ProgrammableMarketing/trunk.io (opens in a new tab)
PDF

Views

  • Plan
  • Roadmap
  • Research

On this page

  • The brief
  • Start here
  • Idea portfolio
  • Effort vs impact
  • This week · 6
  • This quarter · 9
  • Big bets · 3
  • Wildcards · 3
  • How we can help
Generated Oct 18, 2026
  1. Programmable
  2. Playbooks
  3. Trunk
Trunk logo

growth playbook · Oct 18, 2026

trunk.io (opens in a new tab)

Trunk growth playbook

Diagnosis

Trunk has strong proof and a narrow, painful problem: customer stories show Brex cutting time to merge 30%, Faire saving 330 hours in 35 days and Caseware cutting merge time 75% on day one.

Its hiring signal is thin and points one way: the only posts are for a Developer Advocate, judged on developer awareness that converts to pipeline through talks, content, community and docs, and those posts were last reposted 4 and 11 months ago, with no open roles on its job board.

So this playbook leans on awareness that a small team can produce with data, and on fixing the paths from a visit to a paid plan: the site asks developers to book a demo for a product with a free plan, the pricing page contradicts itself, and only one of twelve stories quotes a named engineer.

Growth Score

6/10

Biggest Opportunity

Measure flakiness in public: CI re-run rates for open-source projects can be computed from GitHub's own data, which gives Trunk original content, a badge for READMEs and a lead list, and AI-written PRs make the problem bigger every month.

Biggest Risk

GitHub's own merge queue, Datadog and merge-queue rivals bundle "good enough" flaky-test handling, and with no open GTM roles the team may not have the capacity to make Trunk the default.

01

Start here

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

  1. 1

    Add "Start free" next to "Book a demo" for a product that already has a free plan

    Change the homepage hero buttons today and add an event for each.

    High impact First result in 2 weeks
  2. 2

    A page and post on CI for agent-written PRs, built on the Descript quote

    Ask Descript for their PR volume before and after agent PRs.

    High impact First result in 3 weeks
  3. 3

    Measure CI re-run rates on the 100 busiest public GitHub repos and publish the ranking

    Write the script that reads workflow runs for one repository and computes its re-run rate.

    High impact First result in 3 weeks

02

Your idea portfolio

21 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.

  • 6 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 · 6Fill-ins · 6Reconsider · 7
Effort →

Showing 21 of 21 ideas

This week 6

Quick wins with a clear first action.

  • 01

    The Developer Advocate role has one objective, developer awareness, and Trunk's whole pitch is that flaky tests waste CI time. GitHub exposes every public workflow run with its attempt count, so the re-run rate of any open-source project can be measured from outside, including projects run by Trunk's own logo customers such as Supabase and PostHog.

    High impactMedium effort First result in 3 weeks Senior Developer Advocate

    First action

    Write the script that reads workflow runs for one repository and computes its re-run rate.

    How to measure it

    Free Flaky Tests installs from the report

    Now
    –
    Target
    25 projects
    By
    6 weeks

    Steps

    1. 1

      Write the collector

      GitHub Actions runs and attempts for 90 days.

      Senior Developer Advocate 2 days

    2. 2

      Pick the 100 repos

      Most active by pull requests, across languages.

      Senior Developer Advocate 2 hours

    3. 3

      Compute the numbers

      Re-run rate, CI minutes lost, worst workflows.

      Senior Developer Advocate 1 day

    4. 4

      Write the method

      What counts as a re-run, and the limits.

      CTO 2 hours

    5. 5

      Publish the report

      Ranked table, charts, method and a free-install offer.

      Senior Developer Advocate 2 days

    6. 6

      Tell maintainers

      A friendly note to each top-20 project with their number.

      Senior Developer Advocate 1 day

    7. 7

      Hand leads to Sales

      Company-backed projects go to Sales with the data.

      Sales team ongoing

    First result
    3 weeks
    Cost
    $0, about a week of work
    Owner
    Senior Developer Advocate

    Smallest test

    Publish ten repositories' numbers in one blog post.

    Why it works

    Pull 90 days of GitHub Actions runs for the 100 most active public repositories, compute the share of runs that needed a re-run and the CI minutes lost, publish a ranked report with the method, and offer each maintainer a free Flaky Tests setup.

    It is original data no competitor has, it names projects developers care about (which gets shared), and every listed project is a lead: maintainers and the companies behind them see their own number.

    Risks and how to handle them

    • Projects dislike being ranked

      Lead with the method, offer help, and let maintainers add context.

    • Re-runs have other causes

      Say so in the method and show it as an estimate.

    Evidence

    • Search resultsGitHub's REST API exposes workflow runs with a run_attempt count for public repositories, so re-run rates can be measured from outside.Source
    • HiringSenior Developer Advocate: speak at developer and platform engineering conferences (about 30% travel) and own the event strategy with Sales on lead capture; own blog posts, webinars, demo videos and case studies; "Show up where platform engineers talk shop"; "Docs are a conversion lever - you're accountable for their quality"; one objective, "drive developer awareness", with top-of-funnel metrics that convert to pipeline.Source
    • Your websiteHomepage: "Keep CI Green: Eliminate flaky tests and merge bottlenecks"; Flaky Tests (automatic quarantine, PR summaries, Linear and Jira tickets, AI failure analysis) and a Merge Queue sold as an upgrade to GitHub's (anti-flake protection, batches of up to 100 PRs, parallel queues); logos include Brex, Zillow, Exa, Supabase, PostHog, Faire, Ramp, Cockroach Labs, Retool and Google X; the main call to action is "Book a demo".Source
    • Your websiteCustomer stories: Brex (P90 time to merge down 30%, half of merge-queue failures eliminated), Faire (330 hours saved in 35 days), BetterUp (CI reliability from 70% to 90%+), Caseware (75% merge-time reduction on day one), TheyDo (94% of failing CI jobs now go green), Glydways (160 PRs a week). Only the Descript story quotes a named person: "Agent-based PRs is more of a thing now. And also non-engineers opening PRs is a thing now."Source
  • 02

    Trunk has a Free plan for up to five committers with soft limits, but the homepage's commercial call to action is "Book a demo". Developers who would try it in ten minutes are asked to book a call.

    High impactLow effort First result in 2 weeks Marketing lead

    First action

    Change the homepage hero buttons today and add an event for each.

    How to measure it

    Website-driven free installs

    Now
    –
    Target
    3× the current rate
    By
    4 weeks

    Steps

    1. 1

      Add the button

      Homepage, product and comparison pages.

      Marketing lead 2 hours

    2. 2

      Shorten the install

      GitHub app to first quarantined test in minutes.

      CTO 2 days

    3. 3

      Track both paths

      Clicks on start-free and book-demo.

      Marketing lead 1 hour

    4. 4

      Follow up installs

      An automated day-2 email with a quick win.

      Marketing lead 2 hours

    5. 5

      Route big orgs

      Installs from companies over 50 engineers alert Sales.

      Sales team 1 hour

    6. 6

      Review weekly

      Installs, activations and demos.

      Marketing lead 30 minutes a week

    First result
    2 weeks
    Cost
    $0
    Owner
    Marketing lead

    Smallest test

    Add the button to the homepage only for two weeks.

