A 200-Review Profile Is Now Losing to a 60-Review Competitor
Here is the call we take most often in 2026. A contractor has been in business fourteen years, has 214 Google reviews at a 4.8 average, and has quietly slid out of the map pack in his own city. The company that replaced him has sixty-one reviews. He wants to know what he did wrong. The honest answer is that he did nothing wrong. He stopped, and the algorithm noticed.
Reviews were always a ranking signal. What changed after the March 2026 core update is how those reviews are counted. Correlation studies across large samples of local listings now point the same direction: a review's influence decays with age. Reviews collected in the last thirty days appear to carry close to full weight, while reviews past the six-month mark retain only a fraction of what they once contributed. The 214 reviews are still on the profile. They still convince homeowners. They are just no longer doing the ranking work the owner thinks they are.
That is the mental model shift this post is about. Your review count is a bank balance. Your review recency is income. Google appears to be increasingly interested in the income statement. A competitor collecting six reviews a week is generating more fresh signal every month than a fourteen-year archive that has gone quiet, and in a tight three-slot map pack, that difference decides who shows up.
None of this comes from a Google announcement. Google does not publish weights. It comes from third-party studies of thousands of listings, and it lines up with what we see when we take over a stalled profile and restart the review engine. Treat the numbers below as strong correlation, not gospel, and treat the direction as settled.
The Four Review Signals Google Weighs, Ranked by How Fast You Control Them
"Get more reviews" is not a strategy, because reviews are not one signal. They are four, and they respond on completely different timelines. Knowing which lever moves this month matters more than knowing the theory.
- Recency. How long ago your most recent reviews arrived. This is the signal most studies now rank first, and it is the one that decays whether or not you do anything. You cannot fix it retroactively. You fix it forward, starting this week.
- Velocity. The rate at which reviews arrive, and whether it looks natural. Steady beats spiky. Forty reviews in one weekend after two silent years reads as a campaign, not a business, and campaigns invite filtering.
- Response rate. The percentage of reviews you have replied to. Listings that reply to the large majority of their reviews consistently outperform comparable listings that do not. This is the fastest lever you own, because it does not depend on a single customer doing anything.
- Rating and volume. Your average star rating and lifetime count. Still real, still worth protecting, but the saturating signal. Going from 4.6 to 4.8 is worth less than going from silent to weekly.
Look at that list as an operator, not a marketer. Three of the four require a customer to act. One requires only you, sitting down for forty minutes with your profile. That is why the first move in every review rescue we run is not a review request. It is a reply backlog.
Not Sure How Far Your Reviews Have Decayed?
Book a free strategy call. We'll pull your profile, count your last ninety days, check your reply rate against the competitors beating you, and show you the exact cadence gap.
→ Book Free Strategy CallHow Many New Reviews a Week Your Trade Actually Needs
There is no universal number, and anyone selling you one is guessing. The number you need is competitive, not absolute: enough fresh reviews to stay in the top group in your city, for your category. But you can calculate your own target in about fifteen minutes, and it is the single most useful number in your local marketing.
Open Google Maps, search your main service plus your city, and take the three businesses in the map pack. For each one, sort their reviews by newest and count how many arrived in the last ninety days. Write the three numbers down. The highest of the three, divided by thirteen, is the weekly pace that is currently winning your market. If the leader has thirty-nine reviews in ninety days, the pace is three a week. If they have six, the pace is one every two weeks and your market is soft.
Then be realistic about your job volume. A roofing company doing eight jobs a week can support three reviews a week at a 40 percent ask-to-review conversion rate. A custom builder doing nine projects a year cannot, and should not try. In low-volume, high-ticket trades the winning play is different: you cannot out-pace anyone, so you compete on recency alone by making sure there is never a ninety-day gap on your profile. One review a month, every month, beats eleven reviews in March and silence until the following spring.
The failure mode we see most often is the burst. An owner reads an article like this one, texts every customer from the last three years on a Saturday, and collects twenty-two reviews in four days. Some get filtered, the ones that stick decay together, and six months later the profile is back to silent with a strange spike in its history. Spread the same twenty-two reviews across five months and you get five months of fresh signal instead of one weekend of it.
