Why there is no magic number (and who keeps selling you one)
Contractors ask us for a target because a target is easy to chase. Google will not give you one, because reviews are not a threshold you cross. They are one weighted input among many, scored against your competitors, and the score is local. Fifty reviews wins in a rural county where the top plumber has thirty. Fifty reviews is invisible in a metro where three HVAC companies each carry four hundred. Same fifty, opposite outcome, because the number was never the point.
The number-sellers exploit that. A blog says "you need 100 reviews" because 100 is a clean, quotable figure that ranks for the search you just ran. It is not wrong because 100 is too high or too low. It is wrong because it answered a relative question with an absolute number and never looked at your actual market. Ask that same blog what the roofers in your ZIP code are sitting at and it goes quiet, because it never checked and never could.
Here is the mechanic in plain terms. When a homeowner searches "electrician near me," Google assembles the three-pack from businesses that are close, relevant, and prominent. Reviews feed two of those. Volume and rating build prominence. Keywords and freshness in the review text feed relevance. So the right question is never "how many do I need," it is "how many does it take to pass the third-place profile Google shows right now, and can I hold that lead as they keep collecting too."
That reframe changes what you measure. You stop staring at your own count and start reading the leaderboard. It also changes how you should hear any pitch: an agency that quotes you a review target before it has looked at your market is guessing, and the one thing to bury for good is the idea that a number can be promised in advance. It cannot, because the number lives in your competitors' profiles, not in a best-practices PDF. We cover how to read those profiles next.
Read the leaderboard: how to find your real target
Your target is sitting in public. Open Google, search the exact term a customer would use ("roof repair Fort Myers," "drain cleaning [your city]"), and look at the three profiles in the map pack. Write down each one's review count, star rating, and the date of their most recent review. That table is your target. You are not aiming at 100. You are aiming to pass the profile in position three, then position two.
| What to record | Why it matters |
|---|---|
| Review count | Raw volume. The most visible signal to both Google and the homeowner reading results. |
| Star rating | A 4.9 with 80 reviews often outperforms a 4.4 with 200. Rating gates trust before count does. |
| Most recent review date | Freshness. A profile that stopped collecting a year ago is beatable even at high volume. |
| Reviews mentioning the service | Relevance. Reviews that name the exact job feed the keyword match Google reads. |
Run the search from your actual service area, not from your office if it sits in a different suburb. The map pack changes block to block. Check it from two or three towns you serve, because your target shifts across your footprint and your weakest map pack is the one costing you jobs. A roofer who looks strong from the shop can be buried three towns over, and that is exactly the search a homeowner over there is running.
Now do the math the useful way. If position three shows 110 reviews at 4.6 with its last review a month ago, and you sit at 70 at 4.8 with a review yesterday, you are closer than the raw count says: your rating and freshness already beat theirs, so the gap is smaller than 40 reviews of raw work. Read all three columns together, never the count alone. The leaderboard tells you which lever is actually short.
Do this quarterly. Your competitors are not frozen. The profile you needed to beat in January collected forty more reviews by April, and a target you hit once and stopped chasing is a target you lose. This is exactly the ongoing monitoring a reputation system exists to run, so you are not doing it by hand at 9pm on a Sunday, and so a competitor's quiet surge shows up on a dashboard instead of in your missing phone calls.
Rating and recency beat raw count more often than owners think
Owners fixate on the big count because it is the number printed next to the stars. But two other levers move you up the pack, and they are cheaper to pull than grinding out another two hundred reviews.
Rating first. A homeowner scanning three results reads the stars before the count. A 4.9 reads as "safe." A 4.3 reads as "why the low ones." Google knows this and weights rating heavily in prominence. A 4.9 at 90 reviews frequently sits above a 4.5 at 250, because the trust signal is cleaner. So the play for a lot of contractors is not "get more," it is "stop the bad ones and lift the average," which means fixing the job and the follow-up, not just the ask.
Recency second. A steady drip of fresh reviews signals a business that is busy right now. A wall of five-star reviews that all landed in 2022 signals a business that peaked and coasted. Google reads the timeline. Homeowners read it too: the "most recent" review is often the first one they actually open. A profile earning six to ten reviews a month, every month, beats a dormant profile with a bigger lifetime total.