    Why it works

    Put "Start free" beside "Book a demo" on the homepage, the product pages and the comparison pages, leading to a GitHub install that sets up Flaky Tests on one repository.

    Platform engineers evaluate tools by installing them. A self-serve path captures the developers the Developer Advocate is hired to reach, and the free tier's soft limits already create an upgrade moment.

    Risks and how to handle them

    • Fewer demos

      Keep "Book a demo" for large teams and route large installs to Sales.

    Evidence

    • Your websiteHomepage: "Keep CI Green: Eliminate flaky tests and merge bottlenecks"; Flaky Tests (automatic quarantine, PR summaries, Linear and Jira tickets, AI failure analysis) and a Merge Queue sold as an upgrade to GitHub's (anti-flake protection, batches of up to 100 PRs, parallel queues); logos include Brex, Zillow, Exa, Supabase, PostHog, Faire, Ramp, Cockroach Labs, Retool and Google X; the main call to action is "Book a demo".Source
    • Your websitePricing: Free (up to 5 committers), Team and Enterprise (custom); free-tier limits are soft (an upgrade banner, not a block); a Team or Enterprise trial means emailing [email protected]; on-premise is in private preview. The Free plan says "Up to 5M Test Spans per Month" while the comparison table says 1M per committer.Source
    • HiringSenior Developer Advocate: speak at developer and platform engineering conferences (about 30% travel) and own the event strategy with Sales on lead capture; own blog posts, webinars, demo videos and case studies; "Show up where platform engineers talk shop"; "Docs are a conversion lever - you're accountable for their quality"; one objective, "drive developer awareness", with top-of-funnel metrics that convert to pipeline.Source
  • 03

    The Free plan says up to 5M test spans a month while the table says 1M per committer, the Team plan shows no price, and a Team trial means emailing sales. Engineers who reach pricing are ready to decide and find contradictions.

    Medium impactLow effort First result in 2 weeks Marketing lead

    First action

    Decide the real free-tier limit and the Team price with the CEO today.

    How to measure it

    Pricing page to trial or install

    Now
    –
    Target
    +30% relative
    By
    4 weeks

    Steps

    1. 1

      Pick the limits

      One test-span rule.

      CEO 30 minutes

    2. 2

      Publish a Team price

      Per committer, monthly and yearly.

      CEO 1 hour

    3. 3

      Add a trial button

      Self-serve 14-day Team trial.

      CTO 2 days

    4. 4

      Rewrite the FAQ

      Answer committer counting and limits plainly.

      Marketing lead 1 hour

    5. 5

      Track

      Trial starts from pricing.

      Marketing lead 1 hour

    6. 6

      Review

      Compare before and after.

      Marketing lead 30 minutes

    First result
    2 weeks
    Cost
    $0
    Owner
    Marketing lead

    Smallest test

    Fix the test-span contradiction today.

    Why it works

    State one test-span limit, show the Team price or "from" price, and replace "email sales for a trial" with a self-serve 14-day Team trial button.

    Clear pricing converts developer tools; contradictions send buyers to competitors whose pricing is plain.

    Risks and how to handle them

    • Public pricing anchors enterprise deals low

      Keep Enterprise custom and show Team as "from".

    Evidence

    • Your websitePricing: Free (up to 5 committers), Team and Enterprise (custom); free-tier limits are soft (an upgrade banner, not a block); a Team or Enterprise trial means emailing [email protected]; on-premise is in private preview. The Free plan says "Up to 5M Test Spans per Month" while the comparison table says 1M per committer.Source
  • 04

    Trunk's stories have strong numbers (Faire saved 330 hours in 35 days; Caseware cut merge time 75% on day one) but only one of twelve quotes a named person. Engineers trust other engineers more than a number on a card.

    Medium impactLow effort First result in 3 weeks Sales team

    First action

    Email the champions at Faire, Brex and Caseware for a quote this week.

    How to measure it

    Customer stories with a named quote

    Now
    –
    Target
    10 of 12
    By
    4 weeks

    Steps

    1. 1

      List the champions

      One per story.

      Sales team 1 hour

    2. 2

      Ask for quotes

      One sentence, name, title.

      Sales team 1 day

    3. 3

      Update the stories

      Quote, name and photo on each.

      Senior Developer Advocate 1 day

    4. 4

      Rotate on the homepage

      Show the strongest three.

      Marketing lead 1 hour

    5. 5

      Use in outreach

      Link the matching story in emails.

      Sales team ongoing

    6. 6

      Measure

      Demo requests from story pages.

      Marketing lead monthly

    First result
    3 weeks
    Cost
    $0
    Owner
    Sales team

    Smallest test

    Add named quotes to three stories.

    Why it works

    Ask the champion at each story's company for one quote and permission to use their name, title and photo, and add it to the story and homepage testimonials.

    The Developer Advocate owns case studies and Sales needs proof; named engineers make the numbers believable and give each story a person to share it.

    Risks and how to handle them

    • Champions cannot be named

      Use title and company size instead.

    Evidence

    • Your websiteCustomer stories: Brex (P90 time to merge down 30%, half of merge-queue failures eliminated), Faire (330 hours saved in 35 days), BetterUp (CI reliability from 70% to 90%+), Caseware (75% merge-time reduction on day one), TheyDo (94% of failing CI jobs now go green), Glydways (160 PRs a week). Only the Descript story quotes a named person: "Agent-based PRs is more of a thing now. And also non-engineers opening PRs is a thing now."Source
    • HiringSenior Developer Advocate: speak at developer and platform engineering conferences (about 30% travel) and own the event strategy with Sales on lead capture; own blog posts, webinars, demo videos and case studies; "Show up where platform engineers talk shop"; "Docs are a conversion lever - you're accountable for their quality"; one objective, "drive developer awareness", with top-of-funnel metrics that convert to pipeline.Source
  • 05

    Descript's engineer says agent-based PRs and PRs from non-engineers are now common. More PRs means more merges, more CI runs and more flaky failures, which is exactly what Trunk's merge queue and quarantine handle.

    High impactLow effort First result in 3 weeks CEO

    First action

    Ask Descript for their PR volume before and after agent PRs.

    How to measure it

    Demos and installs from the agent-PR page

    Now
    –
    Target
    40
    By
    6 weeks

    Steps

    1. 1

      Get the numbers

      PR volume and CI runs from one or two customers.

      Sales team 1 week

    2. 2

      Write the post

      The change, the cost, the fix.

      CEO 1 day

    3. 3

      Build the page

      Positioning plus install and demo links.

      Marketing lead 1 day

    4. 4

      Launch

      Hacker News, LinkedIn, developer newsletters.

      Senior Developer Advocate 1 day

    5. 5

      Talk track

      One slide for Sales.

      Sales team 1 hour

    6. 6

      Measure

      Visits and conversions.

      Marketing lead weekly

    First result
    3 weeks
    Cost
    $0
    Owner
    CEO

    Smallest test

    Publish the post with Descript's quote alone.

    Why it works

    Publish "CI when agents open the PRs": the volume change, why flaky tests hurt more, and how batching and quarantine absorb it, with Descript's story and an install link.

