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

Views

  • Plan
  • Roadmap
  • Research

On this page

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

growth playbook · Nov 10, 2026

rainbrand.com (opens in a new tab)

Rainbrand growth playbook

Diagnosis

Rainbrand (Bloom until October 15, 2026) is a three-person YC P26 team with a sharp position, "the brand layer for agents", more than 500,000 assets generated and two customers that build it into their own products. It also has an unusual top-of-funnel asset: 201 brand-guideline pages for brands like Adidas and 7-Eleven, plus a free no-signup Brand Kit Generator.

The two are not connected: guideline pages end in a generic sign-in button, and the generator is not linked from them or from the blog. This playbook first turns that library into a "now make yours" funnel and makes the 1,000-credit signup grant land a finished first task.

Next come the pages that product builders and agencies need before they buy: real numbers in customer stories, a trust page, an embed pricing path and an agency page. After that it widens the product where the site already sells (video Review, Figma, brand freshness), and the bigger bets are an open Brand Skill format, a public on-brand benchmark and a pull-request brand check.

Growth Score

7/10

Biggest Opportunity

Convert the people already reading 200 famous brands' guidelines into users who build their own brand, then let embedding customers carry Rainbrand into hundreds of brands each.

Biggest Risk

Brandfetch, Frontify and Context.dev all now pitch brand data or guidelines to AI agents, two with free tiers or MCP servers, so a three-person team has to win on maintained brand context and Review before larger players copy the position.

01

Start here

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

  1. 1

    Turn the 200 brand-guideline pages into "now make yours": a website box on every page that builds the visitor's own brand

    Add a plain "Make your own brand kit" link from the Adidas and 7-Eleven pages to the Brand Kit Generator and count clicks for a week.

    High impact First result in 2 weeks
  2. 2

    Put the free Brand Kit Generator in the nav, the blog and the use-case pages, and hand each kit straight into a free brand

    Add the generator link to the nav and footer.

    Medium impact First result in 2 weeks
  3. 3

    Make the 1,000-credit signup grant finish one real task: a guided first brief, image and review

    Count how many new accounts reach a first Review today.

    High impact First result in 3 weeks

02

Your idea portfolio

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

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

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

Effort vs impactEach number is an idea. Select one to open it.
Impact →
Quick wins · 3Big bets · 7Fill-ins · 11Reconsider · 2
Effort →

Showing 23 of 23 ideas

This week 7

Quick wins with a clear first action.

  • 01

    Rainbrand's largest content section is 201 brand-guideline pages (Adidas, 7-Eleven and others) where people searching for a famous brand's logo, colours and fonts can download a Brand Skill. Rainbrand already has a free, no-signup Brand Kit Generator that does the obvious next thing a visitor wants, showing the same for their own website. None of the guideline pages link to it, and the only product step on the page is "Get started", which goes to a sign-in screen. A marketer who has just studied Adidas's guidelines has no one-click way to see their own brand the same way.

    High impactLow effort First result in 2 weeks Founding Engineer

    First action

    Add a plain "Make your own brand kit" link from the Adidas and 7-Eleven pages to the Brand Kit Generator and count clicks for a week.

    How to measure it

    Guideline-page visitors who create their first brand

    Now
    –
    Target
    2% of guideline-page visitors make a kit and 15% of kit makers sign up
    By
    60 days

    Steps

    1. 1

      Baseline

      Sessions, Brand Skill downloads and signups for each guideline page over the last 30 days.

      Founder & CEO 1 day

    2. 2

      Add the block

      One website field with "See your brand like this" in the shared template, under the hero and after the Brand Skill section.

      Founding Engineer 2 days

    3. 3

      Render inline

      Run the generator in place and show the result in the same logo, colour and font layout as the page.

      Founding Engineer 3 days

    4. 4

      Prefill signup

      Pass the website to sign-in so the free brand creation starts from it.

      Founding Engineer 2 days

    5. 5

      Protect capacity

      Keep the three-a-day cap and queue rather than fail when busy.

      Founding Engineer 1 day

    6. 6

      Write category lines

      Different prompt copy for retail, AI and developer-tool pages.

      Founder & CEO 1 day

    7. 7

      Measure

      Funnel by page and category; add brands in the categories that convert.

      Founder & CEO weekly

    First result
    2 weeks
    Cost
    About a week of engineering; generator runs are already free
    Owner
    Founding Engineer

    Smallest test

    Add a static link to the generator on the ten most-visited guideline pages and compare signups for two weeks.

    Why it works

    Add a "See your brand like this" block to the shared guideline template, under the hero and again after the Brand Skill section, on all 201 pages. It is one website field that runs the Brand Kit Generator in place and renders the visitor's logos, colours and fonts in the same layout as the page they are reading. Under the result, offer "Turn this into a maintained brand for your agents": sign-in with the website prefilled, so the one free brand creation in the signup grant starts from the URL they already typed. Track page view, then kit, then signup, then first brief, for each guideline page and category, and use the winners to choose which brands to add next.

    Visitors arrive with the intent "what would this look like for us", and a one-field box converts that intent before they leave. The kit is a free preview of the paid product, it is already capped at three a day, and it uses no credits until the visitor signs up and claims the free brand.

    Ready to use

    ChecklistGuideline-page "make yours" block

    [ ] Website field under the hero and after the Brand Skill section
    [ ] Inline kit in the same layout as the page
    [ ] "Turn this into a maintained brand" with the website prefilled at sign-in
    [ ] Free brand creation starts from that URL
    [ ] Three-a-day cap and a queue message
    [ ] Disclaimer kept: the featured guide is not official
    [ ] Events: kit_started, kit_done, signup, first_brief, tagged by page and category

    Risks and how to handle them

    • Many visitors are students or fans, not buyers

      Measure by business websites entered and signups, not kit count.

    • Generator load spikes

      Cache results by domain and keep the daily cap.

    • Confusion with the featured brand

      Keep the not-official disclaimer and label the block clearly as "your brand".

    Evidence

    • Your websiteBrand guidelines library: 201 URLs under /brand-guidelines/ in the sitemap (latest November 4, 2026), headed "Brand guidelines examples" and covering "200 brands, with more coming". Each page, such as "Adidas brand guidelines: logo, colors and fonts", is built from the brand's own website, offers a downloadable Brand Skill ZIP, and says it "isn't Adidas's official guide" and that Rainbrand "isn't affiliated with or endorsed by adidas AG".Source
    • Your website7-Eleven guidelines page (checked November 10, 2026): links to download the Brand Skill and logo files, three related brands, "Get started" (the sign-in page), the docs and plugin and connector setup. There is no link to the Brand Kit Generator and no field to enter your own website.Source
    • Your websiteBrand Kit Generator: "Enter a website and get its brand kit: logos for light and dark backgrounds, colors with their codes, font names and the tagline." "No signup", 3 kits per visitor per day, delivered as a ZIP; the next step offered is "Get started". Neither the blog index nor the guideline pages link to it.Source
    • Your websitePricing: Plus $20/mo (5,000 credits, 1 seat), Pro $35/mo (10,000 credits, unlimited seats), Max $50/mo (20,000), Scale $65/mo (30,000), a volume tier "From $360/mo" and "Custom" through "Talk to us" by email. A new brand costs 500 credits, a brief 100, a 2K image 100, a review 200 and a brand edit 300. New sign-ups get 1,000 credits and one free brand creation, "a one-time signup grant, not a monthly allowance"; there is no free tier. If credits run out mid-task the request stops and links to the pricing page. Annual billing saves 10%.Source
  • 02

    The Brand Kit Generator is Rainbrand's only no-signup tool, yet it appears in the sitemap and nowhere else that was checked: the blog index, the guideline pages and the use-case pages do not link to it. The blog already has posts such as "How to turn your website into brand guidelines AI can use", which is exactly what the tool previews.

    Medium impactLow effort First result in 2 weeks Founder & CEO

    First action

    Add the generator link to the nav and footer.

    How to measure it

    Brand kits made

    Now
    –
    Target
    3x the current weekly kits, with 10% signing up
    By
    45 days

    Steps

    1. 1

      Measure today

      Kits made per week and where visitors came from.

      Founder & CEO 1 day

    2. 2

      Nav and footer

      Add "Free brand kit".

      Founding Engineer 1 hour

    3. 3

      Embed in posts

      The website and PDF guideline posts first.

      Founder & CEO 1 day

    4. 4

      Use-case pages

      A "start with your website" block on all 11.

      Founding Engineer 1 day

    5. 5

      Handoff button

      Prefilled sign-in for the free brand after each kit.

      Founding Engineer 1 day

    6. 6

      Track

      Kits and signups by source page.

