Niche down to expand

Built from the system Nick Velkovski, co-founder and CEO at HeyReach, described on GTM Vault GTM 14 (https://pro.gtmvault.co/episodes/gtm-14). Published free by GTM Vault. Numbers below are the operator's own, from the episode; your results will differ.

When to use this

Use this when an early stage B2B product has users but flat or erratic revenue, the team is selling to several audiences at once, and nobody has decided who the product is for. It fits founders who have run enough sales calls to have a record, and who hold a strong opinion about the best segment that the record may not support. Nick describes HeyReach, his own LinkedIn outreach product, as stuck below $8,000 MRR while selling to agencies, salespeople, individuals and recruiters at the same time (GTM 14, 4:18). The system assumes the team can change its website copy, pricing and positioning within weeks, and has a way to log why deals are lost and why customers churn.

The system on the record

Nick's view is that niching down is the way to expand and grow, and that finding the niche is a bumpy road a team has to go through: try everything, see where the product fits, and double down on what works (GTM 14, 5:23 and 9:38). He calls it a mindset rather than a growth lever, paired with being customer centric (GTM 14, 10:41). It runs in six parts:

1. Validate the differentiator with the likely buyer. While pivoting from Howitzer, a Reddit automation tool, the team ran interviews with potential customers, most of them lead gen agencies. The tools they had used were built for one LinkedIn account on one campaign. When they described multiple LinkedIn senders on one outreach sequence, they got "the wow reaction," and that was enough to decide the feature set (GTM 14, 1:09).
2. Launch before it is perfect and measure payment, not usage. HeyReach opened a free 30 day beta. It drew 200 users, and when the beta ended one person paid (GTM 14, 3:14 and 4:18). Nick's general advice is to ship something good enough that solves a problem, then listen to the market and iterate (GTM 14, 13:50).
3. Test every segment and keep the calls. The team tried outreach to agencies, salespeople, individuals and recruiters. They hired an experienced head of sales, who left, worked with a consultant on the ICP, and tried battle cards. Nick says nothing changed (GTM 14, 4:18 and 5:23). During this period they were running out of money and could not raise on their MRR, so they launched a lifetime deal on AppSumo to fund the company (GTM 14, 4:18).
4. Let the call record overrule the founder. Nick was "95% sure" recruiters were the right vertical because they reacted well on his own calls and work mostly on LinkedIn. His revenue lead, Elia, reviewed 597 sales calls and argued for lead gen agencies, because the product in its current state was not built for recruiters (GTM 14, 6:27). They went with the call record.
5. Reposition everything at once. They changed the website wording to address lead gen agencies, changed the pricing to suit lead gen agencies, and started saying the product is a LinkedIn outreach tool made for lead gen agencies (GTM 14, 6:27). This was in October 2023 (GTM 14, 4:18 and 5:23).
6. Run a loss and churn loop into the roadmap. The team breaks down how much revenue and how many customers they lost and the reasons why, using closed lost deals and churn reasons. That list goes into the developers' roadmap. Once shipped, they reach out to the closed lost customers again (GTM 14, 9:38 and 10:41). Nick says to listen to how people talk about the tool and use their wording in what you build (GTM 14, 14:52).

Two decisions sit alongside this. First, buy adjacent tooling instead of building it: the team started building an email outreach tool internally, then found Instantly on AppSumo solving the same problem, killed the internal project and bought it (GTM 14, 2:10). Second, stay on one channel and integrate: HeyReach does not build email or cold calling to become a sales engagement platform, and instead integrates with the outbound tools buyers already use (GTM 14, 8:35 and 9:38).

The result

1. Before the niche: 200 beta users converted to one paying customer (GTM 14, 3:14 and 4:18). The first paying customer came in April or May 2023, and revenue did not pass $8,000 MRR until October 2023 (GTM 14, 4:18).
2. AppSumo launch: HeyReach became the number one bestselling tool on the platform and made "total of 400,000 in seven days" (GTM 14, 4:18). He did not state the currency, a net figure after AppSumo's share, or how many buyers converted to ongoing revenue. He named the AppSumo audience as the first adopters, but said the company was still struggling until it niched down (GTM 14, 5:23).
3. After the repositioning: "the next month we 2.5x'd our revenue" (GTM 14, 6:27). He gave the monthly figures as 8K MRR in October, 19 in November and 13 in December, and called it fluctuating revenue with clear signs from the market (GTM 14, 7:33). The December figure is lower than November and he did not explain the drop.
4. Organic demand: 2,000 free trial signups per month, which he attributes to people talking about the product and sharing their results. He says growth has been 100% organic so far and the team had only just restarted outbound (GTM 14, 10:41).
5. He gave no figures for churn reduction, win rate on re-contacted closed lost deals, or cost saved by buying Instantly instead of building.