    It ties Trunk to the biggest change in how code gets written and gives buyers a new reason to fund CI reliability.

    Risks and how to handle them

    • Seen as AI hype

      Lead with numbers from real customers.

    Evidence

    • Your websiteCustomer stories: Brex (P90 time to merge down 30%, half of merge-queue failures eliminated), Faire (330 hours saved in 35 days), BetterUp (CI reliability from 70% to 90%+), Caseware (75% merge-time reduction on day one), TheyDo (94% of failing CI jobs now go green), Glydways (160 PRs a week). Only the Descript story quotes a named person: "Agent-based PRs is more of a thing now. And also non-engineers opening PRs is a thing now."Source
    • Your websiteHomepage: "Keep CI Green: Eliminate flaky tests and merge bottlenecks"; Flaky Tests (automatic quarantine, PR summaries, Linear and Jira tickets, AI failure analysis) and a Merge Queue sold as an upgrade to GitHub's (anti-flake protection, batches of up to 100 PRs, parallel queues); logos include Brex, Zillow, Exa, Supabase, PostHog, Faire, Ramp, Cockroach Labs, Retool and Google X; the main call to action is "Book a demo".Source
    • HiringSenior Developer Advocate: speak at developer and platform engineering conferences (about 30% travel) and own the event strategy with Sales on lead capture; own blog posts, webinars, demo videos and case studies; "Show up where platform engineers talk shop"; "Docs are a conversion lever - you're accountable for their quality"; one objective, "drive developer awareness", with top-of-funnel metrics that convert to pipeline.Source
  • 06

    Trunk sells to platform and security-minded engineers and holds SOC 2 Type 2 on paid plans, but its marketing site scores F on security headers. Buyers who run a quick check notice.

    Low impactLow effort First result in 1 week CTO

    First action

    Run the site through Mozilla Observatory and list the missing headers.

    How to measure it

    Mozilla Observatory grade

    Now
    –
    Target
    A
    By
    1 week

    Steps

    1. 1

      Scan

      List missing headers.

      CTO 30 minutes

    2. 2

      Add headers

      CSP, HSTS, frame and referrer policies.

      CTO 1 day

    3. 3

      Test the site

      Check nothing breaks.

      CTO 2 hours

    4. 4

      Rescan

      Confirm the grade.

      CTO 15 minutes

    5. 5

      Mention it

      Trust page lists the controls.

      Marketing lead 30 minutes

    6. 6

      Monitor

      Monthly rescan.

      CTO 15 minutes a month

    First result
    1 week
    Cost
    $0
    Owner
    CTO

    Smallest test

    Add HSTS and frame headers first.

    Why it works

    Add a content security policy, HSTS and the other standard headers to the marketing site and recheck it.

    It removes a small credibility gap in front of exactly the audience that looks.

    Risks and how to handle them

    • A strict policy breaks scripts

      Start in report-only mode.

    Evidence

    • programmable.infoSite scan: 13 "Trunk vs" pages (Buildkite, BrowserStack, BuildPulse, CircleCI, Codecov, Gradle, LambdaTest, Qase, Rainforest QA, Sauce Labs, Testim, Testmo, Testomatio) plus Aviator, GitHub Merge Queue and Mergify pages; 21 blog posts, 21 learn pages, 20 testing pages and 12 customer stories; 40 Hacker News mentions; security-header grade F; about 620 GitHub stars across repositories.Source
    • Your websitePricing: Free (up to 5 committers), Team and Enterprise (custom); free-tier limits are soft (an upgrade banner, not a block); a Team or Enterprise trial means emailing [email protected]; on-premise is in private preview. The Free plan says "Up to 5M Test Spans per Month" while the comparison table says 1M per committer.Source

This quarter 9

The core plays for the next three months.

  • 07

    The Developer Advocate owns content end to end and is asked to show up where platform engineers talk shop. Concrete teardowns of real flaky tests (race conditions, time zones, test order) are the content these engineers share.

    Medium impactLow effort First result in 1 month Senior Developer Advocate

    First action

    Pick four teardowns from the public-repo report.

    How to measure it

    Newsletter subscribers

    Now
    –
    Target
    1,000
    By
    end of quarter

    Steps

    1. 1

      Collect cases

      From open source and customers.

      Senior Developer Advocate ongoing

    2. 2

      Write the format

      Symptom, cause, fix, detection.

      Senior Developer Advocate 1 day

    3. 3

      Publish weekly

      Post and short video.

      Senior Developer Advocate 1 day a week

    4. 4

      Share in communities

      Where platform engineers gather.

      Senior Developer Advocate 1 hour a week

    5. 5

      Build a newsletter

      Weekly email of the teardown.

      Marketing lead 1 day

    6. 6

      Measure

      Subscribers and installs.

      Marketing lead weekly

    First result
    1 month
    Cost
    $0
    Owner
    Senior Developer Advocate

    Smallest test

    Publish four teardowns and count shares.

    Why it works

    Each week, one short post and video on a real flaky test (from open source or a customer with permission): the symptom, the root cause, the fix, and how Trunk flagged it.

    It builds an audience on a recurring format and ranks for the specific error messages engineers search.

    Risks and how to handle them

    • Customer code exposure

      Use open source or anonymised snippets with permission.

    Builds on

    Evidence

    • HiringSenior Developer Advocate: speak at developer and platform engineering conferences (about 30% travel) and own the event strategy with Sales on lead capture; own blog posts, webinars, demo videos and case studies; "Show up where platform engineers talk shop"; "Docs are a conversion lever - you're accountable for their quality"; one objective, "drive developer awareness", with top-of-funnel metrics that convert to pipeline.Source
    • Your websiteHomepage: "Keep CI Green: Eliminate flaky tests and merge bottlenecks"; Flaky Tests (automatic quarantine, PR summaries, Linear and Jira tickets, AI failure analysis) and a Merge Queue sold as an upgrade to GitHub's (anti-flake protection, batches of up to 100 PRs, parallel queues); logos include Brex, Zillow, Exa, Supabase, PostHog, Faire, Ramp, Cockroach Labs, Retool and Google X; the main call to action is "Book a demo".Source
  • 08

    Trunk has 16 comparison pages, mostly against test-management tools, but not against the newer merge queues and test-reliability features engineers now consider.

    Medium impactLow effort First result in 6 weeks Marketing lead

    First action

    Check search demand for each and write the Graphite page first.

    How to measure it

    Conversions from comparison pages

    Now
    –
    Target
    +50%
    By
    end of quarter

    Steps

    1. 1

      Check demand

      Search volume per comparison.

      Marketing lead 2 hours

    2. 2

      Write three pages

      Factual, with a feature table.

      Marketing lead 1 week

    3. 3

      Technical review

      Engineers check accuracy.

      CTO 2 hours

    4. 4

      Build the hub

      Grouped by problem.

      Marketing lead 1 day

    5. 5

      Add start-free

      On every comparison page.

      Marketing lead 1 hour

    6. 6

      Measure

      Visits and conversions.

      Marketing lead monthly

    First result
    6 weeks
    Cost
    $0
    Owner
    Marketing lead

    Smallest test

    Publish the Graphite page.

