Welcome folks! 👋

This edition of The Product-Led Geek will take 8 minutes to read and you’ll learn:

• How GitHub transformed user behaviour data into a powerful product-led sales engine driving enterprise growth.

• The three-phase approach they used to build product qualification models that converted 2-3x better than traditional marketing leads.

• A practical playbook for creating context-rich sales signals that turn product interactions into meaningful sales conversations.

Let’s go!

TOGETHER WITH DELVE

🚀 Delve just raised a $32M Series A at a $300M valuation, led by Insight Partners! We’re redefining compliance with an AI-native platform that gets companies like Lovable and Bland audit-ready in just 15 hours.

To celebrate:

• $15K candidate referral bonus

• $2K off setup for new customers

• Free custom doormat if you repost + comment here

Thank you for your support—this is just the beginning.

Please support our sponsors!

GEEK LINKS

GEEK OUT

Product-Led Sales at GitHub: Turning User Behaviour Into Sales Intelligence

If you’ve followed this series so far, you know the story: GitHub was famous for its open-source roots, and slow to touch anything that might mess with the developer experience.

In Part 1, we saw how Thibault Imbert built a growth team by focusing on neglected “abandoned lands” - starting at the edges, proving value with data, and slowly earning trust.

In Part 2, we shared how a single honest signup flow change unlocked $1M in ARR.

But neither of those moves was the final act.

Those wins opened a new, much bigger question:

What if you could turn everything happening inside GitHub - the millions of user actions every day - into a real sales engine?

This is the playbook for how GitHub did it: how the team built out their product-led sales motion, created their PQL/PQA models, and made user behaviour the strongest driver of enterprise growth.

The Big Question: Could Self-Serve Drive Enterprise Growth?

After the early wins with growth - rebuilding neglected flows, fixing the pricing page, and making signup more transparent - the obvious next question for Thibault and team was: What if self-serve could be more than just a source of upgrades? What if it could meaningfully contribute to the enterprise sales pipeline?

Remember how the team simplified the pricing page?

When they did that they crucially added a 'start a trial' button to the Enterprise plan card:

That unlocked executive support for the next big idea:

Could self-serve become a major driver for enterprise pipeline?

But the answer required something new - a way to spot the right moments in user behaviour, at scale, and actually get them to Sales at just the right time.

Why Product-Led Sales? Why Now?

The buying world had shifted.

Developers, engineering managers, even Fortune 500 CTOs were digital natives.

They wanted to try, deploy, and activate before ever talking to a sales rep.

GitHub had a unique advantage: 95% of its traffic was organic.

Millions of users, experimenting, learning, and pushing the boundaries of the product every day.

But none of that behavioural gold was flowing back to Sales.

Meanwhile, competitors were using self-serve as a wedge.

That was the challenge: could GitHub harness all of this in-product behaviour to fuel pipeline, not just product adoption?

The exec team acknowledged this too:

Having exec sponsorship for any GTM changes or experiments is critical for them to stand any chance of ensuring success.

Phase 1: Starting Simple

The team started from first principles.

There was no out-of-the-box PLS solution.

No one in the market had a mature product-led sales product - certainly not one tuned to developer tools.

So the early days were manual and messy.

They partnered closely with sales leaders to ask: What are the real signals of buying intent? What actually matters, not just in theory, but in the day-to-day grind of trying to close deals?

Here’s are some of the simple signals of high propensity they identified first:

• Time to add seats to expand team
  The faster the team expands, the more likely the team is going fast and is highly technical and likely to adopt more

• The size of the company
  The bigger, the more likely to be willing to talk to a sales person for questions about expansion

• Feature adoption
  The more features used, the more likely the the company is interested in truly using the product

• Reaching usage limits
  Companies with high usage are more likely to buy

The initial approach was a mix of hand-built rules, custom reports, and a lot of patience.

Weekly Iteration: Feedback From the Front Lines

The real progress didn’t come from dashboards.

It came from sitting with Sales, running pilots, and gut-checking every signal.

Each week, the Integrated Marketing team would:

• Review closed deals and lost leads, tracing them back to product activity

• Ask Sales for stories - what signals matched actual readiness?

