Favors.dev
The fastest way to get your first reviews is to stop waiting for customers to volunteer them. With zero users, reviews are a cold-start problem, not a marketing one: you hand-pick the few people who have genuinely used your product, ask each one directly at the right moment with a specific, honest script, and route them to the one platform that matters for your stage. Do that — and never cross the line into fake or positivity-gated reviews — and you can go from zero to a credible handful in a couple of weeks.
This is the version for a solo founder with no customer base, no review-collection budget, and no reputation to trade on yet. It is part of the no-budget toolkit — if you haven't read the pillar, start with SaaS marketing on no budget for the full picture. Here we go deep on the one asset that now pulls double duty.
Why your first reviews now do double duty
Reviews used to do one job: convince a hesitant human to click buy. They still do that — but in 2026 they do a second job that most founders miss. Answer engines cite them.When ChatGPT or Perplexity assembles an answer about "best tool for X," it reaches disproportionately for community discussion, named-author articles, and reviews — the same raw material that reassures a buyer.
The data is lopsided enough to change your priorities. Similarweb's analysis of the most-cited domains by LLMs found Wikipedia (13.15%) and Reddit (11.97%) alone account for more than a quarter of U.S. ChatGPT citations — a finding echoed in the 5W State of AI Citations research. Dedicated review platforms and honest peer discussion feed that same appetite. In other words, a review isn't just social proof sitting on a page — it's a citation source that helps a machine decide whether to name your product at all.
That's the strategic reason to treat your first ten reviews as infrastructure, not vanity. The tactical reason is simpler: a brand with a visible rating converts, and a brand with none looks like a risk. Both jobs start with the same hard part — getting the very first ones when nobody knows you exist. This same off-page logic is why answer engines matter for founders at all; I unpack the full mechanics in how to get cited by ChatGPT and Perplexity.
The chicken-and-egg of zero reviews
The cold-start trap is real: you need reviews to get users, and users to get reviews. Buyers wait for other buyers to go first, so a page with zero reviews stays at zero by default. Nobody wants to be the first pancake.
You break the loop by rejecting its premise. You do not need customers to get honest reviews — you need people who have genuinely used the product. Early on those are different groups. Your first legitimate reviewers are the beta testers who kicked the tires, the handful of free users who hit the aha moment, the founder you swapped feedback with last week. None of them paid you, and every one of them can write a truthful review of their real experience. The FTC's rule doesn't require a reviewer to be a paying customer; it requires the review to be honest and any relationship disclosed.
So the cold-start move is to make a list — literally, a spreadsheet — of everyone who has actually reached your product's core value, and work it one person at a time. Ten hand-collected reviews from real users will out-earn a hundred you tried to manufacture, and they're the only kind that survive contact with the rules below.
Who to ask, when, and how (with scripts)
The order matters: ask the right person, at the right moment, with a request that lowers the effort to almost nothing. Get any of the three wrong and even happy users go quiet.
- Who: people who have hit the outcome your product promises — not everyone who signed up. A user who ran their first report is a reviewer; a user who created an account and vanished is not.
- When: right after a value milestone, not on signup day. The single most common mistake is asking too early (or on a 21-day autopilot). A window of roughly 7–10 days after the aha moment, while the value is fresh, converts far better.
- How:one specific ask, a direct link, and explicit permission to be critical. "Anything you write is fine" beats "please leave a 5-star review" — it's more honest, it's compliant, and it actually gets written.
Steal these three scripts. Fill in the brackets and keep them short — the whole point is that replying costs the user almost nothing.
The email ask
Send 7–10 days after a user hits a real milestone — not on signup day.
Hi [name] — I saw you [did the thing that proves value: ran your first report / connected your account / shipped with it]. I'm a solo founder and honest reviews are how a brand-new tool earns trust. Would you leave a short one here? → [direct link]. Anything you write is fine — the critical stuff helps me more than the praise. Takes about two minutes, and I really appreciate it.
The peer DM
For a founder who actually tried the product — not a cold stranger.
Hey [name], thanks for kicking the tires on [product] the other day. If it was useful, a two-line review on [platform] would mean a lot for a tool this new — and I'll happily return the favor on yours. If it wasn't useful, tell me why instead; that's worth more than a star rating.
The in-app prompt
Fire once, right after the user completes the action that delivers the aha.
Nice — that's [the core outcome] done. If [product] just saved you time, a quick review helps other founders decide whether to try it. [Leave a review] · [Not now]. (No pop-up on signup, no nagging: one prompt, at the moment it earned the right to ask.)
One reminder a few days later, to non-responders only, typically recovers a meaningful share of the people who meant to and forgot. After that, stop — a second nag costs more goodwill than the review is worth.