    Why it works

    Add factual pages for Graphite, GitLab merge trains and Buildkite Test Engine, plus a hub that groups all comparisons by problem (flaky tests or merge bottlenecks).

    Comparison searches come from buyers near a decision; a hub also helps search engines understand the set.

    Risks and how to handle them

    • Pages go stale

      Date each and review quarterly.

    Evidence

    • programmable.infoSite scan: 13 "Trunk vs" pages (Buildkite, BrowserStack, BuildPulse, CircleCI, Codecov, Gradle, LambdaTest, Qase, Rainforest QA, Sauce Labs, Testim, Testmo, Testomatio) plus Aviator, GitHub Merge Queue and Mergify pages; 21 blog posts, 21 learn pages, 20 testing pages and 12 customer stories; 40 Hacker News mentions; security-header grade F; about 620 GitHub stars across repositories.Source
    • Competitor research2026 roundups place Trunk and BuildPulse among dedicated flaky-test platforms, Datadog among observability tools (noted as expensive at high test volumes) and Mergify's Test Insights as tied to its merge queue (GitHub-only on the merge side).Source
  • 09

    Customer stories already prove the size of the waste (Faire saved 330 hours in 35 days). A calculator lets a platform team put its own number in front of the person who signs the budget.

    High impactMedium effort First result in 1 month Marketing lead

    First action

    Write the formula using the public-repo re-run rates as defaults.

    How to measure it

    Installs from the calculator

    Now
    –
    Target
    100
    By
    end of quarter

    Steps

    1. 1

      Define the formula

      Inputs and assumptions, stated plainly.

      Marketing lead 1 day

    2. 2

      Build the page

      Inputs, result, share link.

      Marketing lead 1 week

    3. 3

      Add GitHub connect

      Real numbers from the team's repos.

      CTO 1 week

    4. 4

      Promote

      Blog, DevRel talks, comparison pages.

      Senior Developer Advocate 1 week

    5. 5

      Route leads

      High-waste teams to Sales.

      Sales team ongoing

    6. 6

      Measure

      Completions, installs and demos.

      Marketing lead weekly

    First result
    1 month
    Cost
    $0, about two weeks of work
    Owner
    Marketing lead

    Smallest test

    Ship the static calculator without GitHub connect.

    Why it works

    A free calculator: engineers, PRs a day, re-run rate and CI cost per minute in, hours and dollars lost per month out, with an option to connect GitHub for real numbers and a link to Trunk.

    It gives the champion a business case and gives Trunk a lead with context; the GitHub connect turns it into a free install.

    Risks and how to handle them

    • Numbers look inflated

      Show assumptions and let users change them.

    Evidence

    • Your websiteCustomer stories: Brex (P90 time to merge down 30%, half of merge-queue failures eliminated), Faire (330 hours saved in 35 days), BetterUp (CI reliability from 70% to 90%+), Caseware (75% merge-time reduction on day one), TheyDo (94% of failing CI jobs now go green), Glydways (160 PRs a week). Only the Descript story quotes a named person: "Agent-based PRs is more of a thing now. And also non-engineers opening PRs is a thing now."Source
    • Your websiteHomepage: "Keep CI Green: Eliminate flaky tests and merge bottlenecks"; Flaky Tests (automatic quarantine, PR summaries, Linear and Jira tickets, AI failure analysis) and a Merge Queue sold as an upgrade to GitHub's (anti-flake protection, batches of up to 100 PRs, parallel queues); logos include Brex, Zillow, Exa, Supabase, PostHog, Faire, Ramp, Cockroach Labs, Retool and Google X; the main call to action is "Book a demo".Source
    • Search resultsGitHub's REST API exposes workflow runs with a run_attempt count for public repositories, so re-run rates can be measured from outside.Source
  • 10

    The Developer Advocate is to speak at platform engineering conferences and own lead capture with Sales. A booth or talk that offers attendees their own CI health number is a better reason to talk than swag.

    Medium impactMedium effort First result in Next event Senior Developer Advocate

    First action

    Pick the next two platform-engineering conferences and propose a talk built on the public-repo report.

    How to measure it

    Qualified opportunities per event

    Now
    –
    Target
    10
    By
    next two events

    Steps

    1. 1

      Pick events

      Two platform-engineering conferences.

      Senior Developer Advocate 1 week

    2. 2

      Submit talks

      Built on the public-repo data.

      Senior Developer Advocate 1 week

    3. 3

      Build the kiosk

      The calculator with GitHub connect.

      Marketing lead 1 week

    4. 4

      Prepare Sales

      Two-day follow-up with the report.

      Sales team 2 hours

    5. 5

      Run the event

      Talk, booth, health checks.

      Senior Developer Advocate event days

    6. 6

      Measure

      Leads, meetings and pipeline per event.

      Sales team after each event

    First result
    Next event
    Cost
    $5,000 to $15,000 per event
    Owner
    Senior Developer Advocate

    Smallest test

    Run the health check at one meetup.

    Why it works

    At each event, attendees connect a repository or enter a few numbers and get a one-page CI health report emailed to them; Sales follows up within two days with the report.

    It turns event traffic into qualified leads with data, and it connects the awareness role to pipeline, which is how the post says the role is judged.

    Risks and how to handle them

    • Events produce weak leads

      Count only health checks run and meetings booked.

    Builds on

    Evidence

    • HiringSenior Developer Advocate: speak at developer and platform engineering conferences (about 30% travel) and own the event strategy with Sales on lead capture; own blog posts, webinars, demo videos and case studies; "Show up where platform engineers talk shop"; "Docs are a conversion lever - you're accountable for their quality"; one objective, "drive developer awareness", with top-of-funnel metrics that convert to pipeline.Source
    • Your websiteHomepage: "Keep CI Green: Eliminate flaky tests and merge bottlenecks"; Flaky Tests (automatic quarantine, PR summaries, Linear and Jira tickets, AI failure analysis) and a Merge Queue sold as an upgrade to GitHub's (anti-flake protection, batches of up to 100 PRs, parallel queues); logos include Brex, Zillow, Exa, Supabase, PostHog, Faire, Ramp, Cockroach Labs, Retool and Google X; the main call to action is "Book a demo".Source
  • 11

    Trunk positions its merge queue as an upgrade to GitHub's, and its buyers live in GitHub. Discovery and install inside GitHub removes the website from the path.

    Medium impactMedium effort First result in 6 weeks CTO

    First action

    Check the current listing (if any) and its install numbers.

    How to measure it

    Marketplace installs

    Now
    –
    Target
    150
    By
    end of quarter

    Steps

    1. 1

      Audit the listing

      Current state and installs.

      CTO 1 hour

    2. 2

      Write the listing

      Free plan, screenshots, proof.

      Marketing lead 1 day

    3. 3

      Smooth the install

      One click to first quarantined test.

      CTO 1 week

    4. 4

      Add reviews

      Ask happy users to review.

      Senior Developer Advocate 1 week

    5. 5

      Track

      Marketplace installs to paid.

      Marketing lead weekly

    6. 6

      Iterate

      Test listing copy monthly.

      Marketing lead monthly

    First result
    6 weeks
    Cost
    $0
    Owner
    CTO

    Smallest test

    Update the listing copy and screenshots.