      Founder & CEO weekly

    First result
    2 weeks
    Cost
    $0, staff time only
    Owner
    Founder & CEO

    Smallest test

    Embed the tool in one blog post and compare its signups for two weeks.

    Why it works

    Add the generator to the main nav and footer as "Free brand kit", embed it in the blog posts about turning a website or PDF into guidelines, and add it to the 11 use-case pages as "start with your website". After a kit is made, show one button, "Keep this brand up to date for your agents", that opens sign-in with the website prefilled for the free brand creation.

    A free tool only grows signups if people can find it, and the posts that explain the problem are the best place to offer the instant answer.

    Risks and how to handle them

    • Kit seekers never sign up

      Show what the maintained brand adds: briefs, edits and Reviews.

    Evidence

    • Your websiteBrand Kit Generator: "Enter a website and get its brand kit: logos for light and dark backgrounds, colors with their codes, font names and the tagline." "No signup", 3 kits per visitor per day, delivered as a ZIP; the next step offered is "Get started". Neither the blog index nor the guideline pages link to it.Source
    • Your websiteBlog: 52 URLs in the sitemap and 42 articles in 8 topics (including Agencies, Comparisons, Review and Developers); the index shows no dates or authors and has no newsletter signup; one title reads "11 brand management tools for AI agents (2027)".Source
    • Your websiteUse cases: 11 pages, including static ads, social graphics, product imagery, slides, campaign email, product video, UGC video, motion graphics and email signature GIF. The static ads page shows no prices, links to "View pricing" and asks visitors to "Make your first ad set".Source
  • 03

    There is no free tier: new users get one free brand plus 1,000 credits once. That buys roughly one brief (100), a few 2K images (100 each) and one review (200), and when credits run out mid-task the request simply stops. A new user who explores freely can spend the grant before seeing the full Brief, generate and Review loop that makes Rainbrand different.

    High impactLow effort First result in 3 weeks Founding Engineer

    First action

    Count how many new accounts reach a first Review today.

    How to measure it

    First-session full loop

    Now
    –
    Target
    50% of new accounts
    By
    45 days

    Steps

    1. 1

      Measure the loop

      Share of new accounts reaching brief, image and Review.

      Founder & CEO 1 day

    2. 2

      Design the task

      Three-image ad set plus one Review, within 600 credits.

      Founder & CEO 1 day

    3. 3

      Build the guide

      One flow after brand creation, in the app and through MCP.

      Founding Engineer 3 days

    4. 4

      Show costs

      Credit cost before each action during the grant.

      Founding Engineer 1 day

    5. 5

      End screen

      Balance left and the matching plan.

      Founding Engineer 1 day

    6. 6

      Compare cohorts

      Paid conversion for guided against unguided accounts.

      Founder & CEO monthly

    First result
    3 weeks
    Cost
    A week of engineering
    Owner
    Founding Engineer

    Smallest test

    Email ten new users a one-line prompt for the guided task and compare their conversion.

    Why it works

    After the free brand is created, offer one guided first task picked from the use cases: "Make a 3-image ad set and check it". It runs one brief, three images and one review (600 credits), shows the Review verdict on each image, and ends with the remaining balance and the plan that fits. Show the credit cost of each action before it runs during the grant period.

    Users who see a Review catch an off-brand image understand the product in one session, and showing costs up front stops the grant from running out in the middle of a task.

    Risks and how to handle them

    • Agent users never open the app

      Offer the same guided task as an MCP prompt.

    Builds on

    Evidence

    • Your websitePricing: Plus $20/mo (5,000 credits, 1 seat), Pro $35/mo (10,000 credits, unlimited seats), Max $50/mo (20,000), Scale $65/mo (30,000), a volume tier "From $360/mo" and "Custom" through "Talk to us" by email. A new brand costs 500 credits, a brief 100, a 2K image 100, a review 200 and a brand edit 300. New sign-ups get 1,000 credits and one free brand creation, "a one-time signup grant, not a monthly allowance"; there is no free tier. If credits run out mid-task the request stops and links to the pricing page. Annual billing saves 10%.Source
    • Your websiteHomepage: "The brand layer for agents"; sources In Files, In Website and In Figma & more; outputs MCP ("Use it with your agents") and REST API ("Embed it into your product"); Briefs, Generation through an AI Gateway, and Reviews that return pass, fail, uncertain or not applicable; testimonials from Revnu, Letter and The GTM Co-Founder, and no usage figures in the page text.Source
    • Your websiteUse cases: 11 pages, including static ads, social graphics, product imagery, slides, campaign email, product video, UGC video, motion graphics and email signature GIF. The static ads page shows no prices, links to "View pricing" and asks visitors to "Make your first ad set".Source
  • 04

    Rainbrand's real rival for most small teams is not Frontify but copying brand rules into each assistant's custom instructions or project. The FAQ already answers "why not custom instructions", but none of the 11 comparison pages covers it, and those pages are dated, sourced and well built.

    Medium impactLow effort First result in 1 month Founder & CEO

    First action

    Turn the FAQ answer into an outline.

    How to measure it

    Comparison-page signups

    Now
    –
    Target
    30 a month
    By
    90 days

    Steps

    1. 1

      Outline

      From the FAQ answer and the Frontify page format.

      Founder & CEO 2 hours

    2. 2

      Worked example

      One guideline change across three tools.

      Founding Engineer 1 day

    3. 3

      Write

      Fair, dated and sourced.

      Founder & CEO 1 day

    4. 4

      Migration step

      Build a brand from pasted instructions.

      Founding Engineer 2 days

    5. 5

      Publish and link

      From the FAQ and the compare index.

      Founder & CEO 1 hour

    6. 6

      Refresh

      When assistants change their features.

      Founder & CEO quarterly

    First result
    1 month
    Cost
    Staff time only
    Owner
    Founder & CEO

    Smallest test

    Publish the page without the migration step and track signups.

    Why it works

    Publish "Rainbrand vs custom instructions" in the same format as the Frontify page: what each does, how updates spread across tools, a worked example where one guideline change reaches every assistant through the next Brief, and what Review adds. Close with a short migration: paste your current instructions and Rainbrand builds the brand from them.

    It meets the most common objection at the moment of search, in the format the team already uses for comparisons.

    Risks and how to handle them

    • Assistant features change fast

      Date the page and recheck sources quarterly.

    Evidence

    • Your websiteComparison pages: 11 (Frontify, Adobe, Canto, Canva Brand Kit, Jasper, Marq, Typeface, Writer, Brandfetch, Brandfolder, Bynder), dated and sourced; the Frontify page was updated November 3, 2026 and its CTA is "Connect your agents". None compares Rainbrand with keeping guidelines in each assistant's custom instructions, or with web-context APIs.Source
    • Your websiteFAQ (21 questions): credits, agencies ("Keep each client as a separate brand in your Rainbrand team, with its own guidelines, brand voice and assets"; no cap on brands, but adding one uses credits), MCP vs REST, and why not custom instructions or a single assistant's project. No question covers security, data handling or compliance; help is by email to [email protected].Source
  • 05

    Agencies are a named target in the YC launch and have their own blog topic, and the FAQ says each client can be a separate brand. But there is no agency landing page, and the pricing page makes agencies work out how many client brands fit: each new brand is 500 credits, and plans include 10, 20, 40 or 60 new brands.

    Medium impactLow effort First result in 1 month Founder & CEO

    First action

    Write the cost table for 10, 20 and 60 client brands.

    How to measure it

    Teams with 5+ brands

    Now
    –
    Target
    25 new teams
    By
    90 days

    Steps

    1. 1

      Cost table

      Brands and monthly assets per plan.

      Founder & CEO 2 hours

    2. 2

      Write the page

      Math, separation and approvals.

      Founder & CEO 1 day

    3. 3

      Add proof

      The several-brands customer story.

      Founder & CEO 1 hour

    4. 4

      Bulk import

      A list of client websites in one go.

      Founding Engineer 3 days

    5. 5

      Link it

      From pricing, FAQ and the agencies blog topic.

      Founding Engineer 1 hour

    6. 6

      Track

      Teams creating five or more brands.

      Founder & CEO monthly

    First result
    1 month
    Cost
    Staff time only
    Owner
    Founder & CEO

    Smallest test

    Publish the cost table on the agencies blog topic page first.

    Why it works

    Build /agencies with three things: the per-client math (for example 20 client brands and monthly ad sets on Pro), how clients stay separate and get approvals, and the GTM Co-Founder story of one agent across several brands. Add an "agency starter" that imports a list of client websites in one go.

