Industry and Headcount Filters Find Prospects. Intent Signals Find Buyers.

01 /The Concept

Most B2B prospecting starts with a list pulled from a database that filters by industry and headcount. Firebender’s premise is that those filters are nearly useless – and that the signals that actually predict a purchase are buried across 100+ sources that no sales rep has time to check manually.

An AI engine under the hood pulls from more than 100 data sources to surface the highest-probability prospects out of millions of companies. Searches can be run on nuanced, informal criteria – for example: "AI product startups that have raised pre-seed or seed funding," "customer support leaders at companies that publish a support phone number on their website," "companies with more than 10 employees selling branded food products through retail," "startups providing medical services that use electronic health records," or "logistics companies with IT systems that handle customs clearance."

The underlying problem is that most B2B databases only filter by formal attributes like industry category and headcount. Sales teams then have to manually dig through company websites, LinkedIn profiles, job postings, and other sources to identify which companies actually match their ideal customer profile. Firebender does that research automatically.

The AI crawler continuously scans multiple sources, assembles the information into a unified database, keeps it fresh, and lets users run natural-language queries – which can then be refined with more structured filters.

One client used the platform to identify thousands of companies selling directly to consumers, with an FAQ page on their website, and more than 10 support staff. They also received decision-maker contact information – LinkedIn profiles and email addresses – for the relevant people at each company.

Pricing starts at $199/month or $599/month depending on query volume, covering a base of 100,000 companies. Access to the promised millions requires the enterprise tier, priced on request.

Firebender was founded last year and recently graduated from Y Combinator, receiving the standard $500K.

02 /Why It Matters

AI-powered company discovery turned out to be surprisingly popular as a category. Just within this Y Combinator batch, at least three startups were building variations of it.

Openmart ([related review](/review/vstan-mezhdu-do-hrena-i-do-figa)) helps manufacturers and national distributors find local resellers – shops, salons, dentists, hotels, restaurants, cafes, bars, tour operators, museums, regional distributors, and marketplaces – who might carry their products.

OffDeal ([covered here](/review/globalnyj-peredel-sobstvennosti)) helps national companies and private equity funds build target lists of local acquisition candidates.

The pattern is clear: AI-powered prospect discovery built on informal, multi-source signals is a real and growing need. YC doesn't pick three startups in the same space by accident.

The real differentiation in this category is the choice of signals. And Firebender did something interesting in its hero text that the rest of the copy didn't fully develop – it promised to find not just potential buyers, but "early adopters."

The term comes from the classic technology adoption curve. First come the innovators – enthusiasts who seek out anything new. Then come early adopters – people willing to try something unproven, but who need someone to bring it to them. Then comes a prolonged "chasm" that most products never cross. Products either fall into it, or eventually the early majority picks them up and growth accelerates. Late majority and laggards follow.

In other words, a product's success depends not just on what it does, but on where it is in its adoption lifecycle. Most people won't buy something simply because it's new. That's just how it works.

For a new product, early adopters are the critical audience:

- They're willing to buy before something is established. - There are meaningfully more of them than true innovators. - How many early adopters you acquire determines whether the product can survive the chasm – building enough of a base to still be around when the early majority finally shows up.

The problem is that early adopters don't behave like innovators. Innovators are self-directed hunters – they find new products on their own. Early adopters need to be found and approached.

But finding them is hard and expensive:

- Being an early adopter is an informal behavioral trait, not a category you can target in an ad platform. - Because they're a minority, running broad campaigns by formal criteria mostly reaches people who aren't ready yet – which tanks conversion rates and inflates acquisition costs.

Finding people and companies likely to be early adopters of a specific product is a genuinely valuable problem. Firebender named it clearly in its headline. It's less clear whether the platform actually solves it.

03 /Opportunities

The clear direction: build AI-powered discovery platforms that help sellers find buyers (or resellers, or partners) based on informal signals that matter in a specific domain. This doesn't have to be limited to B2B – the same approach could work in B2C.

In which industries do sellers spend a disproportionate amount of time qualifying leads? What informal criteria do they use? What sources do they pull from manually? What other signals could reasonably imply buying intent? How much of this could be automated?

A more specific angle: platforms for finding early adopters rather than generic prospects.

It's worth connecting "early adopters" to "superconsumers." In almost every product category, 10% of buyers account for 30–70% of revenue – they buy more frequently and at higher values. But here's the non-obvious part: superconsumers in one category are typically superconsumers in nine other categories too, often unrelated ones. It's less about passion for a specific product and more about a general disposition toward engagement and the financial means to act on it.

Which means: if you can identify superconsumer behavior in one category, you can make a reasonable inference about their receptivity in another. The same logic applies to early adopters – if someone was an early adopter of one product, they're more likely to be an early adopter elsewhere.

In a B2B context, if a small-company founder recently adopted a startup's product, they're probably open to trying another one from a different space. Or if a customer support director at a large company consistently buys the latest and most unusual consumer tech, your early-stage support automation tool probably has a shot with them.

Even as a thought experiment, this illustrates how unexpected informal signals can become the foundation of a genuinely interesting product in this space.

So – what set of informal signals, in which domain, could you build around?