    Why it works

    Publish (or improve) a GitHub Marketplace listing with a free plan, screenshots, the strongest customer number and a one-click install that quarantines the first flaky test.

    Developers already search the Marketplace for CI tools; a listing with a free plan converts without a sales step.

    Risks and how to handle them

    • Marketplace billing rules

      Keep billing on Trunk if Marketplace billing does not fit.

    Evidence

    • Your websiteHomepage: "Keep CI Green: Eliminate flaky tests and merge bottlenecks"; Flaky Tests (automatic quarantine, PR summaries, Linear and Jira tickets, AI failure analysis) and a Merge Queue sold as an upgrade to GitHub's (anti-flake protection, batches of up to 100 PRs, parallel queues); logos include Brex, Zillow, Exa, Supabase, PostHog, Faire, Ramp, Cockroach Labs, Retool and Google X; the main call to action is "Book a demo".Source
    • Your websitePricing: Free (up to 5 committers), Team and Enterprise (custom); free-tier limits are soft (an upgrade banner, not a block); a Team or Enterprise trial means emailing [email protected]; on-premise is in private preview. The Free plan says "Up to 5M Test Spans per Month" while the comparison table says 1M per committer.Source
  • 12

    The Free plan's limit is five committers and soft, so growing teams keep using it with only a banner. That banner is the best buying signal Trunk has.

    High impactMedium effort First result in 4 weeks Sales team

    First action

    List free orgs over five committers today.

    How to measure it

    Paid conversions from over-limit free orgs

    Now
    –
    Target
    15%
    By
    end of quarter

    Steps

    1. 1

      Find over-limit orgs

      Committers and spans per org.

      CTO 2 hours

    2. 2

      Build the alert

      Account, usage and value found.

      CTO 2 days

    3. 3

      Write the note

      Value found plus a trial offer.

      Sales team 1 hour

    4. 4

      Follow up

      Sales contacts within two days.

      Sales team ongoing

    5. 5

      Track

      Conversions by band.

      Sales team weekly

    6. 6

      Adjust

      Thresholds monthly.

      Sales team monthly

    First result
    4 weeks
    Cost
    $0
    Owner
    Sales team

    Smallest test

    Email the top ten over-limit orgs by hand.

    Why it works

    Alert Sales when a free organisation passes five committers or a test-span limit, with company, repositories, flaky tests found and hours saved, and send the admin a short note offering a Team trial.

    It converts usage into revenue at the moment the team is getting value, without blocking the developers.

    Risks and how to handle them

    • Users feel nagged

      One note, from a person, with their own numbers.

    Builds on

    Evidence

    • Your websitePricing: Free (up to 5 committers), Team and Enterprise (custom); free-tier limits are soft (an upgrade banner, not a block); a Team or Enterprise trial means emailing [email protected]; on-premise is in private preview. The Free plan says "Up to 5M Test Spans per Month" while the comparison table says 1M per committer.Source
    • Your websiteCustomer stories: Brex (P90 time to merge down 30%, half of merge-queue failures eliminated), Faire (330 hours saved in 35 days), BetterUp (CI reliability from 70% to 90%+), Caseware (75% merge-time reduction on day one), TheyDo (94% of failing CI jobs now go green), Glydways (160 PRs a week). Only the Descript story quotes a named person: "Agent-based PRs is more of a thing now. And also non-engineers opening PRs is a thing now."Source
  • 13

    The Developer Advocate post says "Docs are a conversion lever" and makes the role accountable for their quality, but there is no stated measure.

    Medium impactMedium effort First result in 6 weeks Senior Developer Advocate

    First action

    Add docs-page events and pull the top 20 docs pages by traffic.

    How to measure it

    Activation of new installs

    Now
    –
    Target
    +20 points
    By
    end of quarter

    Steps

    1. 1

      Instrument docs

      Page views to install and activation.

      Senior Developer Advocate 2 days

    2. 2

      Find drop-offs

      Five worst setup pages.

      Senior Developer Advocate 1 day

    3. 3

      Fix them

      Clearer steps, copy-paste config.

      Senior Developer Advocate 1 week

    4. 4

      Add help links

      On setup pages.

      Senior Developer Advocate 1 hour

    5. 5

      Recheck

      Drop-off after four weeks.

      Senior Developer Advocate 1 hour

    6. 6

      Repeat

      Next five pages.

      Senior Developer Advocate monthly

    First result
    6 weeks
    Cost
    $0
    Owner
    Senior Developer Advocate

    Smallest test

    Fix the single most-visited setup page.

    Why it works

    Track docs visits to installs and first quarantined test, find the five pages with the highest drop-off, fix them, and add a "stuck? talk to us" link on setup pages.

    Most developer-tool trials fail in setup; fixing the worst pages raises activation without new traffic.

    Risks and how to handle them

    • Measuring docs adds tracking

      Use privacy-friendly analytics without personal data.

    Evidence

    • HiringSenior Developer Advocate: speak at developer and platform engineering conferences (about 30% travel) and own the event strategy with Sales on lead capture; own blog posts, webinars, demo videos and case studies; "Show up where platform engineers talk shop"; "Docs are a conversion lever - you're accountable for their quality"; one objective, "drive developer awareness", with top-of-funnel metrics that convert to pipeline.Source
    • Your websitePricing: Free (up to 5 committers), Team and Enterprise (custom); free-tier limits are soft (an upgrade banner, not a block); a Team or Enterprise trial means emailing [email protected]; on-premise is in private preview. The Free plan says "Up to 5M Test Spans per Month" while the comparison table says 1M per committer.Source
  • 14

    Trunk creates and updates Linear and Jira tickets for flaky tests automatically. Linear's users are the fast-moving engineering teams in Trunk's logo list.

    Medium impactMedium effort First result in 2 months Marketing lead

    First action

    Ask Linear's partnerships team about the integration directory and a joint post.

    How to measure it

    Installs from partner channels

    Now
    –
    Target
    50
    By
    end of quarter

    Steps

    1. 1

      Pitch Linear

      Joint post and listing.

      Marketing lead 1 week

    2. 2

      Polish the integration

      Ticket templates and auto-close.

      CTO 1 week

    3. 3

      Write the post

      With a shared customer.

      Senior Developer Advocate 1 week

    4. 4

      List it

      Linear integrations directory.

      Marketing lead 1 day

    5. 5

      Repeat with Slack

      Slack App Directory listing.

      Marketing lead 2 weeks

    6. 6

      Measure

      Partner-attributed installs.

      Marketing lead monthly

    First result
    2 months
    Cost
    $0
    Owner
    Marketing lead

    Smallest test

    Publish the Linear integration guide on the Trunk blog.

    Why it works

    Propose a joint post and integration listing with Linear showing flaky tests becoming tickets and closing when fixed, plus a template for Linear workspaces.

    Partner directories and joint posts reach teams already shaped like Trunk's best customers, at almost no cost.

    Risks and how to handle them

    • Partner declines

      Publish the guide independently.