    Agencies buy on cost per client and on separation; answering both on one page shortens the decision, and each agency brings many brands.

    Risks and how to handle them

    • Bulk import burns credits by surprise

      Show the total cost before import.

    Evidence

    • YC launchLaunch post: "We launched v1 in December and went viral" and "Since then we've generated over 175,000 on-brand assets for our customers." Rainbrand ingests websites, decks, Figma, socials and codebases and exposes them through an API, MCP and a forthcoming CLI; it asks to hear from builders of marketing, coding and e-commerce agents, products that need brand context, and agencies and marketing teams.Source
    • Your websiteFAQ (21 questions): credits, agencies ("Keep each client as a separate brand in your Rainbrand team, with its own guidelines, brand voice and assets"; no cap on brands, but adding one uses credits), MCP vs REST, and why not custom instructions or a single assistant's project. No question covers security, data handling or compliance; help is by email to [email protected].Source
    • Your websitePricing: Plus $20/mo (5,000 credits, 1 seat), Pro $35/mo (10,000 credits, unlimited seats), Max $50/mo (20,000), Scale $65/mo (30,000), a volume tier "From $360/mo" and "Custom" through "Talk to us" by email. A new brand costs 500 credits, a brief 100, a 2K image 100, a review 200 and a brand edit 300. New sign-ups get 1,000 credits and one free brand creation, "a one-time signup grant, not a monthly allowance"; there is no free tier. If credits run out mid-task the request stops and links to the pricing page. Annual billing saves 10%.Source
    • Your websiteBlog: 52 URLs in the sitemap and 42 articles in 8 topics (including Agencies, Comparisons, Review and Developers); the index shows no dates or authors and has no newsletter signup; one title reads "11 brand management tools for AI agents (2027)".Source
  • 06

    The careers page says more than 500,000 assets have been generated with Rainbrand, and Revnu alone has created hundreds of brands through it. Yet the homepage text carries no usage figures, and none of the four customer stories has a results metric.

    Medium impactLow effort First result in 2 weeks Founder & CEO

    First action

    Pull assets, brands and Reviews to date from the database.

    How to measure it

    Homepage signup rate

    Now
    –
    Target
    +15%
    By
    60 days

    Steps

    1. 1

      Pull the numbers

      Assets, brands and Reviews to date.

      Founding Engineer 2 hours

    2. 2

      Add the proof line

      Homepage and pricing.

      Founding Engineer 2 hours

    3. 3

      Ask customers

      One number each from the four stories.

      Founder & CEO 1 week

    4. 4

      Update stories

      The number on each title card.

      Founder & CEO 1 day

    5. 5

      Keep it live

      Refresh the counts monthly.

      Founding Engineer monthly

    6. 6

      Measure

      Signup rate before and after.

      Founder & CEO monthly

    First result
    2 weeks
    Cost
    $0
    Owner
    Founder & CEO

    Smallest test

    Add the 500,000 assets line to the homepage this week.

    Why it works

    Add a proof line to the homepage and pricing page (assets generated, brands created, Review checks run). Go back to each of the four customers for one number: brands served, assets a month, time saved per campaign or Review failures caught before shipping. Put the number in each story's title card.

    Product builders choosing an infrastructure vendor look for scale and production use, and the numbers already exist in Rainbrand's own data.

    Risks and how to handle them

    • Customers will not share numbers

      Use counts Rainbrand can see, with their approval.

    Evidence

    • HiringFounding Engineer (San Francisco, in person, $140K-$200K base plus 0.5%-3% equity), the one open role: "We're a three-person team in YC P26", backed by Guillermo Rauch and Nebular VC, and "More than 500,000 assets have been generated with Rainbrand". The role builds the agents, ingestion of websites, social media and files, retrieval, AI-quality evaluations, and the API and MCP that other products and agents use.Source
    • Your websiteCustomer stories: four (Revnu, Letter, The GTM Co-Founder, Autumn), two "Built into their product" and two "In their agents", none with results metrics. Revnu (customer since April 2026) uses the REST API and has created hundreds of brands through Rainbrand; Letter (since June 2026): "We pushed Rainbrand to production as our default image solution."Source
    • Your websiteHomepage: "The brand layer for agents"; sources In Files, In Website and In Figma & more; outputs MCP ("Use it with your agents") and REST API ("Embed it into your product"); Briefs, Generation through an AI Gateway, and Reviews that return pass, fail, uncertain or not applicable; testimonials from Revnu, Letter and The GTM Co-Founder, and no usage figures in the page text.Source
  • 07

    Google's real-visitor data shows the homepage failing Core Web Vitals: p75 LCP is 2.9 s and INP 211 ms. The largest element is the hero paragraph itself, so it is waiting on scripts rather than images, and the page ships 1.3 MB of JavaScript with 435 KB unused. The chat widget (232 KB) and the Facebook pixel (213 KB) load on every first visit.

    Medium impactLow effort First result in 1 month Founding Engineer

    First action

    Switch the chat widget to load on click.

    How to measure it

    Core Web Vitals

    Now
    –
    Target
    LCP under 2.5 s and INP under 200 ms
    By
    45 days

    Steps

    1. 1

      Defer chat

      Load on click.

      Founding Engineer 2 hours

    2. 2

      Delay the pixel

      After the page is interactive.

      Founding Engineer 1 hour

    3. 3

      Server-render the hero

      Text visible without client JavaScript.

      Founding Engineer 1 day

    4. 4

      Split bundles

      Remove the unused JavaScript from the homepage route.

      Founding Engineer 1 day

    5. 5

      Retest

      PageSpeed mobile and desktop.

      Founding Engineer 1 hour

    6. 6

      Watch field data

      28-day Core Web Vitals.

      Founding Engineer monthly

    First result
    1 month
    Cost
    Two days of engineering
    Owner
    Founding Engineer

    Smallest test

    Defer the chat widget and rerun PageSpeed today.

    Why it works

    Render the hero text without waiting for client JavaScript, load the chat widget only when a visitor clicks to open it, load the ad pixel after the page is interactive, and split the unused JavaScript off the homepage route. Retest in PageSpeed and watch the field data over 28 days.

    A site that sells to developers is judged on speed, and these changes need no redesign; PageSpeed estimates about 0.75 s from unused JavaScript alone.

    Risks and how to handle them

    • Fewer chat conversations

      Keep a visible chat button and compare conversations.

    Evidence

    • PageSpeedHomepage (November 10, 2026): mobile performance 80, desktop 98. Real Chrome visitors' p75 LCP is 2.9 s and INP 211 ms, so the page fails Core Web Vitals on both; lab LCP 3.2 s, with the hero paragraph as the LCP element; 2.2 MB in total and 1.3 MB of JavaScript, 435 KB of it unused (about 0.75 s); the chat widget loads 232 KB and the Facebook pixel 213 KB.Source

This quarter 11

The core plays for the next three months.

  • 08

    Half of Rainbrand's published customers build it into their own products through the REST API, and that is the fastest-growing kind of account (one has created hundreds of brands). Product builders get asked by their own customers where brand data goes. Rainbrand's FAQ has 21 questions but none on security, data handling or compliance, while Context.dev, which also sells brand data to agents, advertises SOC 2 Type 2.

    High impactMedium effort First result in 1 quarter Founder & CEO

    First action

    List every service that touches brand data.

    How to measure it

    New API-embedding customers

    Now
    –
    Target
    5 a quarter
    By
    2 quarters

    Steps

    1. 1

      Map data

      What is stored, where and for how long.

      Founding Engineer 2 days

    2. 2

      List subprocessors

      Hosting, storage, model providers.

      Founding Engineer 1 day

    3. 3

      Write the page

      Plain answers and a security contact.

      Founder & CEO 2 days

    4. 4

      Add status

      Public uptime page.

      Founding Engineer 1 day

    5. 5

      Answer pack

      Standard questionnaire answers for embedders.

      Founder & CEO 3 days

    6. 6

      Plan the audit

      Scope and cost, triggered by the first deal that needs it.

      Founder & CEO 1 week

    First result
    1 quarter
    Cost
    Staff time; a later audit costs extra
    Owner
    Founder & CEO

    Smallest test

    Add three security questions to the FAQ this week.

    Why it works

    Publish a /trust page that states plainly what is stored, where, for how long, which subprocessors and model providers see brand data, whether customer data trains models, uptime and status, and a security contact. Add a short security questionnaire answer pack for embedders, and plan a compliance audit when the first larger deal asks for it.

    Embed deals stall on security questions, not features; a clear page answers most of them before a call.

    Risks and how to handle them

    • Promising more than is true

      State only current practice and date the page.

