ProgrammableMarketing/parametric.company (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 · 12
  • Big bets · 3
  • Wildcards · 2
  • How we can help
Generated Nov 10, 2026
  1. Programmable
  2. Playbooks
  3. Parametric
Parametric logo

growth playbook · Nov 10, 2026

parametric.company (opens in a new tab)

Parametric growth playbook

Diagnosis

Parametric (Y Combinator Fall 2025, seven people, a public benefit corporation in San Francisco) has strong proof for its stage: Bit went from 0% to 75% autonomous robot time between February and July 2026, with 130,000 manipulation episodes and 3,500+ hours on real robots, a fleet app (Bastion) and a hardware line that is moving from prototype to production.

The website does not yet sell any of it. The search listing promises "the robot you hire for warehouse work", but the site has nine pages, no explanation of what hiring Bit means, no price shape, no task pages, no customer story, and a "Hire Bit" form that asks only for name, email and a message.

This playbook first makes hiring Bit easy to understand and qualify (a scoping form, the hire terms, task pages, what happens in the 25% of time Bit asks for help), then builds proof (a first case study, a bring-your-own-box trial, customer reports from Bastion, a service commitment backed by the new technician), and finally bets on per-unit pricing, the robot laundromat as a product, and a public task benchmark built from the team's evaluation research.

Growth Score

6/10

Biggest Opportunity

Turn real operating data (autonomy, hours, help requests, units) into the buyer's case: what Bit does, what it costs, how fast it learns their task and what happens when it gets stuck.

Biggest Risk

Better-funded rivals such as Formic and Dexterity already publish uptime, lead times, big-name customers and a clear deployment offer, and a seven-person team can overpromise on tasks Bit cannot yet do reliably.

01

Start here

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

  1. 1

    Turn "Hire Bit" into a short scoping form about the work, with a "how hiring Bit works" section above it

    Write the four "how hiring Bit works" steps and the eight scoping questions, and agree a reply time the team can keep.

    High impact First result in 2 weeks
  2. 2

    Publish the shape of the hire: what a month of Bit includes and what the customer provides

    Write down the terms of the current customer deployments and pick the unit you want to sell by.

    High impact First result in 1 month
  3. 3

    Explain the other 25%: show buyers how Bit asks for help and who answers

    Write three sentences on who answers help requests today and how quickly.

    Medium impact First result in 2 weeks

02

Your idea portfolio

24 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 · 2Big bets · 8Fill-ins · 8Reconsider · 6
Effort →

Showing 24 of 24 ideas

This week 7

Quick wins with a clear first action.

  • 01

    Parametric's search listing promises "the robot you hire for warehouse work", and every call to action on the site says "Hire Bit". But the page behind it is a general contact form: an inquiry-type menu that mixes "ordering a robot" with press and careers, then name, email and a free-text message. Nothing on the site explains what hiring Bit involves: no pilot steps, no timeline, no price shape, no site requirements. A warehouse manager has to write an open-ended email to find out whether Bit fits their work, and the team gets messages it cannot qualify without a call. Rivals spell it out: Formic offers one fixed monthly price, a 12-week average lead time and an on-site evaluation; Dexterity says "Start a deployment".

    High impactLow effort First result in 2 weeks Founder/CEO

    First action

    Write the four "how hiring Bit works" steps and the eight scoping questions, and agree a reply time the team can keep.

    How to measure it

    Hire requests with task, volume and site filled in

    Now
    –
    Target
    3x the current number of buyer inquiries
    By
    60 days

    Steps

    1. 1

      Read the last inquiries

      Go through recent "ordering a robot" messages and list what the team always had to ask before it could answer.

      Founder/CEO half a day

    2. 2

      Write the four steps

      Tell us the work, we try it on Bit, first robot on site with Bastion, scale with more robots. Include rough durations.

      Founder/CEO 2 hours

    3. 3

      Write the scoping questions

      Task, items or box sizes, units per shift, shifts per day, people on the task today, site city, start date. Keep it under two minutes.

      Founder/CEO 2 hours

    4. 4

      Build the Hire Bit page

      A new page for hiring, with the steps above the form; leave press and careers on the general contact form.

      Founder/CTO 1 day

    5. 5

      Set the reply promise

      State a reply time on the page and send each request to a shared inbox and a pipeline sheet tagged by task.

      Founder/CEO half a day

    6. 6

      Point every call to action there

      Header, footer, homepage buttons and the blog's "Hire Bit" link.

      Founder/CTO 1 hour

    7. 7

      Review weekly

      Count requests by task and volume, and note which ones Bit can do today.

      Founder/CEO weekly

    First result
    2 weeks
    Cost
    $0, about two days of team time
    Owner
    Founder/CEO

    Smallest test

    Add the four steps and three questions (task, units per shift, site city) to the current form this week, and compare the next ten inquiries with the last ten.

    Why it works

    Split /contact in two. Keep a short general form for press, careers and questions. Make "Hire Bit" its own page with three parts. (1) How hiring Bit works, in four steps the team already follows: tell us the work, we try it on Bit (from a sample or a video), a first robot at your site with Bastion, then more robots as autonomy rises. (2) A scoping form of six to eight questions: which task (box assembly, pack out, taping, folding, other), what items or box sizes, units per shift, shifts per day, current headcount on that task, site city, and when they would want to start. (3) What happens next: a reply within a set time, with a yes, no or "send us a sample". Route answers into a simple pipeline sheet, tagged by task, so the team can see which work buyers actually bring.

    Robot buyers need to know if their task fits before they will talk to sales. Specific questions make the form feel like a fit check rather than a cold email, and they give a seven-person team the details to reply with a useful answer instead of a call request. Tagging by task also shows which of the four tasks to build pages, demos and pricing around first.

    Ready to use

    ChecklistHire Bit scoping form

    [ ] Which work? Box assembly / Pack out / Taping / Folding / Other
    [ ] What items or box sizes?
    [ ] Units per shift today
    [ ] Shifts per day
    [ ] People on this task today
    [ ] Site city
    [ ] When would you want to start?
    [ ] Email (required)
    
    How hiring Bit works:
    1. Tell us the work
    2. We try it on Bit
    3. A first robot at your site, run with Bastion
    4. More robots as autonomy rises
    
    We reply within [X] working days.

    Risks and how to handle them

    • A longer form means fewer messages

      Make only task and email required, and keep a one-line "just ask a question" link to the general form.

    • Buyers ask for tasks Bit cannot do yet

      Reply honestly and keep the request as demand data for the roadmap.

    Evidence

    • Your websiteContact page "Hire Bit": "Tell us about the work, or ask us anything." The form asks for an inquiry type (general inquiry, ordering a robot, press, careers or other), name, email and message, plus an optional how-you-heard and website field. No pricing, no FAQ and no reply time; after sending, it promises a reply by email.Source
    • Your websiteSearch listing: title "Parametric | Bit, the robot you hire for warehouse work"; description "Parametric builds Bit, a two-armed mobile robot that businesses hire for repetitive warehouse work such as box assembly, pack out, taping and folding." The homepage headline changed from "AUTOMATE REPETITION TODAY" to "Meet Bit." over the past year.Source
    • Your websiteHomepage: H1 "Meet Bit." and "Commercial autonomy made simple."; under "Bit learns repetitive work." it lists "Box assembly", "Pack out", "Taping" and "Folding"; "75%" of robot time "was autonomous in July 2026. In February it was 0%."; "130,000" episodes of manipulation data and "3,500+" hours of real-world robot operation. The page gives no price, pilot process, throughput, site requirements, customer logos or testimonials, and "Learn more" links back to the same page.Source
    • Competitor researchRivals' own homepages: Formic sells "Full Service Automation for mid-market manufacturers. No CapEx. No Risk. No Problems." at one fixed monthly price, and shows 99.30% uptime over the last 12 months, a 12-week average lead time and an on-site evaluation offer; Dexterity shows a FedEx endorsement and "Start a deployment"; Verne Robotics ("Physical AI for building a better world") offers "Book a demo" and fine-tuning its model to your tasks.Source
  • 02

    "Hire" is the core promise, but the site never says what is being hired: by the hour, by the month, with or without remote help, maintenance and Bastion. Formic leads with "No CapEx" and one fixed monthly price. Buyers comparing the two will assume the unknown option is the expensive one.

    High impactLow effort First result in 1 month Founder/CEO

    First action

    Write down the terms of the current customer deployments and pick the unit you want to sell by.

    How to measure it

    Request-to-site-visit rate

    Now
    –
    Target
    +30%
    By
    90 days

    Steps

    1. 1

      List current terms

      What existing customers pay for, and what Parametric covers.

      Founder/CEO half a day

    2. 2

      Choose the unit

      Per robot per month or per robot-hour; one unit only.

      Founder/CEO 1 day

    3. 3

      List inclusions

      Robot, setup, Bastion, remote help, maintenance and repairs.

      Founder/CTO 2 hours

    4. 4

      List customer duties

      Space, power, network, materials and a contact on site.

