Chapter co-founders on rigorous research, no-BS sales, and finding the perfect customer segment

Chapter co-founders on rigorous research, no-BS sales, and finding the perfect customer segment

Engineers are the hardest customers to sell to. For sales teams pitching developer tools, their first approach needs to be just right. However, instead of focusing on those crucial meaningful conversations with potential customers, sales agents spend a lot of time researching leads and stitching data together. Chapter wants to change that. 

Chapter is an AI tool that helps go-to-market teams scale sales outreach by sending them enriched, pre-qualified leads every day. AI agents scour the internet for rich, referenced data specific to developer tools and save each user around 6 hours per week. 

We spoke to Chapter’s co-founders, Robin Greenwood and Shrirang Anil Bhate, about how they identified their perfect customer segment and how more meaningful outreach results in more closed sales.

Can you explain what you’re developing at Chapter?

Robin: We’re developing AI agents that automate lead generation and research specifically for outbound sales of developer tools.

Selling to engineers can be especially challenging because they require trustworthy, detailed, and relevant data. If a sales agent’s initial contact is meaningful and well-informed, it maximizes conversion potential down the line. Existing tools often fall short in this regard because they are typically designed as general-purpose solutions.

For example, existing platforms rely on users starting with a pre-existing list of leads. Users then use credits to enrich these leads and determine if they’re relevant. However, our customers often find they need to do additional manual research to uncover the granular details required for effective outreach.

We take a different approach. Instead of starting with an existing list, we scrape the internet on behalf of our customers.

Devs aren’t exactly famous for being receptive to sales pitches…

Robin: Exactly—that’s the key. There’s a very specific language and mindset we embraced: the "no-BS approach." 

We didn’t want to drag people into yet another tool that required endless clicking and setup. From the beginning, we had these core mantras and they really resonated with our audience.

Fair. But how have you managed to communicate value to the people who need this product? 

Robin: Most developer tools use self-serve or open source as their initial GTM strategy. This can work really well since they’re typically solving a problem they’ve experienced directly and can find individual users to try out and test. 

There are a few limitations however. First, users need to know you exist and this can take time. Then, there's very little viability in the broader team usage. For example, having a single user engage with a product doesn't tell you anything about their team needs, structure, or budget. And because of this, the feedback loop can be really slow. 

During the program we refined our focus to companies that understand the potential of a sales-led motion. Typically these are companies that already have one established. 

We realized that if we started the conversation too early, teams would often conflate lead-gen with making actual sales and would define the success of our initial product by whether or not they closed a sale and not by time saved. It's far easier to create value for those that recognize it. And that segment is growing.

Can you give a specific example of how you can provide high-quality leads?

Sure! So let’s say I'm a developer tool company building documentation software. 

I can use Chapter to give me leads that include info like:

  • Whether they use existing outdated or incompatible doc software
  • If they’re hiring technical writers
  • If they have a public-facing documentation page
  • If they release product updates frequently which means upkeep is important
  • If their engineering teams grow quickly so onboarding is important 

Combining all this data makes outreach targeted, relevant, and yields a high win-rate. 

Have you been able to define a clear market segment? 

Robin: Our ideal customers are teams that have just started a sales process, typically with 1–5 people on their sales team. These teams spend the bulk of their time on research but would rather focus on meaningful conversations with high-value customers. That’s where we come in—delivering leads with context so they can shift their time toward closing deals.

It’s been challenging to identify and target this narrow segment, and much of our time at Techstars has been spent refining this focus.

Shrirang, how has it been on the technical side? Are you able to stick to a long-term vision or are you having to adapt week-to-week? 

Shrirang: We’ve had to adapt week by week based on what we learn. The main reason for this was that initially, we were focusing solely on founders but we’ve since expanded to include go-to-market teams including account executives and VPs of marketing. We definitely have a lot to learn but we go into every call knowing that customers are the experts on their needs. We aim to uncover those. We’re starting to see consistent patterns emerge and our customers have helped validate those patterns.

What we’re building is addressing real pain points—pain points that people are willing to pay for. However, the data we currently provide is just the starting point. We haven’t even touched GitHub data yet so there’s a huge opportunity in this space. We’ve only scratched the surface. 

The way to do that is by continuing conversations, studying how our existing customers use the product, and growing our customer base. As that happens, our data points and learnings will expand too.

In terms of the vision for the next 6 to 12 months, the outcome is becoming clearer. Our goal is to enable a founder or sales team to receive leads complete with all the information they need to craft a personalized message—without even a single click.

We believe future billion dollar companies will be run by teams of fewer than 10 people. We also believe that developer tools will grow in volume and usage beyond our wildest expectations. We are the ones that will enable these teams to reach mid-market and enterprise customers through relevant outreach. This means having the perfect lead, at the perfect time, and being able to craft the perfect relevant message will be key.

Did you come to Techstars with a particular goal? 

Robin: Our original goal was to have 10 paying customers. We started with two on day one, and signed a third last week.

That said, we’re optimistic about reaching five. We have a couple of warm leads we’ve already demoed with, and we’re hopeful they’ll convert.

To add to your earlier question about goals and processes: we’re focused on selling before building. Over the past couple of weeks, Shri has been creating custom demos for every prospect. Here’s how it works:

  1. Shri ensures we can provide value
  2. He then creates an account for them and runs our app to find leads
  3. I record a Loom demo tailored to their use case
  4. I send a demo and login access to our prospects

The goal is to identify the most critical data or features people are willing to pay for right now and double down on delivering that value.

That’s been our process recently—creating tailored experiences to test what resonates most. While we aimed for 10 customers, we’re optimizing for learning at this point and would rather learn the hard lessons early. 

For any potential customer that might read this, what would you like them to understand about Chapter? 

Shri: What I want them to understand is that their time is better spent interacting with customers, not clicking around and stitching data together from different browser tabs. Our goal is to take that burden off their plate so they can focus on meaningful engagement with their customers.

Robin: Well said. I’d just add one more point based on our experience with an early customer: they realized that to build a scalable and reliable business, they couldn’t rely on self-serve forever. Our goal is to help bridge that transition.

We understand that most of our prospects begin with a self-serve model—it makes sense for their audience and their stage. But when they’re ready to start selling to larger companies, we want to ensure they don’t face a cold start problem. We aim to make that shift as seamless as possible.

Thanks so much, Robin and Shrirang! Wishing you every success!

Henry Diep

CEO @ DearFlow (Techstars '24) | On a mission to make all administrative tasks disappear ✨

1d

Let's goooo Robin & Shrirang! 🚀

To view or add a comment, sign in

More articles by Techstars Berlin

Explore topics