    Evidence

    • Your websiteHomepage: "Keep CI Green: Eliminate flaky tests and merge bottlenecks"; Flaky Tests (automatic quarantine, PR summaries, Linear and Jira tickets, AI failure analysis) and a Merge Queue sold as an upgrade to GitHub's (anti-flake protection, batches of up to 100 PRs, parallel queues); logos include Brex, Zillow, Exa, Supabase, PostHog, Faire, Ramp, Cockroach Labs, Retool and Google X; the main call to action is "Book a demo".Source
    • Your websiteCustomer stories: Brex (P90 time to merge down 30%, half of merge-queue failures eliminated), Faire (330 hours saved in 35 days), BetterUp (CI reliability from 70% to 90%+), Caseware (75% merge-time reduction on day one), TheyDo (94% of failing CI jobs now go green), Glydways (160 PRs a week). Only the Descript story quotes a named person: "Agent-based PRs is more of a thing now. And also non-engineers opening PRs is a thing now."Source
  • 15

    Large monorepos are where flaky tests and merge bottlenecks hurt most (Descript runs 150,000+ tests in a JS/TS monorepo). Teams search by build tool.

    Medium impactMedium effort First result in 8 weeks Senior Developer Advocate

    First action

    Write the Bazel page with a working config.

    How to measure it

    Installs from build-tool pages

    Now
    –
    Target
    60
    By
    end of quarter

    Steps

    1. 1

      Pick the tools

      Bazel, Nx, Turborepo, Gradle.

      Senior Developer Advocate 1 hour

    2. 2

      Write configs

      Working samples per tool.

      Senior Developer Advocate 1 week

    3. 3

      Write pages

      Causes, setup, settings.

      Senior Developer Advocate 1 week

    4. 4

      Technical review

      Engineers check samples.

      CTO 2 hours

    5. 5

      Share

      Tool communities and forums.

      Senior Developer Advocate 1 week

    6. 6

      Measure

      Installs per page.

      Marketing lead monthly

    First result
    8 weeks
    Cost
    $0
    Owner
    Senior Developer Advocate

    Smallest test

    Publish the Bazel page.

    Why it works

    One page per build tool: typical flaky-test causes, how to upload results, merge-queue settings for that tool, and a sample config.

    It targets the teams with the biggest problem and largest budgets, with specific setup help that generic pages lack.

    Risks and how to handle them

    • Configs break with tool updates

      Test samples in CI.

    Evidence

    • Your websiteCustomer stories: Brex (P90 time to merge down 30%, half of merge-queue failures eliminated), Faire (330 hours saved in 35 days), BetterUp (CI reliability from 70% to 90%+), Caseware (75% merge-time reduction on day one), TheyDo (94% of failing CI jobs now go green), Glydways (160 PRs a week). Only the Descript story quotes a named person: "Agent-based PRs is more of a thing now. And also non-engineers opening PRs is a thing now."Source
    • programmable.infoSite scan: 13 "Trunk vs" pages (Buildkite, BrowserStack, BuildPulse, CircleCI, Codecov, Gradle, LambdaTest, Qase, Rainforest QA, Sauce Labs, Testim, Testmo, Testomatio) plus Aviator, GitHub Merge Queue and Mergify pages; 21 blog posts, 21 learn pages, 20 testing pages and 12 customer stories; 40 Hacker News mentions; security-header grade F; about 620 GitHub stars across repositories.Source

Big bets 3

Larger plays that can change the trajectory.

  • 16

    Open-source maintainers suffer flaky CI most and pay least, but their READMEs are read by the engineers Trunk sells to. A badge showing a project's CI reliability puts Trunk in front of them constantly.

    High impactHigh effort First result in 1 quarter CEO

    First action

    Invite the top-20 projects from the report to try the badge.

    How to measure it

    Repos showing the badge

    Now
    –
    Target
    500
    By
    2 quarters

    Steps

    1. 1

      Design the badge

      Score, link, update rule.

      CTO 1 week

    2. 2

      Build public pages

      One reliability page per project.

      CTO 3 weeks

    3. 3

      Set the free plan

      Public repos only.

      CEO 1 day

    4. 4

      Recruit projects

      Start with the top 20.

      Senior Developer Advocate 1 month

    5. 5

      Launch

      Hacker News and open-source communities.

      Senior Developer Advocate 1 day

    6. 6

      Track referrals

      Badge clicks to installs.

      Marketing lead monthly

    First result
    1 quarter
    Cost
    Hosting and support for free projects
    Owner
    CEO

    Smallest test

    Offer the badge to ten projects.

    Why it works

    Offer Flaky Tests free for public repositories with a README badge ("CI reliability 97%") that links to the project's public reliability page on Trunk.

    It turns the public-repo report into a programme, gives maintainers something they want, and creates thousands of links and impressions.

    Risks and how to handle them

    • Low scores embarrass projects

      Badge opt-in, and show trends as well as scores.

    Builds on

    Evidence

    • Search resultsGitHub's REST API exposes workflow runs with a run_attempt count for public repositories, so re-run rates can be measured from outside.Source
    • Your websiteHomepage: "Keep CI Green: Eliminate flaky tests and merge bottlenecks"; Flaky Tests (automatic quarantine, PR summaries, Linear and Jira tickets, AI failure analysis) and a Merge Queue sold as an upgrade to GitHub's (anti-flake protection, batches of up to 100 PRs, parallel queues); logos include Brex, Zillow, Exa, Supabase, PostHog, Faire, Ramp, Cockroach Labs, Retool and Google X; the main call to action is "Book a demo".Source
    • programmable.infoSite scan: 13 "Trunk vs" pages (Buildkite, BrowserStack, BuildPulse, CircleCI, Codecov, Gradle, LambdaTest, Qase, Rainforest QA, Sauce Labs, Testim, Testmo, Testomatio) plus Aviator, GitHub Merge Queue and Mergify pages; 21 blog posts, 21 learn pages, 20 testing pages and 12 customer stories; 40 Hacker News mentions; security-header grade F; about 620 GitHub stars across repositories.Source
  • 17

    Trunk sees flaky-test and merge data across customers, and public repositories add a second dataset. Nobody publishes how often CI fails for reasons unrelated to code.

    High impactHigh effort First result in Launch quarter CEO

    First action

    Confirm what aggregated customer data can be used and write the five questions.

    How to measure it

    Opportunities influenced

    Now
    –
    Target
    30
    By
    6 months after launch

    Steps

    1. 1

      Agree data use

      Aggregates only, customer terms checked.

      CEO 2 weeks

    2. 2

      Set questions

      Five questions engineering leaders ask.

      CEO 1 day

    3. 3

      Analyse

      Customer and public data.

      CTO 3 weeks

    4. 4

      Write and design

      Short report, one chart per finding.

      Marketing lead 3 weeks

    5. 5

      Launch

      Press, Hacker News and a webinar.

      Senior Developer Advocate 1 week

    6. 6

      Reuse

      Talks, posts and sales decks.

      Sales team ongoing

    First result
    Launch quarter
    Cost
    $5,000 to $10,000 for design and launch
    Owner
    CEO

    Smallest test

    Publish one chart from the public-repo data first.

    Why it works

    A yearly report: re-run rates by language and test framework, time lost per engineer, merge-queue effects, and the agent-PR trend, with methodology and a free benchmark against your own repos.