      Robotics Technician 2 hours

    5. 5

      Publish

      On the Hire Bit page, with a "from" price if possible.

      Founder/CTO 1 hour

    6. 6

      Track objections

      Log price questions from each call and adjust the copy.

      Founder/CEO monthly

    First result
    1 month
    Cost
    $0
    Owner
    Founder/CEO

    Smallest test

    Add the inclusions list without any price this week.

    Why it works

    Add a short "What hiring Bit includes" block to the Hire Bit page: the unit of hire (per robot per month, or per robot-hour), what is included (the robot, setup, Bastion, remote help when Bit asks, maintenance and repairs), what the customer provides (floor space, power, network, materials), the minimum term and how to stop. A "from" price or a price range is better than nothing; if the team is not ready, state the unit and what is included.

    Robotics buyers fear hidden capital and integration costs. Naming the unit and the inclusions removes the biggest unknown and filters out buyers who cannot afford it before anyone spends time on a call.

    Risks and how to handle them

    • Locking in a price too early

      Publish the unit and inclusions now and a "from" price only when the team is confident.

    Builds on

    Evidence

    • Your websiteHomepage: H1 "Meet Bit." and "Commercial autonomy made simple."; under "Bit learns repetitive work." it lists "Box assembly", "Pack out", "Taping" and "Folding"; "75%" of robot time "was autonomous in July 2026. In February it was 0%."; "130,000" episodes of manipulation data and "3,500+" hours of real-world robot operation. The page gives no price, pilot process, throughput, site requirements, customer logos or testimonials, and "Learn more" links back to the same page.Source
    • Your websiteContact page "Hire Bit": "Tell us about the work, or ask us anything." The form asks for an inquiry type (general inquiry, ordering a robot, press, careers or other), name, email and message, plus an optional how-you-heard and website field. No pricing, no FAQ and no reply time; after sending, it promises a reply by email.Source
    • Competitor researchRivals' own homepages: Formic sells "Full Service Automation for mid-market manufacturers. No CapEx. No Risk. No Problems." at one fixed monthly price, and shows 99.30% uptime over the last 12 months, a 12-week average lead time and an on-site evaluation offer; Dexterity shows a FedEx endorsement and "Start a deployment"; Verne Robotics ("Physical AI for building a better world") offers "Book a demo" and fine-tuning its model to your tasks.Source
  • 03

    The homepage lists four tasks (box assembly, pack out, taping, folding) as single words, and the sitemap has only nine URLs: there is no page a buyer searching for "box folding robot" or "automated pack out" could land on, and nothing showing what Bit handles in each task.

    High impactMedium effort First result in 1 month Founder/CEO

    First action

    Pull the best recent clip of each task from robot logs.

    How to measure it

    Task-page hire requests

    Now
    –
    Target
    40% of all requests
    By
    90 days

    Steps

    1. 1

      Pick clips

      One clean video per task from recent runs.

      Founder/CTO 1 day

    2. 2

      Pull numbers

      Autonomy and units per hour per task from Bastion.

      Founder/CTO 1 day

    3. 3

      Write limits

      Item sizes and what Bit cannot do yet.

      Founder/CEO 1 day

    4. 4

      Build pages

      Four pages with titles and descriptions written around the task.

      Founder/CTO 3 days

    5. 5

      Preset the form

      Each page opens the scoping form on its task.

      Founder/CTO 2 hours

    6. 6

      Add to sitemap

      And link from the homepage task list.

      Founder/CTO 1 hour

    First result
    1 month
    Cost
    Team time
    Owner
    Founder/CEO

    Smallest test

    Publish the box assembly page first.

    Why it works

    Publish /bit/box-assembly, /bit/pack-out, /bit/taping and /bit/folding. Each page: a 30-second video of Bit doing the task, the items or box sizes it handles today, current autonomy on that task, rough units per hour, what it cannot do yet, and the Hire Bit form preset to that task.

    Buyers search by the job they need done, not by robot name. Task pages capture that search, answer "can it do mine?" and let the scoping form start with the right task.

    Risks and how to handle them

    • Numbers change quickly

      Date each figure and refresh monthly.

    Builds on

    Evidence

    • Your websiteSitemap (9 URLs, checked November 9, 2026): the homepage, the blog index and three posts, team, contact, privacy and terms. There are no pages for individual tasks, industries, pricing, Bastion, customer stories, safety or FAQs; the newest post is dated July 13, 2026.Source
    • Your websiteHomepage: H1 "Meet Bit." and "Commercial autonomy made simple."; under "Bit learns repetitive work." it lists "Box assembly", "Pack out", "Taping" and "Folding"; "75%" of robot time "was autonomous in July 2026. In February it was 0%."; "130,000" episodes of manipulation data and "3,500+" hours of real-world robot operation. The page gives no price, pilot process, throughput, site requirements, customer logos or testimonials, and "Learn more" links back to the same page.Source
  • 04

    The homepage says 75% of robot time was autonomous in July 2026. A buyer reads that as "a quarter of the time it doesn't work". The answer is already on the page but unexplained: the example dashboard shows "Bit asks for help" ("The box flap did not fold after two tries.") with "Ask" and "Take over" buttons.

    Medium impactLow effort First result in 2 weeks Founder/CTO

    First action

    Write three sentences on who answers help requests today and how quickly.

    How to measure it

    Reliability objections on first calls

    Now
    –
    Target
    -30%
    By
    60 days

    Steps

    1. 1

      Describe today's process

      Who sees help requests, response time and handover.

      Founder/CTO 2 hours

    2. 2

      Write the section

      Ask, answer, learn: three short lines under the number.

      Founder/CEO 2 hours

    3. 3

      Add a clip

      A real help request and takeover, sped up.

      Founder/CTO 1 day

    4. 4

      State coverage

      Whether help time is included in the hire.

      Founder/CEO 1 hour

    5. 5

      Publish

      On the homepage and the Hire Bit page.

      Founder/CTO 1 hour

    6. 6

      Listen

      Note reliability questions on calls.

      Founder/CEO monthly

    First result
    2 weeks
    Cost
    $0
    Owner
    Founder/CTO

    Smallest test

    Add one sentence under the 75% figure today.

    Why it works

    Add a short section under the autonomy number: when Bit is unsure it asks for help; a remote operator answers or takes over; the work still gets done; each help request becomes training data, which is why autonomy went from 0% to 75%. Say who answers (Parametric or the customer's staff), how fast, and whether help time is included in the hire.

    It turns the weakest-looking number into the strongest part of the offer: the work is covered 100% of the time, and the robot improves from every help request.

    Risks and how to handle them

    • Buyers worry about remote operators watching their site

      Say what the operator sees and how access is controlled.

    Evidence

    • Your websiteHomepage: H1 "Meet Bit." and "Commercial autonomy made simple."; under "Bit learns repetitive work." it lists "Box assembly", "Pack out", "Taping" and "Folding"; "75%" of robot time "was autonomous in July 2026. In February it was 0%."; "130,000" episodes of manipulation data and "3,500+" hours of real-world robot operation. The page gives no price, pilot process, throughput, site requirements, customer logos or testimonials, and "Learn more" links back to the same page.Source
    • Your websiteHomepage fleet sections: "Bit works in fleets." has no body text; "We'll hand you tools to run your fleet." says only "Bastion is the app we use to run our own robots." above an example dashboard ("7 robots at 2 sites"; tasks "Fold towels", "Fold shipping box" and "Fold T-shirts"; "Bit asks for help": "The box flap did not fold after two tries." with "Ask" and "Take over" buttons). "Grow your business alongside Bit." shows only the hardware roadmap from Bit 0 ("Static torso") to Bit 1 ("Next").Source
  • 05

    The search listing sells "the robot you hire for warehouse work"; the Y Combinator profile says Parametric builds "Physical businesses that run themselves", starting with a laundromat; the blog talks about robots in a laundromat folding T-shirts and towels. A buyer or investor who reads two of these sees two different companies.

    Medium impactLow effort First result in 1 week Founder/CEO

    First action

    Write the paragraph and agree it between the founders.

    How to measure it

    Places using the agreed story

    Now
    –
    Target
    All five (homepage, team, contact, blog, YC profile)
    By
    30 days

    Steps

    1. 1

      Agree the line

      Which market leads, and how laundry fits.

      Founder/CEO 1 hour

    2. 2

      Write the paragraph

      Four sentences, plain words.

      Founder/CEO 1 hour

    3. 3

      Update the homepage

      Under the H1.

      Founder/CTO 1 hour

    4. 4

      Update the YC profile

      Tagline and description.

      Founder/CEO 30 minutes

    5. 5

      Add to blog posts

      A closing note on each.

      Founder/CTO 1 hour

    6. 6

      Check search listing

      Title and description match.

      Founder/CTO 30 minutes

    First result
    1 week
    Cost
    $0
    Owner
    Founder/CEO

    Smallest test

    Update the YC profile and homepage subheading today.