Where your first reviews matter most
Don't spread five reviews across five platforms — that leaves every page looking empty. Concentrate your first asks on the one place your buyers actually check, then expand. Here's how the main options stack up for a cold start, from hardest to easiest to seed.
| Platform | Best for | Bar to start counting | Cold-start difficulty |
|---|---|---|---|
| G2 / Capterra | B2B SaaS buyers actively comparing tools | 10 reviews to appear on the G2 Grid | High |
| App stores (iOS, Play, Chrome) | Consumer & prosumer apps; install-page conversion | A handful lifts you out of the unrated 'new' look | Medium |
| Product Hunt reviews | Launch-day credibility + an evergreen listing | 5–10 during launch week | Medium |
| Community pages (Favors.dev /apps, Reddit) | Founder-buyer trust + AI-citable discussion | One thoughtful review already reads as real | Low |
| Your own testimonial wall | Landing-page conversion | 2–3 named quotes with a face | Low |
For most solo founders the smart cold-start order is right-to-left on that table: earn a few real, thoughtful reviews in low-friction places first — your public project page, a testimonial wall, a Product Hunt listing — then use that proof to make the higher-bar platforms like G2 feel less empty when you get there. The one place designed for exactly this is a founder marketing co-op, where other founders will actually try your product and leave an honest take.
The line you don't cross
Everything above works because the reviews are real. The moment you fake them, you trade a slow-but-durable asset for a fast liability — and since 2024 it's also a legal one. The FTC's final rule banning fake reviews and testimonials (effective October 2024) makes buying, selling, or fabricating reviews — including AI-generated ones from people who never used the product — an enforceable violation with penalties that can reach tens of thousands of dollars per instance. The rule specifically covers undisclosed insider reviews and incentives conditioned on a positive rating.
The practical test is easy to remember. You can reward the act of reviewing; you can never reward the rating. Here's the same line drawn as a checklist:
Fair game
- Ask real users directly, at a moment they've felt the value.
- Seed early reviews from genuine beta testers who actually used it.
- Offer a small thank-you for the act of reviewing — if you'd give it for a critical review too, and the reviewer discloses it.
- Send one reminder to non-responders a few days later.
- Publish the negative ones and reply to them in public.
Over the line
- Buy reviews, or trade five-stars for five-stars.
- Write them yourself, or have teammates post without disclosing the relationship.
- Gate the reward on a rating ("$10 for a 5-star") — that's the illegal part.
- Use AI to generate reviews from people who never touched the product.
- Threaten or suppress honest negative reviews to make them disappear.
Beyond the law, faking reviews defeats their entire purpose. A fabricated five-star tells you nothing about what to fix, converts worse than an honest three-star with specifics, and reads as fake to both buyers and — increasingly — the answer engines that are learning what manufactured praise looks like. The durable move is boring: get fewer reviews, but real ones.
Frequently asked questions
How do I get reviews for a brand-new product with no customers?
Start from the few people who have genuinely used it — beta testers, your first free users, and founders you've traded help with. Reviews with zero customers are a cold-start problem, so you hand-pick rather than broadcast: identify everyone who has actually reached the product's core value, ask each one directly at the right moment with a specific script, and route them to one platform that matters for your stage. Ten honest reviews from real users beats a hundred you had to manufacture.
Can I offer a discount or gift card in exchange for a review?
Only for the act of reviewing, never for a positive one — and it must be disclosed. Under the FTC's 2024 rule on consumer reviews, you cannot condition any incentive on the review being favorable, and insiders must disclose their relationship. Many platforms (G2, Yelp, the app stores) also restrict or ban incentivized reviews outright, so check each one's policy first. The safest first-review strategy skips incentives entirely and simply asks real users well.
Where should a new SaaS get its first reviews?
Where your buyers already look, at the lowest bar you can clear now. B2B tools eventually need G2 or Capterra (10 reviews to show on the G2 Grid), but for a cold start the lower-friction places — your Product Hunt listing, a community directory page, and a testimonial wall on your own site — earn credible proof faster. Pick one primary platform for your category and concentrate your asks there instead of spreading five reviews across five sites.
How many reviews do I need before they start helping?
Fewer than you think for trust, more than you think for rankings. Two or three specific, named reviews already lift a landing page's conversion because they answer a hesitant buyer's objection. Platform effects kick in later — roughly ten to appear on comparison grids, and dozens to rank competitively within a category. For a solo founder, the first goal is a credible handful, not a leaderboard.
Do product reviews help with SEO and AI search?
Yes — reviews are now double-duty proof. Beyond convincing human buyers, review and community content is exactly what answer engines lean on: Similarweb's citation analysis puts Reddit and Wikipedia together above 25% of U.S. ChatGPT citations, and dedicated review discussion feeds the same appetite. A product that's genuinely reviewed and discussed is more likely to be named inside an AI answer than one that only has a polished homepage.