    It makes Trunk the reference for the problem it solves, and press and talks reuse it all year.

    Risks and how to handle them

    • Customer data concerns

      Aggregates only, with a minimum group size.

    Builds on

    Evidence

    • Your websiteCustomer stories: Brex (P90 time to merge down 30%, half of merge-queue failures eliminated), Faire (330 hours saved in 35 days), BetterUp (CI reliability from 70% to 90%+), Caseware (75% merge-time reduction on day one), TheyDo (94% of failing CI jobs now go green), Glydways (160 PRs a week). Only the Descript story quotes a named person: "Agent-based PRs is more of a thing now. And also non-engineers opening PRs is a thing now."Source
    • Search resultsGitHub's REST API exposes workflow runs with a run_attempt count for public repositories, so re-run rates can be measured from outside.Source
    • HiringSenior Developer Advocate: speak at developer and platform engineering conferences (about 30% travel) and own the event strategy with Sales on lead capture; own blog posts, webinars, demo videos and case studies; "Show up where platform engineers talk shop"; "Docs are a conversion lever - you're accountable for their quality"; one objective, "drive developer awareness", with top-of-funnel metrics that convert to pipeline.Source
  • 18

    As agents open more PRs, they also re-run CI when tests fail. A simple API that says "this failure is a known flaky test" saves CI minutes and agent time, and puts Trunk inside every agent workflow.

    High impactHigh effort First result in 1 quarter CTO

    First action

    Prototype an MCP tool that answers "is this failing test flaky?" for one repository.

    How to measure it

    Monthly agent API calls

    Now
    –
    Target
    100,000
    By
    2 quarters after launch

    Steps

    1. 1

      Design the API

      Inputs: repo, test, failure; output: flaky or not, with confidence.

      CTO 2 weeks

    2. 2

      Build the MCP server

      One tool, documented.

      CTO 3 weeks

    3. 3

      Test with a customer

      An agent-heavy team such as Descript.

      Sales team 1 month

    4. 4

      Write the guide

      How to add it to agent workflows.

      Senior Developer Advocate 1 week

    5. 5

      Approach agent tools

      Partnerships for default integration.

      CEO 1 quarter

    6. 6

      Track usage

      Calls and saved re-runs.

      CTO monthly

    First result
    1 quarter
    Cost
    About a quarter of engineering
    Owner
    CTO

    Smallest test

    Ship the MCP tool for internal use and one customer.

    Why it works

    Expose Trunk's flaky-test knowledge through an API and an MCP server that agents can call after a CI failure, and partner with agent tools to make it a default step.

    It moves Trunk from a dashboard engineers check to a service agents call on every failure, growing usage with agent adoption.

    Risks and how to handle them

    • Wrong "flaky" answers hide real bugs

      Return confidence and history, and never auto-merge on a flake answer alone.

    Builds on

    Evidence

    • Your websiteCustomer stories: Brex (P90 time to merge down 30%, half of merge-queue failures eliminated), Faire (330 hours saved in 35 days), BetterUp (CI reliability from 70% to 90%+), Caseware (75% merge-time reduction on day one), TheyDo (94% of failing CI jobs now go green), Glydways (160 PRs a week). Only the Descript story quotes a named person: "Agent-based PRs is more of a thing now. And also non-engineers opening PRs is a thing now."Source
    • Your websiteHomepage: "Keep CI Green: Eliminate flaky tests and merge bottlenecks"; Flaky Tests (automatic quarantine, PR summaries, Linear and Jira tickets, AI failure analysis) and a Merge Queue sold as an upgrade to GitHub's (anti-flake protection, batches of up to 100 PRs, parallel queues); logos include Brex, Zillow, Exa, Supabase, PostHog, Faire, Ramp, Cockroach Labs, Retool and Google X; the main call to action is "Book a demo".Source

Wildcards 3

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

  • 19

    Trunk sells its merge queue as an upgrade to GitHub's and already has a comparison page. Teams on GitHub's queue feel the pain of serial queues and flaky failures but fear the switch.

    Medium impactMedium effort First result in 6 weeks Senior Developer Advocate

    First action

    Write the migration guide and schedule the week.

    How to measure it

    Teams migrated

    Now
    –
    Target
    20
    By
    end of migration week

    Steps

    1. 1

      Write the guide

      Step by step with config.

      Senior Developer Advocate 1 week

    2. 2

      Schedule sessions

      Five daily live sessions.

      Senior Developer Advocate 1 day

    3. 3

      Promote

      Comparison page, social, newsletter.

      Marketing lead 2 weeks

    4. 4

      Staff it

      Engineers on call for switching teams.

      CTO 1 week

    5. 5

      Offer the trial

      30 days of Team.

      CEO 1 hour

    6. 6

      Measure

      Switches and conversions after 30 days.

      Sales team 1 month

    First result
    6 weeks
    Cost
    Engineering time for one week
    Owner
    Senior Developer Advocate

    Smallest test

    Run one live session and see who switches.

    Why it works

    A week of daily live sessions and a step-by-step migration guide, with engineers on hand and a 30-day free Team trial for teams that switch that week.

    Time-limited events create urgency, and live help removes the main objection.

    Risks and how to handle them

    • Few teams can switch in one week

      Keep the guide and offer open afterwards.

    Evidence

    • Your websiteHomepage: "Keep CI Green: Eliminate flaky tests and merge bottlenecks"; Flaky Tests (automatic quarantine, PR summaries, Linear and Jira tickets, AI failure analysis) and a Merge Queue sold as an upgrade to GitHub's (anti-flake protection, batches of up to 100 PRs, parallel queues); logos include Brex, Zillow, Exa, Supabase, PostHog, Faire, Ramp, Cockroach Labs, Retool and Google X; the main call to action is "Book a demo".Source
    • programmable.infoSite scan: 13 "Trunk vs" pages (Buildkite, BrowserStack, BuildPulse, CircleCI, Codecov, Gradle, LambdaTest, Qase, Rainforest QA, Sauce Labs, Testim, Testmo, Testomatio) plus Aviator, GitHub Merge Queue and Mergify pages; 21 blog posts, 21 learn pages, 20 testing pages and 12 customer stories; 40 Hacker News mentions; security-header grade F; about 620 GitHub stars across repositories.Source
    • Your websitePricing: Free (up to 5 committers), Team and Enterprise (custom); free-tier limits are soft (an upgrade banner, not a block); a Team or Enterprise trial means emailing [email protected]; on-premise is in private preview. The Free plan says "Up to 5M Test Spans per Month" while the comparison table says 1M per committer.Source
  • 20

    The public-repo report will show which projects lose the most CI time. Small bounties for fixing specific flaky tests make Trunk the company that fixed them.

    Medium impactMedium effort First result in 2 months Senior Developer Advocate

    First action

    Pick ten flaky tests in five projects and agree the bounties with maintainers.

    How to measure it

    Flaky tests fixed through bounties

    Now
    –
    Target
    30
    By
    2 quarters

    Steps

    1. 1

      Pick tests

      From the report.

      Senior Developer Advocate 1 week

    2. 2

      Agree with maintainers

      They approve and review fixes.