    Evidence

    • Your websiteCustomer stories: four (Revnu, Letter, The GTM Co-Founder, Autumn), two "Built into their product" and two "In their agents", none with results metrics. Revnu (customer since April 2026) uses the REST API and has created hundreds of brands through Rainbrand; Letter (since June 2026): "We pushed Rainbrand to production as our default image solution."Source
    • Your websiteFAQ (21 questions): credits, agencies ("Keep each client as a separate brand in your Rainbrand team, with its own guidelines, brand voice and assets"; no cap on brands, but adding one uses credits), MCP vs REST, and why not custom instructions or a single assistant's project. No question covers security, data handling or compliance; help is by email to [email protected].Source
    • Competitor researchContext.dev (brand.dev now redirects there): "The unified web context API for software and AI agents", with brand data and styleguides, a free plan of 1,000 credits a month with no card, a $20 Developer plan, an MCP server and SOC 2 Type 2.Source
  • 09

    Rainbrand's most valuable accounts create a brand for each of their own customers, but the self-serve plans top out at 60 new brands a month, the volume tier is "From $360/mo", and anything larger goes to "Talk to us" through the support email. A product team sizing a launch has no price per brand to plan with and no way to book a call.

    High impactMedium effort First result in 1 month Founder & CEO

    First action

    Set the per-brand price at 500 and 2,000 new brands a month.

    How to measure it

    Embed calls booked

    Now
    –
    Target
    8 a month
    By
    90 days

    Steps

    1. 1

      Set unit prices

      Per brand and per brief at volume.

      Founder & CEO 2 days

    2. 2

      Build the calculator

      Brands and assets per month.

      Founding Engineer 2 days

    3. 3

      Add booking

      Replace the email-only contact.

      Founder & CEO 1 hour

    4. 4

      Document the pattern

      Brand per customer through the REST API.

      Founding Engineer 2 days

    5. 5

      Link from stories

      The two embed customer stories.

      Founder & CEO 1 hour

    6. 6

      Review deals

      Calculator inputs against deals won.

      Founder & CEO monthly

    First result
    1 month
    Cost
    Staff time only
    Owner
    Founder & CEO

    Smallest test

    Add a booking link next to "Talk to us" this week.

    Why it works

    Add an "Embed" section to pricing: price per new brand and per brief at volume, a calculator ("brands per month, assets per brand"), the sandbox they can build on with the signup grant, and a booking link for the founder. Publish the embed pattern already used by the two customers that build Rainbrand into their products.

    Usage-based buyers plan with unit prices; a calculator and a booking link turn "email us" into a scheduled call with a sized deal.

    Risks and how to handle them

    • Locking in prices too early

      Mark volume prices as starting points.

    Builds on

    Evidence

    • Your websitePricing: Plus $20/mo (5,000 credits, 1 seat), Pro $35/mo (10,000 credits, unlimited seats), Max $50/mo (20,000), Scale $65/mo (30,000), a volume tier "From $360/mo" and "Custom" through "Talk to us" by email. A new brand costs 500 credits, a brief 100, a 2K image 100, a review 200 and a brand edit 300. New sign-ups get 1,000 credits and one free brand creation, "a one-time signup grant, not a monthly allowance"; there is no free tier. If credits run out mid-task the request stops and links to the pricing page. Annual billing saves 10%.Source
    • Your websiteCustomer stories: four (Revnu, Letter, The GTM Co-Founder, Autumn), two "Built into their product" and two "In their agents", none with results metrics. Revnu (customer since April 2026) uses the REST API and has created hundreds of brands through Rainbrand; Letter (since June 2026): "We pushed Rainbrand to production as our default image solution."Source
  • 10

    When credits run out mid-task, the request stops and links back to the pricing page. For a user it is an annoyance. For a product that embeds Rainbrand, a stopped request means a broken experience for their own customer. Auto top-up exists but is opt-in, and plan credits do not roll over.

    Medium impactLow effort First result in 1 month Founding Engineer

    First action

    Count requests stopped for low credits last month.

    How to measure it

    Credit-stopped requests

    Now
    –
    Target
    -80%
    By
    60 days

    Steps

    1. 1

      Count stops

      By account and plan.

      Founding Engineer 1 day

    2. 2

      Balance in replies

      API and MCP responses.

      Founding Engineer 2 days

    3. 3

      Warnings

      Email at 25% and 10%.

      Founding Engineer 1 day

    4. 4

      Finish the task

      Small overdraft at the plan rate.

      Founding Engineer 3 days

    5. 5

      Offer auto top-up

      In the first warning.

      Founder & CEO 1 day

    6. 6

      Measure

      Stops, top-ups and upgrades.

      Founder & CEO monthly

    First result
    1 month
    Cost
    A week of engineering
    Owner
    Founding Engineer

    Smallest test

    Add the 10% warning email first.

    Why it works

    Send low-balance warnings at 25% and 10% by email and in API responses, return the remaining balance in every API and MCP reply, let a running task finish on a small overdraft charged at the plan rate, and offer auto top-up at the moment of the first warning.

    Failing gracefully protects the accounts that depend on Rainbrand in production, and the warning is the best moment to sell auto top-up or the next plan.

    Risks and how to handle them

    • Unpaid overdrafts

      Cap the overdraft and limit it to paid accounts.

    Evidence

    • Your websitePricing: Plus $20/mo (5,000 credits, 1 seat), Pro $35/mo (10,000 credits, unlimited seats), Max $50/mo (20,000), Scale $65/mo (30,000), a volume tier "From $360/mo" and "Custom" through "Talk to us" by email. A new brand costs 500 credits, a brief 100, a 2K image 100, a review 200 and a brand edit 300. New sign-ups get 1,000 credits and one free brand creation, "a one-time signup grant, not a monthly allowance"; there is no free tier. If credits run out mid-task the request stops and links to the pricing page. Annual billing saves 10%.Source
    • Your websiteCustomer stories: four (Revnu, Letter, The GTM Co-Founder, Autumn), two "Built into their product" and two "In their agents", none with results metrics. Revnu (customer since April 2026) uses the REST API and has created hundreds of brands through Rainbrand; Letter (since June 2026): "We pushed Rainbrand to production as our default image solution."Source
  • 11

    Rainbrand's own showcase makes the gap visible: in the Eager example, Review caught an unapproved purple and the wrong typeface on the stills, but the 12-second reel and the 20-second film were not reviewed because Review only checks still images. Meanwhile four of the 11 use-case pages sell video: product video, UGC video, motion graphics and email signature GIFs.

    High impactHigh effort First result in 1 quarter Founding Engineer

    First action

    Run the current image Review on ten frames of the Eager reel by hand.

    How to measure it

    Video Reviews a month

    Now
    –
    Target
    1,000
    By
    2 quarters

    Steps

    1. 1

      Manual test

      Image Review on sampled frames.

      Founding Engineer 1 day

    2. 2

      Frame sampling

      Scene changes, end cards and text frames.

      Founding Engineer 2 weeks

    3. 3

      Timestamped verdicts

      Pass, fail or uncertain with the time and reason.

      Founding Engineer 2 weeks

    4. 4

      Evaluate

      Against a labelled set of on- and off-brand clips.

      Founding Engineer 1 week

    5. 5

      Price

      Credits per second or per clip.

      Founder & CEO 1 day

    6. 6

      Update examples

      Re-run Eager and update the video use-case pages.

      Founder & CEO 1 week

    First result
    1 quarter
    Cost
    A quarter of engineering time and model costs
    Owner
    Founding Engineer

    Smallest test

    Offer frame-sampled Review on request to two customers.

    Why it works

    Ship Review for video by sampling key frames and end cards, checking logo, colours, type and on-screen copy on each, and returning a verdict with timestamps. Price it in credits like image Review, then re-run the Eager example with the reel and film reviewed and update the use-case pages.

    Video is where brand mistakes are most expensive to redo, and Review is the feature competitors' MCP servers do not offer.

    Risks and how to handle them

    • Too many false failures

      Return "uncertain" with frames rather than fail.

    Evidence

    • Your websiteExamples: nine brand mock-ups (Eager, Kaorium, Son of a Tailor, Yumbun, Gut Oggau, Friedhats, Extropic, Heart Aerospace, Sazabi). In the Eager example one prompt produced a landing page, an email, a carousel, a 12-second reel and a 20-second film; Review caught a headline in an unapproved purple and lettering in the wrong typeface, but the reel and film were not reviewed because Review only checks still images.Source
    • Your websiteUse cases: 11 pages, including static ads, social graphics, product imagery, slides, campaign email, product video, UGC video, motion graphics and email signature GIF. The static ads page shows no prices, links to "View pricing" and asks visitors to "Make your first ad set".Source
    • Competitor researchBrandfetch: "The brand data platform" with "50M+ brands", a Logo API "free up to 1M monthly requests" and a Brand Context API. Frontify: "Brand intelligence for the AI era", reaching "every AI agent via MCP and API", with enterprise logos.Source
  • 12

    Rainbrand's promise is a maintained brand, and brand edits are already a paid action (300 credits) that is charged only when the proposed changes are ready. But edits start only when a user asks. Brands change their sites, campaigns and products without telling anyone, so a brand built in April quietly goes stale.

