In the first part of this series, we covered the listings every business should claim in its first week online. This one is about the layer that turns those listings into customers.

A listing tells someone you exist. A review tells them you are worth the risk. And increasingly, reviews are what AI assistants read when someone asks them to recommend a business.

But “get more reviews” is useless advice without knowing where — and the right platforms for a bakery and for a software company have almost nothing in common. So here is the split.

Start with Google

Google is the default for every business, in every sector, everywhere. Free, the most visible, it feeds map results and local search, and it is disproportionately what AI assistants draw on for local businesses.

If you have ten reviews in total, you want them on Google. Everything below is what you add once that is working.

If you sell to consumers

Trustpilot — free tier, works across almost every consumer sector, and ranks well for “[brand] reviews” searches. When someone searches your name plus “reviews,” you want a page you can respond on rather than a forum thread.

Facebook — still where a lot of people check that a business is real, particularly for local services.

Yelp — dominant in the US, meaningful in a few other markets, irrelevant in most. Check whether it ranks for your category in your country first.

Your sector’s platform — usually the one that actually drives bookings. Tripadvisor and OpenTable for hospitality; Booking.com for accommodation; Houzz or your local trades directory for home services; the App Store and Google Play for apps, where the rating directly gates downloads.

One rule: find the platform that ranks on page one when you search what your customers search. That is the one worth your effort, whatever the listicles say.

If you sell to businesses

G2 is the primary B2B software review platform. A free vendor profile costs nothing, and G2 category pages rank for the “best [category] software” queries you would struggle to win on your own site.

Capterra, GetApp and Software Advice form a second cluster. G2 acquired all three from Gartner in a deal that closed in early 2026, but they still operate separately — separate review pools, separate profiles to claim. Claim both sides.

TrustRadius — smaller, longer and more detailed reviews, valued by buyers doing serious evaluation. Gartner Peer Insights — only if you sell to enterprise.

Clutch and GoodFirms — the equivalents for agencies, consultancies and service providers. If you sell services rather than software, this is your G2.

Partner and app marketplaces — if you integrate with a larger platform, its marketplace listing carries reviews and reaches buyers already in a purchasing mindset. The highest-quality review surface available to most B2B products, and the least competed for.

Trustpilot works for B2B too, and gets skipped because it is assumed to be consumer-only. LinkedIn recommendations are the closest equivalent for solo consultants.

Product Hunt is a different animal

Product Hunt is a launch platform first and a review platform second, and treating it like the others will waste it.

Your ranking is decided in a single day, measured midnight to midnight US Pacific time wherever you are. So plan the day around a foreign timezone — and accept that you effectively get one launch. A weak one is worse than none.

What people miss is that the page persists, ranking for your product name and collecting reviews long after launch day. That is what makes it worth doing properly rather than early.

The rule that catches people out: you can tell people you have launched. You cannot ask them to upvote — that breaches Product Hunt’s rules and gets products deranked or pulled. Announce, don’t canvass. Votes from brand-new accounts count for little anyway.

Before launching, you want a landing page that converts, something a visitor can try without talking to sales, a maker comment ready to post first, and some real history in the community. On the day, reply to every comment; engagement feeds the ranking. If any of that is not ready, wait.

How many, and how fast

Around ten reviews is where a platform starts treating you as established; twenty-five is where you appear in category rankings.

Recency and steadiness matter as much as the count. A 4.7 built in 2023 is worth less than a 4.5 with reviews from last month. And a slow trickle beats a burst — thirty reviews in one week looks like exactly what it usually is, and fraud detection agrees.

Which makes this a process, not a campaign. Build the ask into your delivery workflow and it compounds quietly.

How and who to ask

Ask everyone. Not just the customers you expect to be happy. A spread of ratings reads as authentic; a wall of five stars reads as bought. It is also the rule — sending only satisfied customers to the public platform while routing unhappy ones to a private form is called review gating, it has been sold as best practice for years, and it is now prohibited in the US, EU and UK. Google removes reviews or suspends listings over it; Yelp flags the business publicly.

Never tie a reward to a positive review. A thank-you offered to every customer regardless of what they write is fine. A discount for five stars is not.

Don’t post reviews yourself. Nothing from employees, owners or family without disclosing the relationship, and nothing AI-written.

Time it to a visible win — right after you have delivered something the customer is pleased with, not in a monthly newsletter.

Remove the friction. A direct link to the review page, not your homepage. Two clicks, maximum.

Make it routine. The businesses that reach fifty reviews are the ones who ask every customer, every time, without thinking about it.

Respond to everything

Reply to every review, positive and negative — it is visible to every future reader, and platforms treat response rate as an engagement signal.

Negative reviews deserve the most attention. A handled complaint often persuades a reader more than a five-star review does, because it answers the question actually on their mind: what happens when something goes wrong. Never argue publicly.

The checklist