    Why it works

    Write one paragraph that connects them and use it everywhere: Bit learned to fold and handle soft goods in Parametric's own robot-run laundry, and now does the same repetitive packing and folding work for warehouses. Put it on the homepage under "Meet Bit.", on the team page and in the Y Combinator profile, and end each blog post with it.

    A consistent story builds trust, and running its own laundry is proof most robot vendors cannot show.

    Risks and how to handle them

    • Narrowing too early

      Keep the story about the work (packing and folding), not one industry.

    Evidence

    • Your websiteSearch listing: title "Parametric | Bit, the robot you hire for warehouse work"; description "Parametric builds Bit, a two-armed mobile robot that businesses hire for repetitive warehouse work such as box assembly, pack out, taping and folding." The homepage headline changed from "AUTOMATE REPETITION TODAY" to "Meet Bit." over the past year.Source
    • Y CombinatorY Combinator company page (Fall 2025, team size 7, San Francisco): "Physical businesses that run themselves", starting with a laundromat designed for robotic operation. The launch post "Parametric: RL for Robotics" says "Existing robots learn generic averages from massive datasets. Our robots learn from customer feedback.", that the company is live with customers, and that its models can learn new on-site behaviors from less than an hour of data.Source
    • Your websiteBlog: three research posts ("Calibration Precedes Evaluation", "Evaluation Precedes Capabilities" and "Monosemantic Features in Vision-Language-Action Models") written for researchers. "Evaluation Precedes Capabilities" says "Today, robots in a laundromat can fold a t-shirt, medium towel, or shorts" and that "not a single robot today can autonomously fold an entire basket of your laundry", with a roadmap to a robot that takes in dirty laundry and produces clean folded laundry. The posts offer an email signup but no buyer-facing next step beyond the "Hire Bit" link.Source
  • 06

    The homepage says "Bit is getting better every single day", but its proof is a July 2026 figure (75%, up from 0% in February). By November that reads as a stale snapshot, or worse, as progress that stopped.

    Medium impactLow effort First result in 2 weeks Founder/CTO

    First action

    Pull August to October autonomy figures from Bastion.

    How to measure it

    Monthly autonomy updates published

    Now
    –
    Target
    3 in a row
    By
    90 days

    Steps

    1. 1

      Pull the data

      Monthly autonomy, hours, episodes since February.

      Founder/CTO half a day

    2. 2

      Define the measure

      One sentence on how autonomy is counted.

      Founder/CTO 1 hour

    3. 3

      Build the chart

      A simple monthly line on the homepage.

      Founder/CTO 1 day

    4. 4

      Add the note

      What changed this month.

      Founder/CEO monthly

    5. 5

      Share it

      Post the update and email the blog list.

      Founder/CEO monthly

    6. 6

      Keep it honest

      Show dips with a reason, such as a new task.

      Founder/CTO monthly

    First result
    2 weeks
    Cost
    $0
    Owner
    Founder/CTO

    Smallest test

    Change the July figure to the latest month this week.

    Why it works

    Replace the single figure with a small monthly chart of autonomous share of robot time, plus hours and episodes, updated on the first of each month from Bastion. Add a one-line note on what changed (a new task, a new hardware version).

    A rising line is the most convincing proof of a learning robot, and a monthly update gives the team a reason to post on social and email the blog list.

    Risks and how to handle them

    • Autonomy dips when new tasks start

      Show autonomy per task, or mark when a new task was added.

    Evidence

    • Your websiteHomepage: H1 "Meet Bit." and "Commercial autonomy made simple."; under "Bit learns repetitive work." it lists "Box assembly", "Pack out", "Taping" and "Folding"; "75%" of robot time "was autonomous in July 2026. In February it was 0%."; "130,000" episodes of manipulation data and "3,500+" hours of real-world robot operation. The page gives no price, pilot process, throughput, site requirements, customer logos or testimonials, and "Learn more" links back to the same page.Source
    • Your websiteSitemap (9 URLs, checked November 9, 2026): the homepage, the blog index and three posts, team, contact, privacy and terms. There are no pages for individual tasks, industries, pricing, Bastion, customer stories, safety or FAQs; the newest post is dated July 13, 2026.Source
  • 07

    Real visitors already pass Core Web Vitals (LCP 1.2 s at the 75th percentile), so this is not urgent. But in Google's phone test the hero image of Bit appears at 3.3 s, held back by about 0.75 s of render-blocking requests, and half the 251 KB page is Google Fonts; field LCP has also slowed by about 30%.

    Low impactLow effort First result in 1 week Founder/CTO

    First action

    Preload the hero image and self-host the fonts.

    How to measure it

    Phone LCP in PageSpeed

    Now
    –
    Target
    Under 2.5 s
    By
    2 weeks

    Steps

    1. 1

      Preload the hero

      High fetch priority on the Bit image.

      Founder/CTO 1 hour

    2. 2

      Resize the image

      Phone sizes in a modern format.

      Founder/CTO 1 hour

    3. 3

      Self-host fonts

      Fewer weights, swap display.

      Founder/CTO 2 hours

    4. 4

      Remove blocking requests

      Inline the small CSS.

      Founder/CTO 1 hour

    5. 5

      Retest

      PageSpeed phone and desktop.

      Founder/CTO 30 minutes

    6. 6

      Watch field data

      Check the 30% LCP slowdown reverses.

      Founder/CTO monthly

    First result
    1 week
    Cost
    $0
    Owner
    Founder/CTO

    Smallest test

    Preload the hero image and rerun PageSpeed.

    Why it works

    Self-host the fonts (or cut to one weight), preload the hero image with a high fetch priority, and serve it in a modern format at phone sizes. Recheck in PageSpeed.

    The robot is the product; buyers on phones from a trade show or a LinkedIn post should see Bit first, not a blank hero.

    Risks and how to handle them

    • Font change alters the look

      Keep the same family, just served locally.

    Evidence

    • PageSpeedGoogle PageSpeed (November 9, 2026): mobile score 78, desktop 97. In the phone lab test the largest element is the hero image of Bit, shown at 3.3 s (target 2.5 s), with about 0.75 s of render-blocking requests and 124 KB of Google Fonts in a 251 KB page. Real visitors pass Core Web Vitals (75th percentile LCP 1.2 s, INP 95 ms, CLS 0), although field LCP is about 30% slower than in the previous period.Source

This quarter 12

The core plays for the next three months.

  • 08

    "We'll hand you tools to run your fleet" is a strong promise, but the site gives Bastion one sentence ("the app we use to run our own robots") and a screenshot of example data; "Bit works in fleets" has no text at all. Bastion has no page in the sitemap.

    Medium impactMedium effort First result in 1 month Founder/CTO

    First action

    Record a two-minute walkthrough of Bastion on real (anonymised) data.

    How to measure it

    Requests touching the Bastion page

    Now
    –
    Target
    30%
    By
    90 days

    Steps

    1. 1

      Pick the screens

      Fleet view, help request, daily report.

      Founder/CTO 1 day

    2. 2

      Anonymise data

      Remove site and customer names.

      Founder/CTO half a day

    3. 3

      Record

      Two-minute walkthrough.

      Founder/CTO half a day

    4. 4

      Write the page

      Who uses it and for what.

      Founder/CEO 1 day

    5. 5

      Fill the fleet section

      Text under "Bit works in fleets" linking to the page.

      Founder/CEO 1 hour

    6. 6

      Publish and link

      Navigation already lists Bastion; point it to the page.

      Founder/CTO 1 hour

    First result
    1 month
    Cost
    Team time
    Owner
    Founder/CTO

    Smallest test

    Add three lines under "Bit works in fleets" this week.

    Why it works

    Publish /bastion: what an operations manager sees (robots, sites, tasks, shift times), how help requests work (Ask, Take over), what reports it gives (hours, autonomy, units), who on the customer's team uses it, and a short screen recording.

    Operations managers buy control and visibility as much as robots. A fleet app that Parametric runs its own business on is proof of maturity small rivals cannot show.

    Risks and how to handle them

    • Showing unfinished screens

      Show only views customers already use.

    Evidence

    • Your websiteHomepage fleet sections: "Bit works in fleets." has no body text; "We'll hand you tools to run your fleet." says only "Bastion is the app we use to run our own robots." above an example dashboard ("7 robots at 2 sites"; tasks "Fold towels", "Fold shipping box" and "Fold T-shirts"; "Bit asks for help": "The box flap did not fold after two tries." with "Ask" and "Take over" buttons). "Grow your business alongside Bit." shows only the hardware roadmap from Bit 0 ("Static torso") to Bit 1 ("Next").Source
    • Your websiteSitemap (9 URLs, checked November 9, 2026): the homepage, the blog index and three posts, team, contact, privacy and terms. There are no pages for individual tasks, industries, pricing, Bastion, customer stories, safety or FAQs; the newest post is dated July 13, 2026.Source
  • 09

    Parametric says it is live with customers, and the homepage shows robots at two sites, but the site has no case studies, testimonials or logos. Formic shows uptime and Dexterity shows FedEx; Parametric shows nothing a buyer can check.