Run it on your company

1. Ask once, in one message, for: every segment the team has sold to; how many sales calls exist and where they are recorded; which segment the founder believes in and why; paid conversion from any free trial or beta; current MRR by month; the website pages, pricing page and core messaging in use; closed lost reasons and churn reasons with the revenue attached to each; and any adjacent product the team is building or planning to build. Output: a one page intake summary with gaps marked.
2. Read the call record by segment. Code each call by segment, outcome, the product gaps raised and the words the prospect used. Output: a table with columns Segment, Calls, Paid or progressed, Main gaps raised, Fit with the product as it is today (high, medium, low), Sample prospect wording.
3. Name the segment and test it against the founder view. State which segment the record supports and where it disagrees with the founder's pick, as Elia's 597 call review disagreed with Nick's recruiter bet (GTM 14, 6:27). Output: a one paragraph recommendation with the call evidence, and a second paragraph stating the case for the founder's pick and why the record does or does not support it.
4. Rewrite the three surfaces together. Output: new homepage hero and subhead naming the segment, a revised pricing table shaped for how that segment buys (for example by sender or seat count, if that is how they scale), and a one line positioning statement in the form "[category] made for [segment]". Leave the price points for the user to set; Nick gave none on the record.
5. Set the evidence window. Output: a table with columns Month, MRR, New paying customers, Segment share of new customers, covering three months before and three months after the change. Flag that one month of growth can reverse, as HeyReach went from 19 to 13 between November and December (GTM 14, 7:33).
6. Build the loss and churn loop. Output: a table with columns Reason, Source (closed lost or churn), Customers lost, Revenue lost, Roadmap item, Ship date, Re-contact date; and a fenced re-contact message template for closed lost customers once their reason has shipped, with three sample versions for user review.
7. Run a buy or build check on every adjacent tool. Output: a table with columns Need, Built in house (yes or no), Existing tool that solves it, Decision, Reason. Use the Instantly decision as the reference case (GTM 14, 2:10).
8. Decide on a funding bridge only if needed. If runway is short and the company cannot raise, assess a lifetime deal launch as cash, not as ICP proof. Output: a short note listing what the launch would fund, the support load it creates, and how buyers from it will be kept out of the segment analysis in step 2.
9. Set the scorecard: paid conversion by segment, MRR by month, share of new customers from the chosen segment, revenue lost by reason, closed lost deals won back after re-contact, and organic trial signups per month, reviewed monthly.

Tools on the record

1. AppSumo: launch marketplace, $400K in seven days.
2. Instantly: email outreach, bought instead of built in-house.
3. LinkedIn: the outreach channel the system runs on.

Any tool doing the same job works; the method is the point. Full stack by layer: https://pro.gtmvault.co/stack/operator/nick-velkovski

Where it breaks

1. Founder conviction wins. Nick was 95% sure about recruiters because of how they reacted on his own calls (GTM 14, 6:27). A choice made from the calls the founder remembers, instead of the full record, picks the segment with the best demo reaction rather than the best fit.
2. Usage read as demand. 200 beta users produced one paying customer (GTM 14, 3:14 and 4:18). Free signups say little about who will pay.
3. Adding sales talent instead of focus. A senior head of sales, a consultant, battle cards and formal sales technique did not change the result while the segment was undefined (GTM 14, 5:23).
4. Lifetime deal buyers treated as the ICP. The AppSumo audience were the first adopters, yet Nick says the company was still struggling until it niched down (GTM 14, 5:23). Cash from a launch is not evidence of fit.
5. Half a repositioning. HeyReach changed the website wording, the pricing and the messaging together (GTM 14, 6:27). Changing one surface while the others still speak to everyone leaves the old signal in place.
6. Reading one good month as proof. Revenue went from 8K to 19 and then to 13 (GTM 14, 7:33). The niche needs several months of data before the team stops testing.
7. Building what can be bought. The internal email tool was killed once Instantly was found doing the same job (GTM 14, 2:10). Building adjacent channels pulls the team away from the segment it chose.
8. A loss loop with no re-contact. The loop depends on going back to closed lost customers once their reason ships (GTM 14, 10:41). Logging reasons without re-contact turns it into a report nobody acts on.