Your Review Replies Are a Ranking Input, Not Just Good Manners
Most contractors treat replies as customer service, so they reply to the angry ones and ignore the five-star ones. That is backwards for ranking purposes. Every reply is a piece of owner-generated text attached to your profile, written by you, about your services, in your service area. It is one of the few places left on a Business Profile where you control the words.
A reply that earns its keep does three things: it thanks the customer by name, it names the specific service and the town in natural language, and it says nothing that sounds copied. "Thanks Dana. Glad the new gutter guards are holding up through the Vernon fall leaves. Call us if you want the back run done before winter." That reply is worth writing. "Thank you for your review!" pasted forty times is not; it is closer to filler than to signal, and homeowners reading your profile can tell instantly.
Reply within a few days while the job is fresh, reply to every review including the one-line five-stars, and keep the keyword use conversational. Stuffing "best roofing company in Kelowna" into thirty replies is the same instinct that gets business names suspended, and it reads as desperate to the customer who is deciding whether to call you. For negative reviews, reply once, briefly, factually, and move the argument to a phone number. Future customers are your real audience, not the reviewer.
If you have a reply backlog, work newest to oldest and stop after ninety days of history. Old replies matter far less than current ones, and the goal is a profile that looks actively tended today. For the mechanics of asking, and the request scripts that actually convert, see our guide on how to get more Google reviews in 2026. This post is about what happens to them after they land.
The 30-Day Reset for a Stalled Review Engine
If your profile has gone quiet, here is the sequence we run. It front-loads the things that do not require a customer, so you get movement in week one.
- Week 1 — Clear the reply backlog. Reply to every review from the last ninety days, then as far back as you have patience for. Specific, short, human. This is entirely within your control and it is done by Friday.
- Week 2 — Build the ask into the job. Pick the single moment when the customer is happiest, usually final walkthrough or the day after completion, and attach the request to it. A text with the direct review link beats a business card, an email, or a QR code sticker on the invoice. One channel, one moment, every job.
- Week 3 — Work the recent backlog, not the archive. Text the last sixty days of completed jobs, a handful a day, not all at once. These customers remember you. Do not text 2023.
- Week 4 — Make it someone's job. Assign the ask to a person and the follow-up to a calendar. A review engine that depends on the owner remembering will stall again within a quarter, and you will be reading this article a second time next year.
Expect ranking movement to lag the work by four to eight weeks. Reviews are one input among many, and the map pack does not recalculate overnight. What you should see almost immediately is the leading indicator: reviews arriving weekly instead of never, and a reply rate climbing toward the whole profile. If those two are moving and rank is not, the bottleneck is somewhere else on the profile, which is where a broader Google Business Profile audit earns its money.
Audit Your Own Review Decay in Ten Minutes
Do this now, before you close the tab. You need your own profile and the top competitor from your map pack open side by side.
- Sort by newest. On each profile, sort reviews by most recent. Read the date on the top review. If yours is more than thirty days old, recency is actively working against you today.
- Count ninety days. Count how many reviews each profile received in the last ninety days. That ratio, not the lifetime totals, is the honest scoreboard.
- Check the reply rate. Scan the last twenty reviews on each profile and count how many have an owner response. Under 80 percent is a gap you can close this week.
- Look for the gap. Scroll your own history for any stretch longer than ninety days with no reviews. Every one of those gaps is a period where your profile looked dormant to both Google and homeowners.
- Read your last five replies. If they are interchangeable, they are not doing any work. Rewrite the habit, not the old replies.
Whatever the numbers say, resist the urge to fix it in one weekend. The businesses that hold the map pack are not the ones with the biggest archives. They are the ones that never went quiet, replied to nearly everyone, and treated the whole thing as a monthly operation rather than a campaign. If your rankings dropped and reviews were not the cause, our breakdown of the 2026 core update covers the other suspects. For an outside view of how recency is being measured across the industry, Epicware's analysis of review recency and local rankings is a reasonable starting point.