    High impactMedium effort First result in 6 weeks Founder/CEO

    First action

    Ask the longest-running customer for permission.

    How to measure it

    Requests after reading the story

    Now
    –
    Target
    25% of requests
    By
    1 quarter

    Steps

    1. 1

      Ask permission

      Named, or anonymous by industry.

      Founder/CEO 1 week

    2. 2

      Pull data

      Hours, units, autonomy over time.

      Founder/CTO 1 day

    3. 3

      Interview

      The site manager, 30 minutes.

      Founder/CEO 1 day

    4. 4

      Film

      Bit working on site.

      Robotics Technician 1 day

    5. 5

      Write and approve

      Customer signs off.

      Founder/CEO 2 weeks

    6. 6

      Publish and share

      Homepage, Hire Bit page, social.

      Founder/CEO 1 day

    First result
    6 weeks
    Cost
    Team time
    Owner
    Founder/CEO

    Smallest test

    Publish the Parametric laundry itself as story zero.

    Why it works

    Ask one customer for a named or anonymous story: the task, the problem before Bit, setup time, hours worked, units handled, autonomy over time and what staff now do instead. Use Bastion data for every number, and add a 60-second video.

    One real deployment with numbers is worth more than any spec. It also gives sales and social a story to repeat.

    Risks and how to handle them

    • Customer will not be named

      Publish by industry and city with real numbers.

    Evidence

    • Y CombinatorY Combinator company page (Fall 2025, team size 7, San Francisco): "Physical businesses that run themselves", starting with a laundromat designed for robotic operation. The launch post "Parametric: RL for Robotics" says "Existing robots learn generic averages from massive datasets. Our robots learn from customer feedback.", that the company is live with customers, and that its models can learn new on-site behaviors from less than an hour of data.Source
    • Your websiteSite links: the footer links only to Bit, Blog, Team, Hire Bit, the hello@ email, privacy and terms. No LinkedIn, X, YouTube or Instagram profile is linked from the site, and there are no case studies, testimonials or live chat.Source
    • Competitor researchRivals' own homepages: Formic sells "Full Service Automation for mid-market manufacturers. No CapEx. No Risk. No Problems." at one fixed monthly price, and shows 99.30% uptime over the last 12 months, a 12-week average lead time and an on-site evaluation offer; Dexterity shows a FedEx endorsement and "Start a deployment"; Verne Robotics ("Physical AI for building a better world") offers "Book a demo" and fine-tuning its model to your tasks.Source
  • 10

    Bit's four tasks (box assembly, pack out, taping, folding) are the same jobs warehouses and fulfilment centres advertise for again and again. A company posting "pack-out associate" or "kitting" roles every month is telling the world it has the exact work Bit does and cannot keep it staffed.

    High impactMedium effort First result in 1 month Founder/CEO

    First action

    Build a list of 30 local companies hiring for pack-out or kitting this month.

    How to measure it

    Site visits booked

    Now
    –
    Target
    6 a quarter
    By
    1 quarter

    Steps

    1. 1

      Set the searches

      Pack out, kitting, box assembly, folding; within driving distance of the shop.

      Founder/CEO half a day

    2. 2

      Score companies

      Repeat posts, shifts, headcount.

      Founder/CEO weekly

    3. 3

      Find the right person

      Operations or site manager.

      Founder/CEO weekly

    4. 4

      Send a short note

      Their task, a clip of Bit, the hire terms.

      Founder/CEO weekly

    5. 5

      Offer a sample test

      Bring-your-box trial from idea 11.

      Founder/CTO as needed

    6. 6

      Track

      Reply and visit rate by task.

      Founder/CEO monthly

    First result
    1 month
    Cost
    Founder time
    Owner
    Founder/CEO

    Smallest test

    Email ten companies with open pack-out roles this week.

    Why it works

    Each week, search public job boards near San Francisco and any second launch city for pack-out, kitting, box assembly and folding roles. List companies with repeat or multi-shift posts, and send the operations lead a short note: the task, a clip of Bit doing it, and the hire terms. Track reply and site-visit rates by task.

    Outbound works when the timing is right; a job post is a visible, current pain. It also keeps the first deployments near the shop, where the technician can reach them.

    Risks and how to handle them

    • Messaging reads as replacing workers

      Frame Bit as covering hard-to-fill shifts and moving staff to better work.

    Builds on

    Evidence

    • Your websiteHomepage: H1 "Meet Bit." and "Commercial autonomy made simple."; under "Bit learns repetitive work." it lists "Box assembly", "Pack out", "Taping" and "Folding"; "75%" of robot time "was autonomous in July 2026. In February it was 0%."; "130,000" episodes of manipulation data and "3,500+" hours of real-world robot operation. The page gives no price, pilot process, throughput, site requirements, customer logos or testimonials, and "Learn more" links back to the same page.Source
    • HiringRobotics Technician (San Francisco, $60K-$90K, on Y Combinator's job board): the company is moving from prototype to production and redesigns processes around robots rather than retrofitting human workflows. The technician assembles robots from drawings and BOMs, runs calibration and functional tests before robots leave the shop, diagnoses and repairs failures in the shop and at deployment sites, does preventive maintenance, keeps inspection checklists and hardware revision records, and travels to deployment sites.Source
  • 11

    Parametric's launch post says its models can learn new on-site behaviors from less than an hour of data. That is the most persuasive claim it has, and buyers cannot test it anywhere on the site.

    High impactMedium effort First result in 1 month Founder/CTO

    First action

    Write the trial rules: what to ship, how much, and where.

    How to measure it

    Trial-to-deployment rate

    Now
    –
    Target
    1 in 4
    By
    2 quarters

    Steps

    1. 1

      Set the rules

      What to ship, batch size, turnaround.

      Founder/CTO 1 day

    2. 2

      Prepare the shop

      A bench and a robot for trials.

      Robotics Technician 1 week

    3. 3

      Template the readout

      Success rate, cycle time, gaps.

      Founder/CTO 1 day

    4. 4

      Add to the site

      Offer on the Hire Bit and task pages.

      Founder/CEO half a day

    5. 5

      Run trials

      Record, train, film, send.

      Founder/CTO 1 week each

    6. 6

      Follow up

      Call with the video and hire terms.

      Founder/CEO per trial

    First result
    1 month
    Cost
    Shop time and shipping
    Owner
    Founder/CTO

    Smallest test

    Offer the trial to the next three scoped requests.

    Why it works

    Offer a free trial on the Hire Bit page: the buyer ships a small batch of their own boxes, tape or items, Parametric collects under an hour of data in its shop, and sends back a video of Bit doing that task plus a short readout (success rate, cycle time, what needs work). Cap it at a few trials a month.

    Seeing a robot do your own task removes the biggest doubt in robot buying, and the sample itself becomes training data and a qualified lead.

    Risks and how to handle them

    • Trials eat engineering time

      Cap at a set number a month and accept only scoped requests.

    Builds on

    Evidence

    • Y CombinatorY Combinator company page (Fall 2025, team size 7, San Francisco): "Physical businesses that run themselves", starting with a laundromat designed for robotic operation. The launch post "Parametric: RL for Robotics" says "Existing robots learn generic averages from massive datasets. Our robots learn from customer feedback.", that the company is live with customers, and that its models can learn new on-site behaviors from less than an hour of data.Source
    • Your websiteHomepage: H1 "Meet Bit." and "Commercial autonomy made simple."; under "Bit learns repetitive work." it lists "Box assembly", "Pack out", "Taping" and "Folding"; "75%" of robot time "was autonomous in July 2026. In February it was 0%."; "130,000" episodes of manipulation data and "3,500+" hours of real-world robot operation. The page gives no price, pilot process, throughput, site requirements, customer logos or testimonials, and "Learn more" links back to the same page.Source
  • 12

    Bastion already tracks robots, sites, tasks, shift times and help requests for Parametric's own operations. Customers who hire Bit need the same numbers to justify keeping it and adding more robots.

    High impactMedium effort First result in 1 month Founder/CTO

    First action

    Mock up the report from one site's data.

    How to measure it

    Average robots per customer site

    Now
    –
    Target
    +1 robot per site
    By
    2 quarters

    Steps

    1. 1

      Pick the numbers

      Hours, units, autonomy, help requests.

      Founder/CTO 1 day

    2. 2

      Mock up

      One page, one site.

      Founder/CTO 1 day

    3. 3

      Automate

      Weekly email from Bastion.

      Founder/CTO 1 week

    4. 4

      Send

      To the site manager and their manager.

      Founder/CEO weekly

    5. 5

      Monthly review

      Propose the next task or robot.

      Founder/CEO monthly

    6. 6

      Track expansion

      Robots and tasks per customer.

      Founder/CEO quarterly

    First result
    1 month
    Cost
    Engineering time
    Owner
    Founder/CTO

    Smallest test

    Send one hand-made report to one customer this month.

    Why it works

    Generate a one-page weekly email per customer site: hours worked, units done, autonomy share, help requests and how fast they were answered, and the trend since start. Add a monthly review call where the founders propose the next task or robot.