    High impactMedium effort First result in 2 months Founding Engineer

    First action

    Re-crawl 50 brands older than three months and count how many changed.

    How to measure it

    Applied freshness edits

    Now
    –
    Target
    30% of alerts applied
    By
    90 days

    Steps

    1. 1

      Measure drift

      Re-crawl 50 older brands.

      Founding Engineer 2 days

    2. 2

      Change detection

      Logos, colours, type, products and voice.

      Founding Engineer 2 weeks

    3. 3

      Draft edits

      Prepared edits the owner can apply or discard.

      Founding Engineer 1 week

    4. 4

      Alert email

      Before and after, one click to review.

      Founder & CEO 3 days

    5. 5

      API switch

      Freshness on for embedders' customer brands.

      Founding Engineer 1 week

    6. 6

      Measure

      Alerts, applied edits and retention.

      Founder & CEO monthly

    First result
    2 months
    Cost
    Six weeks of engineering plus crawl costs
    Owner
    Founding Engineer

    Smallest test

    Email ten customers a manual "your brand changed" note.

    Why it works

    For each brand built from a website, re-crawl the source monthly, compare logos, colours, type, products and voice, and when something changes, prepare a draft brand edit the owner can apply or discard. Send a short "your brand changed" email with the before and after. Let embedders turn it on for all their customer brands through the API.

    It turns the product promise into a monthly reason to come back, and the edits it proposes are already a paid action.

    Risks and how to handle them

    • Noisy alerts

      Alert only on changes above a threshold.

    • Surprise credit charges

      Ask for opt-in before charging for automatic edits.

    Evidence

    • Your websitePricing: Plus $20/mo (5,000 credits, 1 seat), Pro $35/mo (10,000 credits, unlimited seats), Max $50/mo (20,000), Scale $65/mo (30,000), a volume tier "From $360/mo" and "Custom" through "Talk to us" by email. A new brand costs 500 credits, a brief 100, a 2K image 100, a review 200 and a brand edit 300. New sign-ups get 1,000 credits and one free brand creation, "a one-time signup grant, not a monthly allowance"; there is no free tier. If credits run out mid-task the request stops and links to the pricing page. Annual billing saves 10%.Source
    • HiringFounding Engineer (San Francisco, in person, $140K-$200K base plus 0.5%-3% equity), the one open role: "We're a three-person team in YC P26", backed by Guillermo Rauch and Nebular VC, and "More than 500,000 assets have been generated with Rainbrand". The role builds the agents, ingestion of websites, social media and files, retrieval, AI-quality evaluations, and the API and MCP that other products and agents use.Source
    • Your websiteHomepage: "The brand layer for agents"; sources In Files, In Website and In Figma & more; outputs MCP ("Use it with your agents") and REST API ("Embed it into your product"); Briefs, Generation through an AI Gateway, and Reviews that return pass, fail, uncertain or not applicable; testimonials from Revnu, Letter and The GTM Co-Founder, and no usage figures in the page text.Source
  • 13

    Every guideline page says it is not the brand's official guide and that Rainbrand is not affiliated. The people most likely to visit their own company's page are that brand's marketing and design teams, the exact buyers Rainbrand wants, and today the page gives them nothing to do.

    Medium impactMedium effort First result in 1 quarter Founder & CEO

    First action

    Add the claim link to the 20 most-visited guideline pages.

    How to measure it

    Claimed brand pages

    Now
    –
    Target
    10
    By
    2 quarters

    Steps

    1. 1

      Claim link

      On every guideline page.

      Founding Engineer 1 day

    2. 2

      Verify

      Company email domain check.

      Founding Engineer 3 days

    3. 3

      Claimed brand

      Free brand built from the Brand Skill.

      Founding Engineer 1 week

    4. 4

      Correct or remove

      Simple form and a fast response promise.

      Founder & CEO 2 days

    5. 5

      Badge

      "Maintained by the brand" once claimed.

      Founding Engineer 2 days

    6. 6

      Follow up

      A personal note to each claimer.

      Founder & CEO ongoing

    First result
    1 quarter
    Cost
    About two weeks of engineering
    Owner
    Founder & CEO

    Smallest test

    Add the link to ten pages and see if anyone claims.

    Why it works

    Add a quiet "Work at this brand? Claim this page" link. A verified company email unlocks a free brand built from the page's Brand Skill, the option to correct the public page or take it down, and an "official" badge once they maintain it in Rainbrand. Track claims by brand and follow up personally.

    It turns the main risk of the library (unofficial guides of famous brands) into a sales channel to their brand teams, and corrections improve the pages.

    Risks and how to handle them

    • Takedown requests

      Honour them quickly; it is better than a legal letter.

    Evidence

    • Your websiteBrand guidelines library: 201 URLs under /brand-guidelines/ in the sitemap (latest November 4, 2026), headed "Brand guidelines examples" and covering "200 brands, with more coming". Each page, such as "Adidas brand guidelines: logo, colors and fonts", is built from the brand's own website, offers a downloadable Brand Skill ZIP, and says it "isn't Adidas's official guide" and that Rainbrand "isn't affiliated with or endorsed by adidas AG".Source
    • Your website7-Eleven guidelines page (checked November 10, 2026): links to download the Brand Skill and logo files, three related brands, "Get started" (the sign-in page), the docs and plugin and connector setup. There is no link to the Brand Kit Generator and no field to enter your own website.Source
  • 14

    Higgsfield is the only creative-tool integration on the integrations page, listed as "Higgsfield MCP for on-brand video and images", but unlike the others it has no setup steps. Video is four of Rainbrand's 11 use cases, so this is the partner closest to them, and the page is a dead end.

    Medium impactLow effort First result in 1 month Founder & CEO

    First action

    Write the setup steps.

    How to measure it

    Higgsfield-connected accounts

    Now
    –
    Target
    100
    By
    90 days

    Steps

    1. 1

      Setup steps

      Connect, sign in and pick a brand.

      Founding Engineer 1 day

    2. 2

      Worked campaign

      Brief to finished video.

      Founder & CEO 3 days

    3. 3

      Publish

      On the integration page and a use-case page.

      Founding Engineer 1 hour

    4. 4

      Pitch the partner

      Joint tutorial and a listing.

      Founder & CEO 2 weeks

    5. 5

      Track

      Accounts connected via the integration.

      Founding Engineer monthly

    6. 6

      Repeat

      The same template for the next tool.

      Founder & CEO quarterly

    First result
    1 month
    Cost
    Staff time only
    Owner
    Founder & CEO

    Smallest test

    Publish the setup steps this week.

    Why it works

    Write the Higgsfield page with setup, one worked campaign (brief, video, Review once video Review ships) and pricing in both products' terms. Propose a joint tutorial and a listing in Higgsfield's own integrations or community, and repeat the template for the next creative tool.

    Partner pages bring users who already pay for generation and need it on-brand, and a complete page is the minimum a partner will promote.

    Risks and how to handle them

    • Partner does not promote it

      Publish the tutorial anyway and share it with their community.

    Evidence

    • Your websiteIntegrations: ChatGPT, Claude, Claude Code, Codex, Cursor and Higgsfield (listed as "Higgsfield MCP for on-brand video and images" with no setup steps), plus the API. There is no Figma, Canva, Webflow or automation-tool page, although the homepage lists "In Figma & more" as a source.Source
    • Your websiteUse cases: 11 pages, including static ads, social graphics, product imagery, slides, campaign email, product video, UGC video, motion graphics and email signature GIF. The static ads page shows no prices, links to "View pricing" and asks visitors to "Make your first ad set".Source
  • 15

    The homepage shows three inputs, Files, Website and "Figma & more", and the YC launch lists Figma among the sources Rainbrand ingests. But the integrations page covers only assistants, coding tools and one video tool. A designer searching for a way to get their Figma brand into agents finds no page.

    Medium impactMedium effort First result in 2 months Founding Engineer

    First action

    Write the integration page for the import that already works.

    How to measure it

    Figma-sourced brands

    Now
    –
    Target
    15% of new brands
    By
    2 quarters

    Steps

    1. 1

      Document import

      What already works from Figma today.

