buyers' biggest fear about AI articles is that they read as generic. Sitefire has the measure to prove its articles don't, and isn't using it in the product.

    Medium impactMedium effort First result in 1 month Founder & CTO

    First action

    Score last month's delivered articles and see how many would fail.

    How to measure it

    Articles passing the slop check

    Now
    –
    Target
    95%
    By
    2 quarters

    Steps

    1. 1

      Score past articles

      Baseline from last month's deliveries.

      Founder & CTO 2 days

    2. 2

      Add the check

      Run it before each delivery.

      Founding Product Engineer 1 week

    3. 3

      Show it in the editor

      Score and flagged features.

      Founding Product Engineer 1 week

    4. 4

      Auto-restructure

      Rewrite structure, not just wording, on failure.

      Founder & CTO 2 weeks

    5. 5

      Publish the average

      On pricing and in sales decks.

      Founder & CEO 1 day

    6. 6

      Track edits

      Customer edit rate before and after.

      Founder & CTO monthly

    First result
    1 month
    Cost
    Engineering time and extra model calls per article
    Owner
    Founder & CTO

    Smallest test

    Score ten delivered articles and share the results with their customers.

    Why it works

    Run every generated article through the SlopShape structural check before delivery, show the score and the flagged features in the editor, and auto-restructure articles that fail. Publish the average score of delivered articles on the pricing page, next to the article counts.

    It turns research into a product guarantee no rival can easily copy, answers the "AI slop" objection on sales calls, and gives the Founding Product Engineer's evals and review flows a concrete first metric.

    Risks and how to handle them

    • Optimizing to the score, not the reader

      Track customer edit rates and AI citations alongside the score.

    Builds on

    Evidence

    • Your website"How to Tell AI Slop From Human Writing" (September 2026): 13,500 blog posts analyzed; structure alone separated AI from human posts with a score of 98.0 out of 100; the main call to action is to play the five-round "Spot the Slop" game; no author is named on the post, although the underlying SlopShape study is credited to Sitefire's CEO, with a public verification repository.Source
    • Your websitePricing: Lite "$249/month" (150 tracked prompts, 8 action briefs and 3 AI-optimized articles a month, MCP access; no CMS integration, Brand Kit, Personas or Agent Analytics); Pro "$499/month" (300 prompts, 18 briefs, 8 articles, CMS integration, Agent Analytics, onboarding call); Enterprise custom. "All plans include a 7-day free trial with monthly billing and monthly cancellation." No annual option and no FAQ.Source
    • HiringFounding Product Engineer role text: will "join customer calls", scope agentic features with customers and build evals, feedback loops and customer-facing controls for agents; Sitefire "closed a pilot with one of the largest insurances in the world", is backed by Y Combinator, CDTM Ventures and Sequoia Scout, and must "scale our analytics layer to handle billions of rows".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.

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
ProgrammableDigital

Show Sitefire’s product working

Turn screenshots or the website into an animated product demo in its own brand, for the homepage, launch posts and outreach in this playbook.

Make a demo (opens programmable.digital in a new tab)

More growth playbooks

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

  • Closure Intelligence growth strategyClosure builds an AI "digital analyst" for law enforcement: investigators upload case evidence and Closure transcribes and searches jail…22 growth ideas
  • A Place for Mom growth strategyA senior care advisory marketplace: families search listings and reviews for senior living communities and home care agencies, and local…24 growth ideas
  • M-KOPA growth strategyA fintech that finances smartphones, electric motorbikes and other assets for Every Day Earners on daily, weekly or monthly payments, with…21 growth ideas
  • Parachute Health growth strategyA digital ordering network for home medical equipment (DME/HME): clinicians ePrescribe equipment and supplies, HME providers receive…20 growth ideas
  • Ditto growth strategyAn edge sync platform: an offline-first mobile database with peer-to-peer mesh sync and CRDT conflict resolution (Edge SDK, Cloud, Server…22 growth ideas
  • The HOTH growth strategySEO and AI visibility services: managed SEO (HOTH X), a la carte link building, content, digital PR, earned media, reputation management…21 growth ideas
  • StackAdapt growth strategyAn AI-powered programmatic advertising and marketing platform (DSP) covering CTV, display, native, video, audio, DOOH, in-game, email…21 growth ideas
  • Optro growth strategyThe GRC Intelligence Platform (formerly AuditBoard): software and AI agents for internal audit, SOX, risk, compliance, infosec…21 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. Sitefire did not commission, review or endorse it, and Programmable is not affiliated with Sitefire. 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 Sitefire runs its business.

Something wrong, or do you represent Sitefire? 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 308 ideas

    Restore the research tools, fix llms.txt and the sitemap dates, align the 78% claim, add authors and calls to action, then the first comparison page and pricing value framing.

  2. Days 31 to 6012 ideas

    Annual billing, a free snapshot, German and industry pages, a trust page, YC referrals, CMS directories, MCP discovery, the monthly report, the trial redesign, homepage weight and the slop score in the product.

  3. Days 61 to 903 ideas

    Commit to the big bets: a quarterly citation index, an agency plan and a regulated-industry package.

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 recommendations that say what to do, not checklists. The most cited problem: not knowing why AI answers skip their brand.

  • Articles teams are proud to publish
  • Monitoring tools that stop at dashboards
  • In their words: “AI visibility”

Market

Competitive landscape

2 competitors compared: Peec AI and Profound.

  • Gap to exploit: Proof with control groups rather than before-and-after charts
  • Peec AI: Analytics-led rather than action-led
  • Profound: Broad enterprise scope for smaller teams

Hiring signals

Sitefire is a two-person team hiring its first engineer, a Founding Product Engineer in Munich who will join customer calls, scope agentic features with customers, and build evals, feedback loops and customer-facing…

  • Founding Product Engineer (Munich): First hire: customer-facing agent features, evals and review flows. Ideas 10, 17, 18, 23.

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 firstNew accounts get 100,000 free enrichment credits with code BETA100K (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)
  • ProgrammableMediaAn AI note taker, interpreter and webinar host for your calls (opens programmable.media 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