    Robot hires are renewed on visible value. A weekly number makes that value obvious to the site manager and to their boss, and sets up the expansion conversation.

    Risks and how to handle them

    • Bad weeks are visible

      Explain dips and the fix in the same email.

    Builds on

    Evidence

    • Your websiteHomepage fleet sections: "Bit works in fleets." has no body text; "We'll hand you tools to run your fleet." says only "Bastion is the app we use to run our own robots." above an example dashboard ("7 robots at 2 sites"; tasks "Fold towels", "Fold shipping box" and "Fold T-shirts"; "Bit asks for help": "The box flap did not fold after two tries." with "Ask" and "Take over" buttons). "Grow your business alongside Bit." shows only the hardware roadmap from Bit 0 ("Static torso") to Bit 1 ("Next").Source
    • Y CombinatorY Combinator company page (Fall 2025, team size 7, San Francisco): "Physical businesses that run themselves", starting with a laundromat designed for robotic operation. The launch post "Parametric: RL for Robotics" says "Existing robots learn generic averages from massive datasets. Our robots learn from customer feedback.", that the company is live with customers, and that its models can learn new on-site behaviors from less than an hour of data.Source
  • 13

    Parametric is hiring a Robotics Technician to test robots before they leave the shop, repair failures at deployment sites, do preventive maintenance and keep inspection checklists. That is a service operation. Formic already sells on 99.30% uptime over 12 months; Parametric publishes nothing on uptime or response time.

    Medium impactMedium effort First result in 1 quarter Robotics Technician

    First action

    Start logging every robot stoppage and fix time from today.

    How to measure it

    Uptime across customer sites

    Now
    –
    Target
    Published quarterly
    By
    2 quarters

    Steps

    1. 1

      Log stoppages

      Cause, time to remote fix, time to on-site fix.

      Robotics Technician ongoing

    2. 2

      Standardise checks

      Pre-ship and preventive checklists.

      Robotics Technician 1 month

    3. 3

      Build a field kit

      Spares for the top failure causes.

      Robotics Technician 1 month

    4. 4

      Measure a quarter

      Uptime and response times.

      Founder/CTO 1 quarter

    5. 5

      Set the commitment

      Numbers the team can beat.

      Founder/CEO 1 week

    6. 6

      Publish

      On the Hire Bit page and in contracts.

      Founder/CEO 1 day

    First result
    1 quarter
    Cost
    Technician salary (already budgeted) and spares
    Owner
    Robotics Technician

    Smallest test

    Publish only the remote response time first.

    Why it works

    Once the technician is in place, measure robot uptime and time-to-fix at each site for a quarter, then publish a simple commitment: target uptime, remote response time, on-site response time in the Bay Area, and preventive maintenance cadence. Put it on the Hire Bit page.

    Buyers of hired robots are really buying reliability. A published commitment backed by a real field process answers the question competitors already answer.

    Risks and how to handle them

    • Missing a public target

      Commit below measured performance and report honestly.

    Evidence

    • HiringRobotics Technician (San Francisco, $60K-$90K, on Y Combinator's job board): the company is moving from prototype to production and redesigns processes around robots rather than retrofitting human workflows. The technician assembles robots from drawings and BOMs, runs calibration and functional tests before robots leave the shop, diagnoses and repairs failures in the shop and at deployment sites, does preventive maintenance, keeps inspection checklists and hardware revision records, and travels to deployment sites.Source
    • Competitor researchRivals' own homepages: Formic sells "Full Service Automation for mid-market manufacturers. No CapEx. No Risk. No Problems." at one fixed monthly price, and shows 99.30% uptime over the last 12 months, a 12-week average lead time and an on-site evaluation offer; Dexterity shows a FedEx endorsement and "Start a deployment"; Verne Robotics ("Physical AI for building a better world") offers "Book a demo" and fine-tuning its model to your tasks.Source
  • 14

    The blog is written for researchers (calibration staircases, transcoders, monosemantic features), which helps hiring and credibility, but the only buyer-facing next step is the header link. Each post actually explains a buyer benefit: calibrated evaluation means faster, cheaper onboarding of a new task.

    Medium impactLow effort First result in 1 month Founder/CTO

    First action

    Write the box for "Calibration Precedes Evaluation".

    How to measure it

    Blog-sourced requests

    Now
    –
    Target
    2 a month
    By
    1 quarter

    Steps

    1. 1

      Write three boxes

      One per existing post.

      Founder/CTO half a day

    2. 2

      Link to tasks

      Task page or trial.

      Founder/CTO 1 hour

    3. 3

      Template it

      For every new post.

      Founder/CTO 1 hour

    4. 4

      Email the list

      Buyer version of each post.

      Founder/CEO per post

    5. 5

      Post more often

      One short post a month.

      Founder/CTO monthly

    6. 6

      Measure

      Blog to request path.

      Founder/CEO monthly

    First result
    1 month
    Cost
    $0
    Owner
    Founder/CTO

    Smallest test

    Add the box to the newest post today.

    Why it works

    Add a three-line box to each existing and future post: what the research changes for a customer (faster new tasks, fewer help requests, safer handling) and a link to the relevant task page or the bring-your-box trial. Send the email list a short buyer version of each post.

    Technical posts attract engineers and operations leaders at automation-minded companies; a plain-language takeaway turns some of those readers into requests.

    Risks and how to handle them

    • Researchers dislike sales copy

      Keep it short and clearly separate from the post.

    Builds on

    Evidence

    • Your websiteBlog: three research posts ("Calibration Precedes Evaluation", "Evaluation Precedes Capabilities" and "Monosemantic Features in Vision-Language-Action Models") written for researchers. "Evaluation Precedes Capabilities" says "Today, robots in a laundromat can fold a t-shirt, medium towel, or shorts" and that "not a single robot today can autonomously fold an entire basket of your laundry", with a roadmap to a robot that takes in dirty laundry and produces clean folded laundry. The posts offer an email signup but no buyer-facing next step beyond the "Hire Bit" link.Source
    • Your websiteSitemap (9 URLs, checked November 9, 2026): the homepage, the blog index and three posts, team, contact, privacy and terms. There are no pages for individual tasks, industries, pricing, Bastion, customer stories, safety or FAQs; the newest post is dated July 13, 2026.Source
  • 15

    Parametric records thousands of hours of real robot work, yet the site links to no LinkedIn, X, YouTube or Instagram profile. Robot clips are among the most shared content in operations and automation circles.

    Medium impactLow effort First result in 1 month Founder/CEO

    First action

    Pick four clips for the next four weeks.

    How to measure it

    Social-sourced requests

    Now
    –
    Target
    3 a month
    By
    1 quarter

    Steps

    1. 1

      Claim profiles

      LinkedIn page, X, YouTube.

      Founder/CEO 1 hour

    2. 2

      Link from the site

      Footer icons.

      Founder/CTO 30 minutes

    3. 3

      Build a clip queue

      Four weeks ahead.

      Founder/CTO monthly

    4. 4

      Post weekly

      One clip, one line, one link.

      Founder/CEO weekly

    5. 5

      Founders amplify

      Share from personal accounts.

      Founder/CEO weekly

    6. 6

      Track

      Requests with social as the source.

      Founder/CEO monthly

    First result
    1 month
    Cost
    $0
    Owner
    Founder/CEO

    Smallest test

    Post one clip this week and measure clicks.

    Why it works

    Each week, post one 15-30 second clip from real work (a box folded, a help request handled, a new task learned) on LinkedIn and X, with a one-line caption and a link to the task page. Add the profile links to the footer, and collect clips into a YouTube playlist per task.

    Short real footage builds trust faster than any copy, reaches operations leaders where they read, and costs nothing because the footage already exists.

    Risks and how to handle them

    • Customer sites appear on camera

      Film in the shop or the company's own laundry, or get written approval.

    Evidence

    • Your websiteSite links: the footer links only to Bit, Blog, Team, Hire Bit, the hello@ email, privacy and terms. No LinkedIn, X, YouTube or Instagram profile is linked from the site, and there are no case studies, testimonials or live chat.Source
    • Your websiteHomepage: H1 "Meet Bit." and "Commercial autonomy made simple."; under "Bit learns repetitive work." it lists "Box assembly", "Pack out", "Taping" and "Folding"; "75%" of robot time "was autonomous in July 2026. In February it was 0%."; "130,000" episodes of manipulation data and "3,500+" hours of real-world robot operation. The page gives no price, pilot process, throughput, site requirements, customer logos or testimonials, and "Learn more" links back to the same page.Source
  • 16

    Box assembly and taping are two of Bit's four tasks. Corrugated box and packaging distributors sell flat boxes and tape to exactly the businesses that assemble them by hand, and their sales reps hear about labor problems on every visit.

    Medium impactMedium effort First result in 1 quarter Founder/CEO

    First action

    List five local packaging distributors and the box styles Bit handles today.