      Founding Engineer 1 day

    2. 2

      Publish the page

      Under integrations.

      Founder & CEO 1 day

    3. 3

      Build the plugin

      Send the library, pull Brief guidance.

      Founding Engineer 3 weeks

    4. 4

      Community listing

      Figma community.

      Founder & CEO 1 week

    5. 5

      Tutorial

      Library to on-brand ad set.

      Founder & CEO 3 days

    6. 6

      Measure

      Brands created from Figma.

      Founding Engineer monthly

    First result
    2 months
    Cost
    About a month of engineering
    Owner
    Founding Engineer

    Smallest test

    Publish the page and count signups from it before building the plugin.

    Why it works

    Publish a Figma integration page showing how to import styles, components and logos into a brand, then build a small plugin that sends selected frames or the library to Rainbrand and pulls Brief guidance back into the canvas. List it in the Figma community.

    Designers own the brand source of truth; meeting them in their tool brings the team that decides on brand tooling.

    Risks and how to handle them

    • Plugin maintenance load

      Start with import only.

    Evidence

    • Your websiteIntegrations: ChatGPT, Claude, Claude Code, Codex, Cursor and Higgsfield (listed as "Higgsfield MCP for on-brand video and images" with no setup steps), plus the API. There is no Figma, Canva, Webflow or automation-tool page, although the homepage lists "In Figma & more" as a source.Source
    • YC launchLaunch post: "We launched v1 in December and went viral" and "Since then we've generated over 175,000 on-brand assets for our customers." Rainbrand ingests websites, decks, Figma, socials and codebases and exposes them through an API, MCP and a forthcoming CLI; it asks to hear from builders of marketing, coding and e-commerce agents, products that need brand context, and agencies and marketing teams.Source
  • 16

    Two of the four customer stories are products that build Rainbrand in: one has created hundreds of brands through it, another made it the default image solution in its product. Each of those products shows Rainbrand's work to its own customers, who never learn where the brand layer came from.

    High impactLow effort First result in 1 quarter Founder & CEO

    First action

    Offer the badge and credit deal to the two embed customers.

    How to measure it

    Partner-sourced signups

    Now
    –
    Target
    50 a month
    By
    2 quarters

    Steps

    1. 1

      Design the badge

      Small and optional.

      Founder & CEO 1 day

    2. 2

      Set the deal

      Credits or a rate cut for showing it.

      Founder & CEO 1 day

    3. 3

      Ask embedders

      Starting with the published customers.

      Founder & CEO 2 weeks

    4. 4

      Directory page

      Products with a Rainbrand brand layer.

      Founding Engineer 2 days

    5. 5

      Co-launch posts

      One per new embedder.

      Founder & CEO ongoing

    6. 6

      Track

      Referral links per partner.

      Founding Engineer monthly

    First result
    1 quarter
    Cost
    Credits given to partners
    Owner
    Founder & CEO

    Smallest test

    Ask one customer to add a settings-page link for a month.

    Why it works

    Offer embedders an optional "Brand by Rainbrand" badge and a link in their brand settings, in exchange for extra credits or a lower volume rate. List them in a public directory of products with a brand layer, and co-write a launch post with each new embedder.

    Every embedder's customers are agent and marketing users, the same buyers Rainbrand wants, and a badge reaches them at the moment they see their brand used correctly.

    Risks and how to handle them

    • Embedders prefer white-label

      Keep it optional and reward it instead.

    Builds on

    Evidence

    • Your websiteCustomer stories: four (Revnu, Letter, The GTM Co-Founder, Autumn), two "Built into their product" and two "In their agents", none with results metrics. Revnu (customer since April 2026) uses the REST API and has created hundreds of brands through Rainbrand; Letter (since June 2026): "We pushed Rainbrand to production as our default image solution."Source
    • YC launchLaunch post: "We launched v1 in December and went viral" and "Since then we've generated over 175,000 on-brand assets for our customers." Rainbrand ingests websites, decks, Figma, socials and codebases and exposes them through an API, MCP and a forthcoming CLI; it asks to hear from builders of marketing, coding and e-commerce agents, products that need brand context, and agencies and marketing teams.Source
  • 17

    Rainbrand already sells credits in 5,000-credit top-ups ($25), and the signup grant is 1,000 credits. Neither the pricing page nor the FAQ mentions any referral offer, although agencies and builders who share brands across tools are natural referrers.

    Medium impactLow effort First result in 1 month Founder & CEO

    First action

    Set the reward and fraud rules.

    How to measure it

    Referred paid teams

    Now
    –
    Target
    10% of new paid teams
    By
    2 quarters

    Steps

    1. 1

      Set rules

      Reward on first payment, limits per account.

      Founder & CEO 1 day

    2. 2

      Build links

      Per-user referral codes.

      Founding Engineer 3 days

    3. 3

      Place prompts

      After a passing Review and in sharing.

      Founding Engineer 1 day

    4. 4

      Announce

      Email to current users.

      Founder & CEO 1 day

    5. 5

      Pay out

      Automatic credit grants.

      Founding Engineer 1 day

    6. 6

      Measure

      Referred signups and paid teams.

      Founder & CEO monthly

    First result
    1 month
    Cost
    Credits only
    Owner
    Founder & CEO

    Smallest test

    Offer it by email to the 50 most active users.

    Why it works

    Give each user a referral link: the new team gets an extra 1,000 credits on signup, and the referrer gets 5,000 credits (one top-up) when the team pays for its first month. Show the link after a passing Review and in the brand-share flow.

    Credits cost Rainbrand less than their price, and they reward the users who already get value from the product.

    Risks and how to handle them

    • Self-referral abuse

      Pay only on a first paid month from a new payment method.

    Evidence

    • Your websitePricing: Plus $20/mo (5,000 credits, 1 seat), Pro $35/mo (10,000 credits, unlimited seats), Max $50/mo (20,000), Scale $65/mo (30,000), a volume tier "From $360/mo" and "Custom" through "Talk to us" by email. A new brand costs 500 credits, a brief 100, a 2K image 100, a review 200 and a brand edit 300. New sign-ups get 1,000 credits and one free brand creation, "a one-time signup grant, not a monthly allowance"; there is no free tier. If credits run out mid-task the request stops and links to the pricing page. Annual billing saves 10%.Source
    • Your websiteFAQ (21 questions): credits, agencies ("Keep each client as a separate brand in your Rainbrand team, with its own guidelines, brand voice and assets"; no cap on brands, but adding one uses credits), MCP vs REST, and why not custom instructions or a single assistant's project. No question covers security, data handling or compliance; help is by email to [email protected].Source
  • 18

    Rainbrand already ships a plugin, a connector, coding-agent plugins and an MCP server, and its YC launch targets builders of marketing, coding and e-commerce agents. Those builders find tools in MCP registries and plugin directories. Competitors are there too: Frontify and Context.dev both advertise MCP servers.

    Medium impactLow effort First result in 1 month Founding Engineer

    First action

    List the registries and directories and check which already list Rainbrand.

    How to measure it

    Directory-sourced signups

    Now
    –
    Target
    100 a month
    By
    90 days

    Steps

    1. 1

      Audit listings

      Where Rainbrand appears today.

      Founding Engineer 1 day

    2. 2

      Write the listing

      One demo prompt with a library brand.

      Founder & CEO 1 day

    3. 3

      Public demo brand

      Read-only library brands without sign-in.

      Founding Engineer 1 week

    4. 4

      Submit

      Registries and directories.

      Founding Engineer 2 days

    5. 5

      Tag sources

      A source parameter per listing.

      Founding Engineer 1 day

    6. 6

      Keep current

      Update with each release.

      Founding Engineer monthly

    First result
    1 month
    Cost
    $0
    Owner
    Founding Engineer

    Smallest test

    Submit to two registries with source tags.

    Why it works

    Submit the MCP server and plugins to the main public MCP registries and agent plugin directories, with a description that leads on one concrete demo ("ask for the Adidas brand and get an on-brand ad set") using a public library brand that works before signup. Keep listings current with each release.

    Directories are where agent builders shop, and a demo that works on a famous brand shows value in one prompt.

    Risks and how to handle them

    • Demo brands used heavily without signup

      Rate-limit and allow briefs only, not generation.