The practical read: if you are close on count but your rating trails the pack, chase rating. If your rating is strong but your last review is three months old, chase recency with a consistent request habit. Raw volume is the lever you pull when the top profiles are genuinely ahead on all three and you have to close a real gap. Chase the cheap levers before the expensive one.
What the target actually looks like by trade and market
Because the number is relative, the only honest way to talk ranges is by market density, not by a single figure. Below is how the target tends to shake out. Treat these as what we typically see, not a promise, and always overwrite them with the real leaderboard from your own search.
| Market type | Rough review target to compete | What usually decides it |
|---|---|---|
| Rural / small-town, low competition | ~40-70 with a 4.7-plus | Rating and recency. Volume gaps are small, so a clean average wins. |
| Suburban, moderate competition | ~80-150 with a 4.7-plus | All three levers. This is where most contractors are fighting. |
| Metro, high competition | 200-400-plus with a 4.8-plus | Volume becomes table stakes; rating and freshness break the tie. |
Trade matters inside those bands. High-ticket, low-frequency work (roof replacement, HVAC install) produces fewer review opportunities per year than high-volume work (drain cleaning, lawn service, garage-door repair), so a roofer's "competitive" count is naturally lower than a plumber's in the same city. Do not compare your roofing profile to a plumber's leaderboard. Compare it to the roofers Google shows.
Seasonal trades add a wrinkle. A landscaper who earns most reviews April through September looks dormant all winter unless the request habit keeps a trickle going in the off months. That winter gap is where a competitor who never stops asking quietly passes you. Plan the ask around your slow season, not just your busy one.
The gap most contractors miss: reviews are also an AI-answer citation
The map pack is no longer the only place your reviews get read. When a homeowner asks ChatGPT or Google's AI answer "who's the best plumber near me," those engines pull from the same review corpus, your count, your rating, and the actual sentences customers wrote. A profile with a hundred detailed, service-specific reviews gives an answer engine something concrete to quote. A profile with thirty vague "great job" reviews gives it almost nothing.
This raises the bar on review quality, not just quantity. Reviews that name the job ("replaced our 30-year-old panel," "fixed the AC the same day in July") are the ones an engine can lift into an answer. That is a reputation asset problem, which is our lane: shaping the request so customers describe the work, then displaying and marking up those reviews so both Google and the answer engines can read them cleanly.
We do not re-teach the ranking mechanics of AI answer engines here; that is its own discipline. The point for this guide is narrower and it changes your target: you are now collecting reviews to satisfy two audiences at once, the map pack algorithm and the answer engine, plus the human who reads them before they call. A number that only chases the map pack undershoots the other two. Detailed, recent, high-rated reviews serve all three, which is one more reason quality and freshness beat a bare count.
How to hit and hold your target without gaming it
Once you know the leaderboard, the work is a habit, not a campaign. The businesses that pass the pack and stay there do four unglamorous things, consistently.
- Ask every satisfied customer, every job, on a system. The single biggest gap is not a bad reputation, it is no asking habit. A text with a direct review link, sent the day the job closes while the customer is happy, is the whole game. Do it by hand and it slips the week you get busy, which is exactly the week you needed it.
- Make the ask produce detail. Prompt for the specific job, not a generic rating, so the review names the service. Those are the reviews that feed relevance and give answer engines something to quote.
- Respond to all of them, fast, good and bad. Responses are a signal to Google and a signal to the next homeowner reading. A calm, specific reply under a one-star does more for trust than the one-star does damage.
- Keep the drip going in the slow season. Freshness is a lever you lose by accident. A trickle every month beats a flood then silence.
What not to do: never gate reviews (filtering happy customers to the link and unhappy ones to a private form is against Google's policy and the FTC's), never buy reviews, never incentivize them. Getting caught erases the profile you spent years building. The whole point of doing this on a real system is that the honest version, asked at volume and consistently, comfortably beats the shortcut anyone else is tempted to take.
That system, the request, the monitoring, the response drafting, and the schema markup that displays it all, is what a reputation service runs so you can stay on the truck. If you want the ongoing month-to-month cost of running it, we break that down separately.