    How to measure it

    Partner introductions

    Now
    –
    Target
    5 a quarter
    By
    2 quarters

    Steps

    1. 1

      List box styles

      Which standard styles Bit folds today.

      Founder/CTO 1 day

    2. 2

      List distributors

      Five in the Bay Area.

      Founder/CEO 1 day

    3. 3

      Write the deal

      Referral fee per hired robot.

      Founder/CEO 1 day

    4. 4

      Make the leaflet

      One page and a clip.

      Founder/CEO 2 days

    5. 5

      Meet reps

      Demo at the shop.

      Founder/CEO 1 month

    6. 6

      Track

      Introductions and hires.

      Founder/CEO monthly

    First result
    1 quarter
    Cost
    Referral fees on closed hires
    Owner
    Founder/CEO

    Smallest test

    One coffee with one distributor's sales manager.

    Why it works

    Approach two or three Bay Area packaging distributors with a simple referral deal: they introduce customers who hand-fold large volumes of boxes, Parametric pays a referral fee per hired robot, and the supplier keeps selling the boxes Bit folds. Give reps a one-page leaflet and a clip of Bit with standard box styles.

    Suppliers gain stickier customers and a fee; Parametric gets warm introductions to the right buyers through someone they already trust.

    Risks and how to handle them

    • Low rep engagement

      Start with one distributor and a rep who already asked about automation.

    Builds on

    Evidence

    • Your websiteHomepage: H1 "Meet Bit." and "Commercial autonomy made simple."; under "Bit learns repetitive work." it lists "Box assembly", "Pack out", "Taping" and "Folding"; "75%" of robot time "was autonomous in July 2026. In February it was 0%."; "130,000" episodes of manipulation data and "3,500+" hours of real-world robot operation. The page gives no price, pilot process, throughput, site requirements, customer logos or testimonials, and "Learn more" links back to the same page.Source
    • Your websiteHomepage fleet sections: "Bit works in fleets." has no body text; "We'll hand you tools to run your fleet." says only "Bastion is the app we use to run our own robots." above an example dashboard ("7 robots at 2 sites"; tasks "Fold towels", "Fold shipping box" and "Fold T-shirts"; "Bit asks for help": "The box flap did not fold after two tries." with "Ask" and "Take over" buttons). "Grow your business alongside Bit." shows only the hardware roadmap from Bit 0 ("Static torso") to Bit 1 ("Next").Source
  • 17

    Operations, safety and IT teams each have to approve a robot on the floor, and the site says nothing about what Bit needs: floor space, power, network, lighting, or how it works around people. The technician role already covers deployment-site work, so this knowledge exists in the team.

    Medium impactLow effort First result in 1 month Robotics Technician

    First action

    Have the technician write the checklist used at the last deployment.

    How to measure it

    Time to deploy

    Now
    –
    Target
    -25%
    By
    2 quarters

    Steps

    1. 1

      Write the checklist

      From the last deployment.

      Robotics Technician 1 day

    2. 2

      Add specs

      Footprint, power, network.

      Founder/CTO half a day

    3. 3

      Describe working near people

      Only practices in use today.

      Founder/CTO half a day

    4. 4

      Review

      Check every claim is accurate.

      Founder/CEO 1 day

    5. 5

      Publish

      Page and printable checklist.

      Founder/CTO half a day

    6. 6

      Send before visits

      Every site visit gets the checklist.

      Founder/CEO ongoing

    First result
    1 month
    Cost
    $0
    Owner
    Robotics Technician

    Smallest test

    Send the checklist to the next prospect before the visit.

    Why it works

    Publish /bit/site-requirements: footprint and reach, power, network, workstation height, materials supply, how Bit works near staff (stops, speeds, zones) and what the customer's team needs to do in week one. Offer a printable checklist. State only safety practices the team actually follows; do not claim certifications it does not hold.

    Answering the safety and IT questions up front shortens approvals, and it signals that Parametric has done real deployments.

    Risks and how to handle them

    • Overstating safety

      Describe practices, not certifications, unless certified.

    Evidence

    • HiringRobotics Technician (San Francisco, $60K-$90K, on Y Combinator's job board): the company is moving from prototype to production and redesigns processes around robots rather than retrofitting human workflows. The technician assembles robots from drawings and BOMs, runs calibration and functional tests before robots leave the shop, diagnoses and repairs failures in the shop and at deployment sites, does preventive maintenance, keeps inspection checklists and hardware revision records, and travels to deployment sites.Source
    • Your websiteSitemap (9 URLs, checked November 9, 2026): the homepage, the blog index and three posts, team, contact, privacy and terms. There are no pages for individual tasks, industries, pricing, Bastion, customer stories, safety or FAQs; the newest post is dated July 13, 2026.Source
  • 18

    Parametric is the rare robot company that runs its own robot-operated business, a laundry designed for robotic operation. That is a live, working showroom rivals cannot match, and nothing on the site invites buyers to see it.

    Medium impactMedium effort First result in 1 month Founder/CEO

    First action

    Pick a date and a visitor route that does not disrupt operations.

    How to measure it

    Visit-to-trial rate

    Now
    –
    Target
    30%
    By
    1 quarter

    Steps

    1. 1

      Plan the route

      Safe viewing, no disruption.

      Robotics Technician 1 day

    2. 2

      Set the agenda

      Watch, Bastion, questions, trial booking.

      Founder/CEO half a day

    3. 3

      Invite

      From requests, outbound and partners.

      Founder/CEO 2 weeks

    4. 4

      Host

      One hour, monthly.

      Founder/CEO monthly

    5. 5

      Follow up

      Within 48 hours with a trial offer.

      Founder/CEO per visit

    6. 6

      Film it

      Clips for social.

      Founder/CTO per visit

    First result
    1 month
    Cost
    Coffee and team time
    Owner
    Founder/CEO

    Smallest test

    Invite five prospects to one visit next month.

    Why it works

    Host a one-hour visit once a month for eight to ten operations and fulfilment leaders: watch Bit fold and handle goods, see Bastion live, ask the team questions, and book a bring-your-box trial on the spot. Invite from the scoping form, outbound lists and partners.

    Seeing robots do real paid work, with real help requests, convinces faster than any video, and a small group visit costs little.

    Risks and how to handle them

    • A bad day for the robots

      Show help requests openly; it is part of the story.

    Builds on

    Evidence

    • Y CombinatorY Combinator company page (Fall 2025, team size 7, San Francisco): "Physical businesses that run themselves", starting with a laundromat designed for robotic operation. The launch post "Parametric: RL for Robotics" says "Existing robots learn generic averages from massive datasets. Our robots learn from customer feedback.", that the company is live with customers, and that its models can learn new on-site behaviors from less than an hour of data.Source
    • Your websiteBlog: three research posts ("Calibration Precedes Evaluation", "Evaluation Precedes Capabilities" and "Monosemantic Features in Vision-Language-Action Models") written for researchers. "Evaluation Precedes Capabilities" says "Today, robots in a laundromat can fold a t-shirt, medium towel, or shorts" and that "not a single robot today can autonomously fold an entire basket of your laundry", with a roadmap to a robot that takes in dirty laundry and produces clean folded laundry. The posts offer an email signup but no buyer-facing next step beyond the "Hire Bit" link.Source
  • 19

    Operations managers talk to peers at other sites, 3PLs and suppliers, and the homepage's "Grow your business alongside Bit" section has no customer voice at all. With few customers, each referral matters.

    Medium impactLow effort First result in 1 quarter Founder/CEO

    First action

    Write the ask and the reward in two lines.

    How to measure it

    Referred hires

    Now
    –
    Target
    2 a half-year
    By
    2 quarters

    Steps

    1. 1

      Set the reward

      One free robot-month per referred hire.

      Founder/CEO 1 hour

    2. 2

      Pick the moment

      After a strong weekly report.

      Founder/CEO ongoing

    3. 3

      Ask personally

      A founder call or email.

      Founder/CEO per customer

    4. 4

      Fast-track referrals

      Straight to a trial.

      Founder/CTO per referral

    5. 5

      Thank publicly

      With permission, in a post.

      Founder/CEO per hire

    6. 6

      Track

      Referrals and hires.

      Founder/CEO quarterly

    First result
    1 quarter
    Cost
    One robot-month per referred hire
    Owner
    Founder/CEO

    Smallest test

    Ask the happiest customer this month.

    Why it works

    When a customer's weekly report shows a good month, ask the site manager for one introduction. If it becomes a hire, the referrer gets a free month on one robot. Keep it personal, from the founders.

    Peer introductions close faster and cost less than outbound, and the reward is cheap for Parametric and valuable for the customer.

    Risks and how to handle them

    • Asking too early

      Only after a strong report and a good review call.