    Evidence

    • Your websiteIntegrations: ChatGPT, Claude, Claude Code, Codex, Cursor and Higgsfield (listed as "Higgsfield MCP for on-brand video and images" with no setup steps), plus the API. There is no Figma, Canva, Webflow or automation-tool page, although the homepage lists "In Figma & more" as a source.Source
    • YC launchLaunch post: "We launched v1 in December and went viral" and "Since then we've generated over 175,000 on-brand assets for our customers." Rainbrand ingests websites, decks, Figma, socials and codebases and exposes them through an API, MCP and a forthcoming CLI; it asks to hear from builders of marketing, coding and e-commerce agents, products that need brand context, and agencies and marketing teams.Source
    • Competitor researchBrandfetch: "The brand data platform" with "50M+ brands", a Logo API "free up to 1M monthly requests" and a Brand Context API. Frontify: "Brand intelligence for the AI era", reaching "every AI agent via MCP and API", with enterprise logos.Source
    • Competitor researchContext.dev (brand.dev now redirects there): "The unified web context API for software and AI agents", with brand data and styleguides, a free plan of 1,000 credits a month with no card, a $20 Developer plan, an MCP server and SOC 2 Type 2.Source

Big bets 3

Larger plays that can change the trajectory.

  • 19

    Rainbrand already ships Brand Skills as downloadable ZIPs (a skill file, a README, reference files per topic, images and logos) for 200 public brands, and its skills and plugins are on GitHub. Brandfetch and Context.dev offer brand data from a lookup API, and Frontify keeps guidelines inside its own portal. Nobody owns the file format agents use to carry a brand.

    High impactHigh effort First result in 1 quarter Founding Engineer

    First action

    Write version 0.1 of the specification from the current ZIP layout.

    How to measure it

    Tools supporting the format

    Now
    –
    Target
    10
    By
    3 quarters

    Steps

    1. 1

      Write the spec

      Files, fields and versioning.

      Founding Engineer 2 weeks

    2. 2

      Validator

      Open-source lint tool.

      Founding Engineer 2 weeks

    3. 3

      Open the library

      200 skills in a public repository with a clear licence.

      Founder & CEO 1 week

    4. 4

      Lint page

      Free check on the website.

      Founding Engineer 1 week

    5. 5

      Recruit adopters

      Agent builders and design tools.

      Founder & CEO 1 quarter

    6. 6

      Govern

      Public change process.

      Founder & CEO ongoing

    First result
    1 quarter
    Cost
    A quarter of engineering
    Owner
    Founding Engineer

    Smallest test

    Publish the spec draft and ask five builders for feedback.

    Why it works

    Publish a versioned Brand Skill specification and validator, open-source the 200 library skills under a clear licence, and offer a free "lint your brand skill" check. Rainbrand stays the place where skills are built, maintained, versioned and reviewed, and the format spreads through every tool that reads it.

    If the format is the standard, every skill made anywhere is a step towards Rainbrand's paid maintenance and Review, and competitors have to read it.

    Risks and how to handle them

    • Competitors adopt the format and compete on it

      Compete on maintenance, versioning and Review, not the file.

    • Trademark issues with public brand skills

      Keep disclaimers and honour takedowns.

    Evidence

    • Your websiteBrand guidelines library: 201 URLs under /brand-guidelines/ in the sitemap (latest November 4, 2026), headed "Brand guidelines examples" and covering "200 brands, with more coming". Each page, such as "Adidas brand guidelines: logo, colors and fonts", is built from the brand's own website, offers a downloadable Brand Skill ZIP, and says it "isn't Adidas's official guide" and that Rainbrand "isn't affiliated with or endorsed by adidas AG".Source
    • Competitor researchBrandfetch: "The brand data platform" with "50M+ brands", a Logo API "free up to 1M monthly requests" and a Brand Context API. Frontify: "Brand intelligence for the AI era", reaching "every AI agent via MCP and API", with enterprise logos.Source
    • Competitor researchContext.dev (brand.dev now redirects there): "The unified web context API for software and AI agents", with brand data and styleguides, a free plan of 1,000 credits a month with no card, a $20 Developer plan, an MCP server and SOC 2 Type 2.Source
  • 20

    Rainbrand has the two things a credible benchmark needs: 200 structured public brands and an automated Review that checks logo, colours, typography and copy and returns pass, fail or uncertain. Its own Eager example shows Review catching an unapproved colour and the wrong typeface. Nobody publishes how often generation tools get a brand wrong, with or without brand context.

    High impactHigh effort First result in 1 quarter Founder & CEO

    First action

    Run a pilot on 20 brands and two tools.

    How to measure it

    Benchmark-sourced signups

    Now
    –
    Target
    500 in the first quarter
    By
    2 quarters

    Steps

    1. 1

      Design the method

      Brands, briefs, tools and scoring.

      Founding Engineer 1 week

    2. 2

      Pilot

      20 brands and two tools.

      Founding Engineer 1 week

    3. 3

      Human check

      Audit a sample of Review verdicts.

      Founder & CEO 1 week

    4. 4

      Full run

      100 brands and the main tools.

      Founding Engineer 2 weeks

    5. 5

      Publish

      Index, method and failure gallery.

      Founder & CEO 1 week

    6. 6

      Pitch

      Design and AI press, and the tool makers.

      Founder & CEO 2 weeks

    7. 7

      Refresh

      Quarterly.

      Founding Engineer quarterly

    First result
    1 quarter
    Cost
    Generation and Review costs plus staff time
    Owner
    Founder & CEO

    Smallest test

    Publish the 20-brand pilot as a blog post.

    Why it works

    Generate the same brief for a sample of library brands across the leading image tools, with and without a Rainbrand Brief, score every output with Review, and publish an "On-brand index" with the method, examples of failures and the lift from brand context. Refresh quarterly as tools change.

    Original data earns links, press and AI-search citations, and the headline result (brand context improves pass rates) is the product's pitch, measured.

    Risks and how to handle them

    • Seen as biased

      Publish the method and data and invite reruns.

    • Brand owners object

      Use mock briefs only and keep disclaimers.

    Evidence

    • Your websiteBrand guidelines library: 201 URLs under /brand-guidelines/ in the sitemap (latest November 4, 2026), headed "Brand guidelines examples" and covering "200 brands, with more coming". Each page, such as "Adidas brand guidelines: logo, colors and fonts", is built from the brand's own website, offers a downloadable Brand Skill ZIP, and says it "isn't Adidas's official guide" and that Rainbrand "isn't affiliated with or endorsed by adidas AG".Source
    • Your websiteExamples: nine brand mock-ups (Eager, Kaorium, Son of a Tailor, Yumbun, Gut Oggau, Friedhats, Extropic, Heart Aerospace, Sazabi). In the Eager example one prompt produced a landing page, an email, a carousel, a 12-second reel and a 20-second film; Review caught a headline in an unapproved purple and lettering in the wrong typeface, but the reel and film were not reviewed because Review only checks still images.Source
    • Your websiteHomepage: "The brand layer for agents"; sources In Files, In Website and In Figma & more; outputs MCP ("Use it with your agents") and REST API ("Embed it into your product"); Briefs, Generation through an AI Gateway, and Reviews that return pass, fail, uncertain or not applicable; testimonials from Revnu, Letter and The GTM Co-Founder, and no usage figures in the page text.Source
  • 21

    The YC launch promises a forthcoming CLI, and Rainbrand already has coding-agent plugins for on-brand frontend design. Coding agents now ship UI changes daily, and nobody checks them against the brand. Review already returns pass, fail or uncertain with evidence for images.

    High impactHigh effort First result in 2 quarters Founding Engineer

    First action

    Prototype the action on Rainbrand's own site repository.

    How to measure it

    Active repositories

    Now
    –
    Target
    200
    By
    3 quarters

    Steps

    1. 1

      CLI core

      Auth, brand selection, brief and review.

      Founding Engineer 3 weeks

    2. 2

      CI action

      Screenshot changed pages.

      Founding Engineer 2 weeks

    3. 3

      PR comment

      Verdict, reason and evidence.

      Founding Engineer 1 week

    4. 4

      Dogfood

      On Rainbrand's own site.

      Founding Engineer 2 weeks

    5. 5

      Launch

      Docs, a demo repository and a launch post.

      Founder & CEO 1 week

    6. 6

      Price

      Reviews per merge in credits.

      Founder & CEO 1 day

    First result
    2 quarters
    Cost
    A quarter of engineering
    Owner
    Founding Engineer

    Smallest test

    Run the check by hand on ten pull requests and share the comments with two customers.

    Why it works

    Ship the CLI and a CI action that screenshots changed pages in a pull request, runs Review against the team's brand, and posts a short comment with the verdict, the reason and the frames. Teams that adopt it pay for Reviews on every merge, which is recurring usage tied to engineering work.

    It puts Rainbrand into the developer workflow where coding agents already work, and makes brand a merge check rather than a design review.