• Kill false positives (leads that looked hot in product, but cold in real life)

• Tune the criteria and try again

This wasn’t an overnight success.

But with each iteration, the signal got clearer, and Sales got more confident in what landed in their queue.

The process of iterating on a PLS model is like training a dog: reward what works, ignore what doesn’t, and never assume you’re done.

Building the Plumbing: Turning Data Into Action

Spotting signals is only half the job.

To make any of this useful, you have to get it into Salesforce.

That meant building pipelines that could take millions of data points and surface just the right ones, with just the right context, at just the right time.

If all you do is surface some usage metric, it becomes noise to sales.

You need to give them the story: ‘This team just started using this feature within x days of signing up, added three people, and hit their limit.’

Now sales knows why to reach out, and the conversation’s about helping - not just selling.

This turned every sales conversation from a cold call into a warm, relevant, and helpful discussion.

Phase 2: ML-Based

Once the basics were in place, things got more sophisticated.

Data science joined in to build propensity models - scoring users and accounts based on a much richer blend of behaviours.

This allowed the team to filter the noise and put better context in front of sales reps, right in the CRM.

Phase 3: Expansion

The final evolution?

Expand this playbook to the whole GitHub portfolio.

Product-led sales didn’t stop with just the core product - it rolled out to add-ons like GitHub Advanced Security, GitHub Copilot, and more.

Each new product meant new signals, new usage stories, and more actionable context for sales.

The team also kept refining what went into Salesforce: not just a lead score, but the why behind the score - top reasons this user or account was likely to convert.

Context.

Request a Feature, Figma Style

The team drew inspiration from something they’d seen Figma do: let users hit a wall and request what they need.

They built the “request a feature” flow right into GitHub: if you tried to use something gated, you could ask for access.

Suddenly, Sales had not just a score, but a direct ask from the user.

It was the clearest signal yet.

Wild Results and Wild Reactions

The impact, when it hit, was obvious to everyone:

• PQAs converted at 2-3x the rate of classic MQLs

• Deals closed faster

What Others Can Learn: A Practical Playbook

You don’t need to be GitHub-sized to use these ideas. Here’s the playbook anyone can start with:

• Start before you’re ready. Manual rules, hand-checked signals, and weekly iteration beat waiting for the perfect system.

• Sit with Sales. Kill false positives, tune signals, and celebrate the deals that close from product, not marketing.

• Pipe stories, not just signals. Sales wants to know why - not just who.

• Let users raise their hand. A request-a-feature button could be your best sales signal.

• Be relentless with feedback. Every week is a chance to get better.

• Share your dashboards, but obsess over results. Data is only useful if it drives action.

• Work cross-functionally. Growth, Sales, Product, Data and Ops all need skin in the game.

The Bottom Line

The power of their approach lay in its simplicity: watch what users do, learn constantly, and build with the teams who know the customers best.

That ground-level focus on behaviour and collaboration turned out to be the most sophisticated sales engine they could have built.

Those qualities, more than any technology or framework, powered their transformation from open-source pioneer to enterprise sales machine.

If you missed the earlier posts, check out Part 1 for how the growth team started, and Part 2 for how a single signup flow change unlocked seven figures in new ARR.

If you’re building anything similar, remember: start with what your users are telling you, and let the results drive what comes next.

THAT’S A WRAP

Before you go, here are 3 ways I can help:

Take the FREE Learning Velocity Index assessment - Discover how your team's ability to learn and leverage learnings stacks up in the product-led world. Takes 2 minutes and you get free advice.

Book a free 1:1 consultation call with me - I keep a handful of slots open each week for founders and product growth leaders to explore working together and get some free advice along the way. Book a call.

Sponsor this newsletter - Reach over 7600 founders, leaders and operators working in product and growth at some of the world’s best tech companies including Paypal, Adobe, Canva, Miro, Amplitude, Google, Meta, Tailscale, Twilio and Salesforce.

That’s all for today,

If there are any product, growth or leadership topics that you’d like me to write about, just hit reply to this email or leave a comment and let me know!

And if you enjoyed this post, consider upgrading to a VIG Membership to get the full Product-Led Geek experience and access to every post in the archive including all guides.

Until next time!

— Ben

PS: Thanks again to our sponsor: Delve