      Senior Developer Advocate 2 weeks

    3. 3

      Post bounties

      On a bounty platform or GitHub issues.

      Senior Developer Advocate 1 day

    4. 4

      Verify fixes

      Trunk data shows the flake is gone.

      Senior Developer Advocate ongoing

    5. 5

      Publish teardowns

      Each fix as a post.

      Senior Developer Advocate ongoing

    6. 6

      Measure

      Fixes, installs and mentions.

      Marketing lead monthly

    First result
    2 months
    Cost
    $5,000 to $10,000 in bounties
    Owner
    Senior Developer Advocate

    Smallest test

    Post three bounties in one project.

    Why it works

    Post bounties (for example $100 to $500) for fixing named flaky tests in top projects, verified by Trunk's data, and publish each fix as a teardown.

    It creates goodwill with maintainers, a stream of teardown content, and attention from engineers at the companies behind the projects.

    Risks and how to handle them

    • Low-quality fixes

      Maintainers approve every fix before payout.

    Builds on

    Evidence

    • Search resultsGitHub's REST API exposes workflow runs with a run_attempt count for public repositories, so re-run rates can be measured from outside.Source
    • HiringSenior Developer Advocate: speak at developer and platform engineering conferences (about 30% travel) and own the event strategy with Sales on lead capture; own blog posts, webinars, demo videos and case studies; "Show up where platform engineers talk shop"; "Docs are a conversion lever - you're accountable for their quality"; one objective, "drive developer awareness", with top-of-funnel metrics that convert to pipeline.Source
  • 21

    A company posting for a developer productivity, build infrastructure or release engineer has just decided CI pain is worth a salary. That person, and their manager, are Trunk's buyers.

    Medium impactLow effort First result in 4 weeks Sales team

    First action

    Pull last month's job posts for developer productivity and build engineers at companies over 100 engineers.

    How to measure it

    Meeting rate from job-post triggers

    Now
    –
    Target
    5 per 100
    By
    6 weeks

    Steps

    1. 1

      Define the trigger

      Role titles and company size.

      Sales team 1 hour

    2. 2

      Pull weekly

      Job boards and careers pages.

      Sales team 2 hours a week

    3. 3

      Find the manager

      Engineering or platform lead.

      Sales team 1 hour a week

    4. 4

      Write the note

      The post, the calculator, a story.

      Sales team 1 hour

    5. 5

      Follow up

      Once after a week.

      Sales team 1 hour a week

    6. 6

      Review

      Meeting rate monthly.

      Sales team monthly

    First result
    4 weeks
    Cost
    $0 to $300 a month for data
    Owner
    Sales team

    Smallest test

    Contact 20 companies from one week of posts.

    Why it works

    Track those job posts weekly, find the engineering manager, and send a short note with the CI waste calculator and a relevant customer story.

    It targets intent with a precise reason to talk, and the new hire is a natural champion once they start.

    Risks and how to handle them

    • Feels like surveillance

      Reference only the public post, once.

    Builds on

    Evidence

    • Your websiteCustomer stories: Brex (P90 time to merge down 30%, half of merge-queue failures eliminated), Faire (330 hours saved in 35 days), BetterUp (CI reliability from 70% to 90%+), Caseware (75% merge-time reduction on day one), TheyDo (94% of failing CI jobs now go green), Glydways (160 PRs a week). Only the Descript story quotes a named person: "Agent-based PRs is more of a thing now. And also non-engineers opening PRs is a thing now."Source
    • HiringSenior Developer Advocate: speak at developer and platform engineering conferences (about 30% travel) and own the event strategy with Sales on lead capture; own blog posts, webinars, demo videos and case studies; "Show up where platform engineers talk shop"; "Docs are a conversion lever - you're accountable for their quality"; one objective, "drive developer awareness", with top-of-funnel metrics that convert to pipeline.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 #21 (opens programmable.info in a new tab)

New accounts: 100,000 free credits with code BETA100K

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

Fill the events in this plan

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

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

Show Trunk’s product working

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

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

More growth playbooks

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

  • Vend Park growth strategyAI-powered parking technology combined with full-service parking operations for commercial real estate, replacing the separate operator…21 growth ideas
  • Klarent growth strategyAn agentic QA platform (formerly fore ai) whose AI agents discover, write, run and self-heal tests for web, iOS and Android apps from…21 growth ideas
  • PointFive growth strategyCloud and AI efficiency software: PointFive OS finds deep waste across cloud, data and AI platforms (500+ detections) and fixes it with…21 growth ideas
  • Pixability growth strategyAn AI-driven YouTube technology and data company: managed and self-serve YouTube and CTV media, contextual and brand-suitable targeting…21 growth ideas
  • Hawthorne Residential Partners growth strategyA Greensboro-based multifamily operator and third-party property manager: 64,000+ units across 275+ properties in eight states, with…20 growth ideas
  • Nostra AI growth strategyEdge agents for ecommerce storefronts: Dash for speed (Edge Delivery), Knox for bot protection (Edge Protect), Sherlock for AI agent…22 growth ideas
  • Xenon Pharmaceuticals growth strategyA neuroscience-focused biopharmaceutical company developing ion channel medicines; its lead investigational candidate, azetukalner, is in…20 growth ideas
  • Apify growth strategyA marketplace of more than 95,000 ready-to-run tools (Actors) for web data, automation and AI agents, plus the platform, proxies and…22 growth ideas

See every playbook and website review

Independent research. Programmable wrote this playbook from public sources, as they stood on October 18, 2026, and cites them throughout. Trunk did not commission, review or endorse it, and Programmable is not affiliated with Trunk. 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 Trunk runs its business.

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

30/60/90 roadmap

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

  1. Days 1 to 308 ideas

    Publish the public-repo re-run report, open the self-serve path, fix pricing and stories, and claim the agent-PR angle.

  2. Days 31 to 6010 ideas

    Build the engines: the waste calculator, conference health checks, the Marketplace listing, over-limit routing, docs fixes, partner and monorepo pages, plus the wildcards.

  3. Days 61 to 903 ideas

    Commit to the big bets: the open-source badge, the annual report and the agent flake-check API.

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 flaky tests quarantined without re-runs. The most cited problem: red main branches and blocked merges.

  • Faster, safer merges (Brex, Caseware)
  • CI cost and engineer time lost to re-runs
  • In their words: “keep CI green”

Market

Competitive landscape

5 competitors compared: GitHub Merge Queue, Mergify, Aviator, Datadog Test Optimization and 1 more.

  • Gap to exploit: Public data on CI reliability
  • GitHub Merge Queue: Serial queue, no flake handling
  • Mergify: GitHub-only merge side

Hiring signals

Trunk's only visible go-to-market hiring is a Developer Advocate, judged on developer awareness that converts to pipeline through conference talks with Sales on lead capture, content, community and docs (ideas 1, 7…

  • Senior Developer Advocate: Talks at platform-engineering conferences, content, community and docs as a conversion lever, judged on awareness that…
  • DevRel Engineer: Earlier version of the same role (content, code samples, developer experience); the board now shows no open postings. Ideas 19, 20.

From the team that wrote this playbook

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

© 2026 AI-Powered Growth Intelligence