    Risks and how to handle them

    • Noisy checks block merges

      Comment only by default, never block.

    Evidence

    • YC launchLaunch post: "We launched v1 in December and went viral" and "Since then we've generated over 175,000 on-brand assets for our customers." Rainbrand ingests websites, decks, Figma, socials and codebases and exposes them through an API, MCP and a forthcoming CLI; it asks to hear from builders of marketing, coding and e-commerce agents, products that need brand context, and agencies and marketing teams.Source
    • Your websiteIntegrations: ChatGPT, Claude, Claude Code, Codex, Cursor and Higgsfield (listed as "Higgsfield MCP for on-brand video and images" with no setup steps), plus the API. There is no Figma, Canva, Webflow or automation-tool page, although the homepage lists "In Figma & more" as a source.Source
    • Your websiteHomepage: "The brand layer for agents"; sources In Files, In Website and In Figma & more; outputs MCP ("Use it with your agents") and REST API ("Embed it into your product"); Briefs, Generation through an AI Gateway, and Reviews that return pass, fail, uncertain or not applicable; testimonials from Revnu, Letter and The GTM Co-Founder, and no usage figures in the page text.Source

Wildcards 2

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

  • 22

    Bloom became Rainbrand on October 15, 2026. The product launched in December, "went viral" and had generated over 175,000 assets by its YC launch, so many early users signed up under the Bloom name for one-off images. The FAQ reassures them that keys and plans carry over, but the changelog has no entry since the rename explaining what is new.

    Medium impactLow effort First result in 2 weeks Founder & CEO

    First action

    Count Bloom-era accounts inactive for 60 days.

    How to measure it

    Bloom-era accounts active again

    Now
    –
    Target
    8% of the segment
    By
    60 days

    Steps

    1. 1

      Segment

      Accounts from before the rename, inactive for 60 days.

      Founding Engineer 1 day

    2. 2

      Write the email

      Name, what is new, their brand.

      Founder & CEO 1 day

    3. 3

      Credit offer

      Enough for one Review.

      Founder & CEO 1 hour

    4. 4

      Send in waves

      Test subject lines on 10% first.

      Founder & CEO 1 week

    5. 5

      Changelog entry

      The same summary, public.

      Founder & CEO 1 hour

    6. 6

      Measure

      Return visits, Reviews and upgrades.

      Founding Engineer 1 month

    First result
    2 weeks
    Cost
    Credits only
    Owner
    Founder & CEO

    Smallest test

    Send to 200 dormant accounts first.

    Why it works

    Segment accounts created under Bloom that have not used the product in 60 days and send one plain email: the new name, what is new since they left (Briefs, Reviews, connectors for their assistants, the brand library), and their brand already waiting. Offer a small credit grant to try one Review. Log the same summary in the changelog.

    Dormant users who already made a brand are the cheapest reactivation pool, and the rename is a natural reason to write.

    Risks and how to handle them

    • Spam complaints

      Send only to users who agreed to product email, with easy opt-out.

    Evidence

    • Your websiteChangelog: "Bloom is now Rainbrand" (October 15, 2026) is the only entry in the last 90 days. The FAQ says existing Bloom API keys still work and Bloom plans keep their pricing; trybloom.ai 301-redirects to the rainbrand.com homepage.Source
    • YC launchLaunch post: "We launched v1 in December and went viral" and "Since then we've generated over 175,000 on-brand assets for our customers." Rainbrand ingests websites, decks, Figma, socials and codebases and exposes them through an API, MCP and a forthcoming CLI; it asks to hear from builders of marketing, coding and e-commerce agents, products that need brand context, and agencies and marketing teams.Source
  • 23

    The docs and API reference are updated almost daily, yet the public changelog has one entry in 90 days, the blog index shows no dates and has no newsletter signup, and one post title reads "(2027)" in November 2026. A fast-moving product looks slower than it is.

    Low impactLow effort First result in 1 month Founder & CEO

    First action

    Backfill the changelog with the last month's releases.

    How to measure it

    Subscribers

    Now
    –
    Target
    1,000
    By
    2 quarters

    Steps

    1. 1

      Backfill

      Last month's releases.

      Founder & CEO 2 hours

    2. 2

      Signup box

      On the blog and changelog.

      Founding Engineer 2 hours

    3. 3

      Dates

      On the blog index; fix future-dated titles.

      Founding Engineer 1 hour

    4. 4

      First issue

      Changes, a tutorial and an example.

      Founder & CEO 1 day

    5. 5

      Routine

      An entry per release, an email each month.

      Founder & CEO monthly

    6. 6

      Measure

      Subscribers and return visits.

      Founder & CEO monthly

    First result
    1 month
    Cost
    $0
    Owner
    Founder & CEO

    Smallest test

    Backfill the changelog this week.

    Why it works

    Publish a short changelog entry for every user-visible change and a monthly email that bundles them with one tutorial and one customer example. Add a signup box to the blog and changelog, show dates on the blog index, and date list posts by the year they were checked.

    Visible momentum reassures buyers betting on a small team, and a monthly email brings existing users back to new features.

    Risks and how to handle them

    • The routine slips

      Generate the draft from release notes each month.

    Evidence

    • Your websiteChangelog: "Bloom is now Rainbrand" (October 15, 2026) is the only entry in the last 90 days. The FAQ says existing Bloom API keys still work and Bloom plans keep their pricing; trybloom.ai 301-redirects to the rainbrand.com homepage.Source
    • Your websiteBlog: 52 URLs in the sitemap and 42 articles in 8 topics (including Agencies, Comparisons, Review and Developers); the index shows no dates or authors and has no newsletter signup; one title reads "11 brand management tools for AI agents (2027)".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.

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.

  • Rudus growth strategyAI takeoff and estimating software for concrete subcontractors. It reads structural and site drawings, finds footings, walls, columns and…23 growth ideas
  • Osynth growth strategyCreates and publishes AI-generated short-form videos for brands on TikTok, Instagram Reels and YouTube Shorts: studies the niche, creates…21 growth ideas
  • Antithesis growth strategyAn autonomous testing platform that runs whole software systems in a deterministic simulation, injects faults, finds bugs other testing…22 growth ideas
  • Capy growth strategyAn AI software engineer that runs fleets of parallel coding agents on cloud machines or paired devices: it takes tasks from Slack, Linear…21 growth ideas
  • Proximitty growth strategyAI agents for commercial lending: document ingestion, financial spreading and credit memos, covenant monitoring and loan servicing, plus…22 growth ideas
  • Charm growth strategyAn ecommerce data intelligence platform tracking DTC brands and daily TikTok Shop sales in 15 countries: revenue estimates, growth and…22 growth ideas
  • Yext growth strategyAn enterprise agentic marketing platform: Scout (AI search visibility and competitive intelligence), Listings, Knowledge Graph, Reviews…22 growth ideas
  • Hanover Park growth strategyAn AI-native fund administrator: a proprietary general ledger, fund administration with expert fund accountants, portfolio intelligence…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. Rainbrand did not commission, review or endorse it, and Programmable is not affiliated with Rainbrand. 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 Rainbrand runs its business.

Something wrong, or do you represent Rainbrand? 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

    Connect the guideline library and Brand Kit Generator to signup, make the signup grant finish a first task, add numbers to proof, ship the agency page and the custom-instructions comparison, and speed up the homepage.

  2. Days 31 to 6013 ideas

    Sell to embedders and teams: trust page, embed pricing path, low-balance handling, video Review, brand freshness, claim-your-brand, Higgsfield and Figma, a powered-by program, referrals and directory listings; plus the Bloom-user email and a monthly changelog.

  3. Days 61 to 903 ideas

    Commit to the big bets: an open Brand Skill format, a public on-brand benchmark and the CLI with a pull-request brand check.

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 one brand that every agent and tool uses, instead of re-pasting guidelines. The most cited problem: brand lives in decks, PDFs and tribal knowledge that agents cannot query.

  • Accurate, on-brand creative fetched per customer inside their own product
  • Image models do not know a brand's products, type or colours
  • In their words: “brand layer”

Market

Competitive landscape

5 competitors compared: Brandfetch, Frontify, Context.dev (formerly brand.dev), Canva Brand Kit and 1 more.

  • Gap to exploit: Maintained, versioned brand context that any agent can call
  • Brandfetch: Looks a brand up rather than maintaining it
  • Frontify: Sales-led pricing

Hiring signals

Rainbrand is a three-person YC P26 team hiring one in-person Founding Engineer in San Francisco.

  • Founding Engineer: Agents, multimodal ingestion, AI-quality evaluations and the API and MCP. Ideas 1, 11, 12, 19, 21.

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