    Builds on

    Evidence

    • Your websiteHomepage fleet sections: "Bit works in fleets." has no body text; "We'll hand you tools to run your fleet." says only "Bastion is the app we use to run our own robots." above an example dashboard ("7 robots at 2 sites"; tasks "Fold towels", "Fold shipping box" and "Fold T-shirts"; "Bit asks for help": "The box flap did not fold after two tries." with "Ask" and "Take over" buttons). "Grow your business alongside Bit." shows only the hardware roadmap from Bit 0 ("Static torso") to Bit 1 ("Next").Source
    • Your websiteSite links: the footer links only to Bit, Blog, Team, Hire Bit, the hello@ email, privacy and terms. No LinkedIn, X, YouTube or Instagram profile is linked from the site, and there are no case studies, testimonials or live chat.Source

Big bets 3

Larger plays that can change the trajectory.

  • 20

    Buyers think in cost per unit: what each packed order or folded box costs in labor. Bastion already tracks tasks and shift times per robot, so Parametric can count units. Per-unit pricing would compare directly with the buyer's labor cost and move the risk of help requests and downtime onto Parametric, where it is easiest to manage.

    High impactHigh effort First result in 1 quarter Founder/CEO

    First action

    Calculate cost per unit at today's autonomy for the two most common tasks.

    How to measure it

    Per-unit vs per-robot win rate

    Now
    –
    Target
    Higher on per-unit
    By
    2 quarters

    Steps

    1. 1

      Model unit cost

      Robot, help time, maintenance per unit at current autonomy.

      Founder/CEO 1 week

    2. 2

      Make counts auditable

      Bastion unit counts the customer can see.

      Founder/CTO 1 month

    3. 3

      Set the price

      Per unit with a monthly minimum.

      Founder/CEO 1 week

    4. 4

      Pilot

      Two customers.

      Founder/CEO 1 quarter

    5. 5

      Compare

      Margin, win rate, churn.

      Founder/CEO 1 quarter

    6. 6

      Decide

      Default model for 2027.

      Founder/CEO 1 week

    First result
    1 quarter
    Cost
    Engineering time; margin risk during the pilot
    Owner
    Founder/CEO

    Smallest test

    Quote one prospect both ways and see which they choose.

    Why it works

    Pilot a per-unit price with two customers: a fixed price per box assembled or per item folded, with a monthly minimum. Bastion counts units and shows them in the weekly report. Compare margin and win rate against per-robot pricing over two quarters.

    It removes the buyer's main fear (paying for a robot that is waiting for help) and makes the comparison with labor cost a single number. As autonomy rises, Parametric's margin rises with it.

    Risks and how to handle them

    • Low-autonomy tasks lose money

      Offer per-unit only on tasks above an autonomy threshold.

    • Disputes over counts

      Customer-visible counts with video samples.

    Builds on

    Evidence

    • Your websiteHomepage fleet sections: "Bit works in fleets." has no body text; "We'll hand you tools to run your fleet." says only "Bastion is the app we use to run our own robots." above an example dashboard ("7 robots at 2 sites"; tasks "Fold towels", "Fold shipping box" and "Fold T-shirts"; "Bit asks for help": "The box flap did not fold after two tries." with "Ask" and "Take over" buttons). "Grow your business alongside Bit." shows only the hardware roadmap from Bit 0 ("Static torso") to Bit 1 ("Next").Source
    • Competitor researchRivals' own homepages: Formic sells "Full Service Automation for mid-market manufacturers. No CapEx. No Risk. No Problems." at one fixed monthly price, and shows 99.30% uptime over the last 12 months, a 12-week average lead time and an on-site evaluation offer; Dexterity shows a FedEx endorsement and "Start a deployment"; Verne Robotics ("Physical AI for building a better world") offers "Book a demo" and fine-tuning its model to your tasks.Source
    • Your websiteHomepage: H1 "Meet Bit." and "Commercial autonomy made simple."; under "Bit learns repetitive work." it lists "Box assembly", "Pack out", "Taping" and "Folding"; "75%" of robot time "was autonomous in July 2026. In February it was 0%."; "130,000" episodes of manipulation data and "3,500+" hours of real-world robot operation. The page gives no price, pilot process, throughput, site requirements, customer logos or testimonials, and "Learn more" links back to the same page.Source
  • 21

    Parametric's stated mission is "Physical businesses that run themselves", starting with a laundromat designed for robotic operation, and its blog lays out a roadmap from folding single items to dirty laundry in, clean folded laundry out. Wash-and-fold operators everywhere have the same folding-table labor problem.

    High impactHigh effort First result in 2 quarters Founder/CEO

    First action

    Publish the laundry's monthly autonomy and cost per pound internally.

    How to measure it

    Laundries running Bit

    Now
    –
    Target
    2
    By
    3 quarters

    Steps

    1. 1

      Measure the own laundry

      Autonomy, pounds per hour, cost per pound.

      Founder/CTO 1 quarter

    2. 2

      Write the playbook

      Layout, staffing, Bastion setup.

      Robotics Technician 1 month

    3. 3

      Set the offer

      Per-pound fee or savings share.

      Founder/CEO 2 weeks

    4. 4

      Find a partner

      One independent operator nearby.

      Founder/CEO 1 month

    5. 5

      Deploy

      Folding first, then loading.

      Founder/CTO 1 quarter

    6. 6

      Publish results

      With the partner's permission.

      Founder/CEO 1 week

    First result
    2 quarters
    Cost
    Robots and field time for one partner site
    Owner
    Founder/CEO

    Smallest test

    Interview five laundry owners about folding labor and a per-pound offer.

    Why it works

    Once the own laundry runs at high autonomy, write down the layout, the robot count, the Bastion setup and the unit economics, then offer it to a few independent wash-and-fold operators as a hired package: robots for folding, Bastion, training, and a share of savings or a per-pound fee. Start with one partner operator near San Francisco.

    It turns Parametric's own business into a repeatable product with proven numbers, and every partner laundry adds data on the hardest soft-goods tasks.

    Risks and how to handle them

    • Splitting focus between warehouses and laundry

      Treat laundry as the same folding task in a second market, with one partner only until it works.

    Builds on

    Evidence

    • Y CombinatorY Combinator company page (Fall 2025, team size 7, San Francisco): "Physical businesses that run themselves", starting with a laundromat designed for robotic operation. The launch post "Parametric: RL for Robotics" says "Existing robots learn generic averages from massive datasets. Our robots learn from customer feedback.", that the company is live with customers, and that its models can learn new on-site behaviors from less than an hour of data.Source
    • Your websiteBlog: three research posts ("Calibration Precedes Evaluation", "Evaluation Precedes Capabilities" and "Monosemantic Features in Vision-Language-Action Models") written for researchers. "Evaluation Precedes Capabilities" says "Today, robots in a laundromat can fold a t-shirt, medium towel, or shorts" and that "not a single robot today can autonomously fold an entire basket of your laundry", with a roadmap to a robot that takes in dirty laundry and produces clean folded laundry. The posts offer an email signup but no buyer-facing next step beyond the "Hire Bit" link.Source
  • 22

    The team's two newest posts argue that evaluation precedes capabilities and that calibrated evaluation lets you "learn more, spend less, and iterate faster". Buyers comparing robot vendors have no standard way to judge claims about box folding or pack out. The team that defines the test shapes the market.

    High impactHigh effort First result in 2 quarters Founder/CTO

    First action

    Write down the internal evaluation used for box folding today.

    How to measure it

    External benchmark users

    Now
    –
    Target
    10
    By
    3 quarters

    Steps

    1. 1

      Pick tasks

      Box folding, taping, garment folding.

      Founder/CTO 1 week

    2. 2

      Standardise items

      Box styles and garments anyone can buy.

      Robotics Technician 2 weeks

    3. 3

      Write scoring

      From the calibration method.

      Founder/CTO 1 month

    4. 4

      Publish Bit's results

      With videos.

      Founder/CTO 1 month

    5. 5

      Write the buyer's guide

      How to evaluate a packing robot.

      Founder/CEO 2 weeks

    6. 6

      Invite others

      Research labs and robot companies.

      Founder/CTO ongoing

    First result
    2 quarters
    Cost
    Research time
    Owner
    Founder/CTO

    Smallest test

    Publish the box-folding scoring method as a blog post.

    Why it works

    Release a small public benchmark: standard box styles, garments and pack-out kits, a scoring method based on the calibration staircase, and Bit's own results updated each quarter. Invite other robot teams and buyers to use it. Turn the results into a buyer's guide to evaluating packing robots.

    It turns research credibility into commercial authority, earns links and press, and gives buyers a reason to start their evaluation on Parametric's site.

    Risks and how to handle them

    • Rivals score higher

      That still makes Parametric the referee; focus on improvement over time.

    Builds on

    Evidence

    • Your websiteBlog: three research posts ("Calibration Precedes Evaluation", "Evaluation Precedes Capabilities" and "Monosemantic Features in Vision-Language-Action Models") written for researchers. "Evaluation Precedes Capabilities" says "Today, robots in a laundromat can fold a t-shirt, medium towel, or shorts" and that "not a single robot today can autonomously fold an entire basket of your laundry", with a roadmap to a robot that takes in dirty laundry and produces clean folded laundry. The posts offer an email signup but no buyer-facing next step beyond the "Hire Bit" link.Source
    • Your websiteTeam page: "a small team in San Francisco, and we're hiring" (applications go through the contact form); backed by Y Combinator, Swell and Mana Ventures; the team comes from Covariant, Meta, Parallel Systems and Berkeley AI Research; Parametric is a public benefit corporation.Source

Wildcards 2

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

  • 23

    The homepage dashboard already shows Bit on "Fold towels", and the blog says robots in a laundromat can fold a medium towel today. Commercial linen laundries fold huge volumes of identical towels every shift, the easiest soft-goods task Bit already does.

