Small-Pond ICP Narrowing
Constrain your first ICP so tightly that customers talk to each other — then let word of mouth light the fire.
- Difficulty
- Moderate
- Time to result
- ~months to results
- Steps
- 4
- Confidence
- 90%
A go-to-market framework for a startup's first customers: define the initial ICP with multiple stacked constraints until the addressable set is small and homogeneous (Gong's first target was ~5,000 companies). A tight, similar cohort shares lingo and talks to each other, so you can build one focused product and ignite a viral word-of-mouth effect rare in B2B. Reshef contrasts this with his prior company's failure of selling to wildly different industries at once.
Origin
Eilon Reshef applies the 'crossing the chasm' / bowling-alley methodology (Geoffrey Moore) — which he ignored at his prior company, to his cost, then adopted deliberately at Gong.
Core principles
- 01A homogeneous cohort shares lingo and technology assumptions, so one focused product fits all of them
- 02Small ponds create viral word-of-mouth because customers in the same niche talk to each other
- 03Heterogeneous early customers can't be scaled — every one has its own language and needs
- 04Narrow first, expand later — starting broad is counterintuitively riskier
How to run it
- 1
Stack multiple constraints to shrink the pond
Define the ICP with several intersecting filters. Gong used: selling in the US, in English, over video conference, using WebEx, selling software priced $1,000-$100,000 — yielding only ~5,000 companies.
Pro tip Keep stacking constraints until the set is small and internally similar, even if it feels alarmingly narrow.
Watch out Resist the 'it's a huge market, it's for everyone' instinct — breadth here prevents word of mouth and dilutes the product.
- 2
Build one focused product for that homogeneous cohort
Because the cohort shares lingo and technology, build a single tightly-fitted product rather than customizing per industry.
Watch out Selling to different industries at once (e.g. cosmetics + financial services + tech) forces per-customer lingo and cannot scale.
- 3
Let the small pond ignite word of mouth
Rely on customers in the niche talking to each other to spread adoption — a viral effect achievable in B2B only when the market is small and connected.
- 4
Expand outward once the fire is lit
After dominating the initial pond, widen the ICP from that validated base.
In the wild
About a year into Gong, a company interviewing a salesperson was told by the candidate that he'd only work for companies that use Gong. That prospect became a Gong customer because of the interview. This viral, peer-to-peer effect only works when customers in a tight niche talk to each other.
→ A new customer acquired purely through niche word of mouth.
At his previous company Reshef read Crossing the Chasm, decided he could do better, and sold to L'Oreal, American Express, and Cisco simultaneously — different industries each with their own lingo. There was no way to scale it.
→ Unscalable go-to-market; a mistake he vowed not to repeat.
Common mistakes
Going horizontal / 'it's for everyone'
Targeting a wide market where customers don't talk to each other kills word of mouth and forces the product to serve incompatible lingos and needs — it cannot scale early.
Scaling sales before a repeatable ICP
At his prior company, investors pushed hiring ~20 salespeople across three unrelated segments before a focused, repeatable product-market fit existed; all failed miserably.
Is it for you?
Best for
Early-stage B2B founders choosing a beachhead market who feel pressure to go broad
Not ideal for
Companies past product-market fit ready to expand, or genuinely horizontal products with no connected niche to seed
From the transcript
“selling their product in the US in English over video conference uh using WebEx which was the big one at the time selling software that…”
“by having a smaller set of kind of customers or I see kind of definition of of customers you can develop like much more focused…”
“that's the sort of the power of a small pond with like you know companies that are like each other because you get this viral…”
“we had one customer in I think it was L'Oreal or I know one of the Cosmetics companies in American Express in Cisco like different…”
From the episode
Inside Gong: How teams work with design partners, their pod structure, autonomy, trust, and more
Eilon Reshef (co-founder and CPO)