first principles

Field notes • October 2026

How great DevTools companies get customers.

Not a theory deck. Real examples from Harness, Rippling, PostHog and seven others—followed by simple templates you can steal and adapt.

10 companies23 original sources12 copyable templates

This guide answers two questions: What did these companies actually do? And how could you use the same idea? Company examples are linked to their source. Templates are fresh examples to adapt—not claims that the company published those exact words.

A note on evidence: a company publishing a popular article or event does not prove it generated a particular amount of revenue. The examples show visible practices, not a secret formula for its first $10M.

Start here

Start with the lesson you need

Different problems need different playbooks. Pick the one closest to your current bottleneck.

See lessons by problem →

Channel playbooks

Thirteen channels, each with its own playbook

SEO, technical content, research, outbound, launches, DevRel, partnerships, events, email and paid — what companies did, how to do it, and a worksheet to copy.

Browse the channel playbooks →

Company playbooks

The ten company playbooks

Each one includes real examples, the underlying lesson, steps to try it and a template you can copy.

What they built

What they actually published or built

Harness

Comparisons, research reports, events, customer proof—and sales conversations built around outcomes.

Lesson to keep: Start with a decision the buyer has to make, not a keyword. Then connect education, proof and evaluation. A good article can be something your champion forwards to a manager.

Full breakdown →

Rippling

Systematic outbound, sharp use cases and partner channels.

Lesson to keep: Pick a very specific audience, reach it directly, and learn which message gets a real conversation. Add people and volume only after you understand conversion.

Full breakdown →

PostHog

High-intent SEO, deep tutorials, honest opinions and a website that converts.

Lesson to keep: A strong technical article answers a specific question so well that a developer shares it—even if they never buy. Then make the next step useful: runnable example, free tool or trial.

Full breakdown →

Supabase

Engineer-written launch posts, demos, coordinated distribution and repeated launches.

Lesson to keep: Shipping is not distribution. Give the release a clear story, a working demo, technical substance and an actual distribution plan.

Full breakdown →

Merge

Talk to users and buyers, show the product early and document what works.

Lesson to keep: Treat early calls as fieldwork. Your website copy, demo and outbound should come from the customer’s actual language.

Full breakdown →

Datadog

Low-friction adoption, additional teams, usage and adjacent products.

Lesson to keep: The next sale can be inside an account that already trusts you. But expansion has to follow delivered value; don’t push a second workload before the first one works.

Full breakdown →

Linear

Selective early users, excellent onboarding and multiple launch moments.

Lesson to keep: A great experience for the right ten users can teach you more than a disappointing experience for a thousand random signups.

Full breakdown →

Tailscale

Individual utility, word of mouth and a clear paid-company use case.

Lesson to keep: Make it easy to get one meaningful result. Design the natural next step for a team—not a random paywall after an arbitrary number of clicks.

Full breakdown →

incident.io

A complete practitioner guide that earns trust.

Lesson to keep: Teach the whole workflow. A practitioner should be able to put the advice into practice without a demo.

Full breakdown →

Grafana Labs

Open standards, integrations, community and a clear paid upgrade boundary.

Lesson to keep: Meet engineers in the tools and standards they already use. Compatibility can make trying something new feel safer.

Full breakdown →

Swipe file

Copy the frameworks

Twelve editable starting points, grouped by the job they do.

Where to start

If you could only try three things

You don’t need ten growth motions. Here’s a small, testable start.

1. Build one page that helps someone choose.

Interview sales and customers. Publish a genuinely fair comparison or migration guide. Add a working checklist.

Use the comparison template →

2. Get it in front of the right 100 accounts.

Choose a buyer and trigger. Test a few messages. Track qualified conversations, not send volume.

Use the outreach template →

3. Put a real customer or engineer on the record.

Record a technical story with details and results. Turn it into a guide, short webinar and reference asset.

Use the customer-story template →

Then decide what to repeat.

Review the pipeline, opportunities, activated users and objections. Keep what works; drop the rest.

Use the discovery-notes template →

Rule of thumb: A good GTM asset should do at least one job: help a prospect discover you, help them make a decision, help them adopt the product, or help them expand its use. If it does none of those, ask why you’re making it.

Sources

Original reading list

Mostly first-party company writing, founder interviews and Datadog’s SEC filing.

See all 23 sources →

First Principles newsletter

New company breakdowns, when there’s something specific worth explaining.

No pop-ups, no gated downloads. Just the next useful lesson.