    Medium impactMedium effort First result in 1 quarter Founder/CEO

    First action

    List ten commercial linen laundries within an hour of the shop.

    How to measure it

    Trials booked

    Now
    –
    Target
    2
    By
    1 quarter

    Steps

    1. 1

      List laundries

      Ten within an hour of the shop.

      Founder/CEO 1 day

    2. 2

      Measure towel speed

      Towels per hour and autonomy.

      Founder/CTO 1 week

    3. 3

      Send the note

      Clip, terms, sample offer.

      Founder/CEO 1 week

    4. 4

      Run samples

      Fold their towels in the shop.

      Robotics Technician per trial

    5. 5

      Decide

      Build a page if two trials start.

      Founder/CEO 1 week

    6. 6

      Report

      What worked and why.

      Founder/CEO 1 day

    First result
    1 quarter
    Cost
    Founder time
    Owner
    Founder/CEO

    Smallest test

    Email five laundries with a towel-folding clip.

    Why it works

    Run a small test outreach to Bay Area commercial laundries serving hotels, gyms and spas: a clip of Bit folding towels, the hire terms, and an offer to fold a sample batch. If two say yes to a trial, build a towel-folding page.

    It sells the most mature task into a market with constant volume and hard-to-fill folding shifts, and bridges the warehouse and laundry stories.

    Risks and how to handle them

    • Industrial folders already exist for flat linen

      Target mixed loads and smaller laundries where machines do not fit.

    Builds on

    Evidence

    • Your websiteHomepage fleet sections: "Bit works in fleets." has no body text; "We'll hand you tools to run your fleet." says only "Bastion is the app we use to run our own robots." above an example dashboard ("7 robots at 2 sites"; tasks "Fold towels", "Fold shipping box" and "Fold T-shirts"; "Bit asks for help": "The box flap did not fold after two tries." with "Ask" and "Take over" buttons). "Grow your business alongside Bit." shows only the hardware roadmap from Bit 0 ("Static torso") to Bit 1 ("Next").Source
    • Your websiteBlog: three research posts ("Calibration Precedes Evaluation", "Evaluation Precedes Capabilities" and "Monosemantic Features in Vision-Language-Action Models") written for researchers. "Evaluation Precedes Capabilities" says "Today, robots in a laundromat can fold a t-shirt, medium towel, or shorts" and that "not a single robot today can autonomously fold an entire basket of your laundry", with a roadmap to a robot that takes in dirty laundry and produces clean folded laundry. The posts offer an email signup but no buyer-facing next step beyond the "Hire Bit" link.Source
  • 24

    Buyers already "hire" pack-out labor by the shift through staffing agencies, and Parametric's own language is that businesses hire Bit. An agency that cannot fill every night shift could offer Bit for the hardest-to-staff shifts, billed the way the client already buys labor.

    Medium impactMedium effort First result in 2 quarters Founder/CEO

    First action

    Ask two staffing agencies which shifts they fail to fill most often.

    How to measure it

    Agency-sold robot shifts

    Now
    –
    Target
    First client live
    By
    2 quarters

    Steps

    1. 1

      Interview agencies

      Unfilled shifts and tasks.

      Founder/CEO 2 weeks

    2. 2

      Design the offer

      Price per robot shift and agency margin.

      Founder/CEO 1 week

    3. 3

      Pick one client

      A pack-out site with night-shift gaps.

      Founder/CEO 1 month

    4. 4

      Deploy

      Site checklist and Bastion setup.

      Robotics Technician 1 month

    5. 5

      Report

      Shifts covered and units done.

      Founder/CTO weekly

    6. 6

      Decide

      Expand to more clients or stop.

      Founder/CEO 1 week

    First result
    2 quarters
    Cost
    Agency margin on robot shifts
    Owner
    Founder/CEO

    Smallest test

    One conversation with one agency branch manager.

    Why it works

    Pilot with one light-industrial staffing agency in the Bay Area: the agency offers Bit to clients with unfilled pack-out or box-assembly shifts, Parametric supplies and runs the robot through Bastion, and the agency earns a margin on robot shifts. Start with one client site.

    It plugs Bit into an existing buying habit and sales force, and the agency gains supply for shifts it cannot fill.

    Risks and how to handle them

    • Agency sees robots as a threat

      Position Bit for shifts the agency cannot fill today.

    Builds on

    Evidence

    • Your websiteSearch listing: title "Parametric | Bit, the robot you hire for warehouse work"; description "Parametric builds Bit, a two-armed mobile robot that businesses hire for repetitive warehouse work such as box assembly, pack out, taping and folding." The homepage headline changed from "AUTOMATE REPETITION TODAY" to "Meet Bit." over the past year.Source
    • Your websiteHomepage fleet sections: "Bit works in fleets." has no body text; "We'll hand you tools to run your fleet." says only "Bastion is the app we use to run our own robots." above an example dashboard ("7 robots at 2 sites"; tasks "Fold towels", "Fold shipping box" and "Fold T-shirts"; "Bit asks for help": "The box flap did not fold after two tries." with "Ask" and "Take over" buttons). "Grow your business alongside Bit." shows only the hardware roadmap from Bit 0 ("Static torso") to Bit 1 ("Next").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 #1 (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)
ProgrammableDigital

Show it, don't just describe it

These ideas need something to show: a demo, a video, an animation, graphics or a deck. Digital turns screenshots or the site into all of them, in the company's own brand.

Make a demo (opens programmable.digital 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.

  • Shuttlebase growth strategyAn autonomous AI conversion optimization platform for online stores: one tracking script finds where a store loses sales, designs and…21 growth ideas
  • Swarmia growth strategyAn engineering intelligence platform: engineering and DORA metrics, AI adoption, impact and cost tracking, developer experience surveys…23 growth ideas
  • Marble growth strategyAn AI-native back-of-house operating system for restaurants: voice and photo inventory counts, demand forecasting, agentic ordering, prep…24 growth ideas
  • Melody Ehsani growth strategyAn LA women-owned sport lifestyle brand (est. 2007) selling jewelry, custom name pieces, apparel and accessories online and from its…20 growth ideas
  • Jester Swink growth strategyA one-artist studio in Boulder Creek, in the Santa Cruz Mountains, making nature-inspired ceramic wall hangings, ceramic and metal clay…21 growth ideas
  • Sol growth strategyA spiritual wellness app with personalized 4-week plans, practices (meditation, breathwork, journaling, prayer, yoga), quizzes and…20 growth ideas
  • SleepyCat growth strategyAn Indian sleep brand selling mattresses (memory foam, latex, hybrid), pillows, toppers, bedding, beds, recliners, ErgoVerse cushions and…21 growth ideas
  • Selfin growth strategyA broker for business event risk: Selfin takes a risk that insurance and futures do not cover (a club's relegation, an airport hub…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. Parametric did not commission, review or endorse it, and Programmable is not affiliated with Parametric. 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 Parametric runs its business.

Something wrong, or do you represent Parametric? 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

    Make hiring Bit understandable: scoping form, hire terms, task pages, the help-request story, one clear market story, a monthly progress number and a faster hero image.

  2. Days 31 to 6012 ideas

    Build proof and pipeline: Bastion page, first case study, pack-out job-post prospecting, bring-your-box trials, customer reports, a service commitment, buyer notes on research, short clips, packaging partners, a site-readiness page, laundromat visits and customer referrals.

  3. Days 61 to 905 ideas

    Decide on the bets: per-unit pricing, the robot laundromat as a product, an open task benchmark, commercial linen laundries and staffing-agency partners.

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 hiring a robot instead of buying and integrating one. The most cited problem: repetitive hand work that is hard to staff: box assembly, pack out, taping and folding.

  • A robot that learns their own task from customer feedback
  • Uncertainty about cost, timeline and what happens when a robot gets stuck
  • In their words: “hire a robot”

Market

Competitive landscape

3 competitors compared: Formic, Dexterity and Verne Robotics.

  • Gap to exploit: A clear, published way to hire a two-armed robot for packing and folding
  • Formic: Built around fixed cells such as palletizing and case packing rather than a learning two-armed robot
  • Dexterity: Aimed at large enterprises, not small packing and folding operations

Hiring signals

Parametric is small (seven people on its Y Combinator profile) and its one open role on Y Combinator's job board says a lot: a Robotics Technician to build robots from drawings and BOMs, test them before they leave the…

  • Robotics Technician: Moving from prototype to production, with repairs and maintenance at deployment sites. Ideas 11, 13, 17.

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