What you fixed last year is quietly losing ground
The website and Google Business Profile work you did last year is silently losing ground while you're focused on running the business. Some signals decay in 30 days. Others in 90. The businesses being recommended by AI right now are the ones keeping all four maintained. Almost nobody is doing it on purpose.
The work that didn't stay fixed
A plumber in Calgary spent four thousand dollars in 2024 on a new website. Clean design, proper local SEO, Google Business Profile filled out at launch. He stopped thinking about any of it because the playbook said the work was done.
Eighteen months later his phone is ringing less than it did the year before he built the site. The site is still there, the reviews from 2023 are still there, his Google Business Profile still has the same eleven photos in the gallery from the day he set it up. From the outside, nothing looks broken.
Three of his four marketing signals have been decaying for fourteen months. He didn't do anything wrong. He stopped doing anything at all, and the work he paid for doesn't survive that.
"He didn't do anything wrong. He stopped doing anything at all."
The mental model that stopped working
Most service business owners are running on a mental model from 2019 that says marketing is something you buy once. You pay a guy to set up the website, another guy to do the SEO, you fill out the Google Business Profile, and then you go back to running the business.
That model is wrong for 2026 in a specific way. In the 2026 local ranking factor surveys, Google Business Profile signals account for 32% of all local pack weight, review signals for another 20%, and behavioral signals for another 8%. All three categories are dominated by recency-weighted measures, not one-time setup work. A review from this month carries more weight than a review from two years ago, a photo from last week carries more than one from 2023, and content from the last thirteen weeks carries more than anything from 2024.
Every signal has a half-life. Every month you're not maintaining the signals, the ones you spent money setting up are getting quietly weaker.
"Local search is a maintenance system, not an asset. Owners treat it like an asset and wonder why it stops working."
Decay curve one of four
Most owners think reviews are cumulative. You collect them, they sit on your profile, they keep working for you. That used to be true. It isn't anymore.
Review recency became the number-one individual ranking factor in the 2026 local search ranking factors survey. Review signals as a category now account for 20% of all local pack weight, up from 17% in 2024. Recency outranks total count, response rate, and review keywords as the highest individual signal Google reads on a business.
The practical effect: a business with 80 fresh reviews from the last quarter now outranks a business with 200 reviews and nothing in the last six months. Sterling Sky's 2026 case study on review recency tracked a business whose rankings dropped sharply the month its review velocity flat-lined, and recovered when the owner restarted the ask cadence. The reviews you collected in 2022 aren't carrying you the way they used to.
The decay shape is rough. Reviews from the last 30 days carry the most weight, the last 90 days still count, and past that the weight starts dropping. By six months a review is largely just there for social proof, and by eighteen months it's barely doing either.
If your last review is older than 30 days, your review signal is decaying in real time. Two years without a fresh ask and you've spent the review capital you didn't realize you had.
"The reviews you collected in 2022 aren't carrying you the way they used to."
Decay curve two of four
The GBP photo gallery is the second decay curve and it moves faster than anyone expects. Google Business Profiles with photos posted in the last month get 45% more direction requests and 31% more website clicks than profiles whose latest photo is older. Google's own internal data, surfaced in the 2026 ranking factor surveys.
Profiles that hadn't posted anything in over 30 days lost measurable impressions during the 2026 algorithm updates. Not a small drop. A measurable one. Behavioral signals as a category now account for 8% of all local pack weight in 2026 and the share is rising every quarter.
The cadence Google reads as "active" in 2026 is one photo or post every two to four weeks. Anything slower and the profile starts reading as dormant. Hours that haven't been confirmed in six months drop your "open now" reliability score. "Open now" during the search is the fifth-strongest individual ranking factor in the 2026 local pack. For emergency queries (best plumber near me, emergency electrician, urgent dental), Google now filters closed businesses out of the pack entirely.
The work isn't hard. It's just relentless. Ten minutes once every two weeks if someone is on the calendar to do it. Six months of silence if nobody is.
"The work isn't hard. It's just relentless."
Decay curve three of four
Schema markup is the small piece of code AI reads on each page to confirm what the page is about: this is a plumbing service, it serves Calgary and these neighborhoods, it has 4.7 stars from 247 reviews. Without schema, AI has to guess. With schema, it knows.
Most service business websites that get schema set up at launch never have it updated again. Then the site gets a redesign in year two, or the owner adds a new service page without telling whoever set up the schema, or the review count grows past 89 while the schema still says 4.7 stars from 89 reviews. The original schema is now describing a business that no longer exists at the same address.
Stale schema is worse than no schema. AI cross-references the schema against everything else on the web about the business and reads inconsistency as low confidence. Low confidence means AI picks the competitor with cleaner data.
The decay curve here isn't time-based. It's edit-based. Every time your site is touched without someone updating the schema to match, the signal degrades a little.
"Stale schema isn't neutral. It's actively pushing AI toward your competitor."
Decay curve four of four
The fourth decay curve is content freshness, and it's the slowest of the four, which is why most owners don't notice it.
AI weights content updated in the last thirteen weeks higher than older content. Pricing, hours, service area descriptions, service detail pages, anything visible on the site that could plausibly have changed. Past the thirteen-week window the page still shows up but with less confidence. By six months AI starts treating it as historical, and by eighteen months the model can read the business as possibly inactive.
The 2026 BrightLocal Local Consumer Review Survey put another number on this. 41% of customers now read reviews and content closely before calling a service business, up from 29% in 2024. That's a 12-point jump in twelve months. The bar for what counts as "current" went up while most owners' content stayed flat.
The fix isn't hard. A small edit on a few service pages every quarter, updated pricing if anything changed, a line about a new neighborhood you're serving, a current-year note in the footer. Maintenance work, not content marketing.
Of the thirty sites I audit a month in 2026, maybe two or three have content that's been touched in the last quarter. Everyone else is running on copy from 2023 or 2024 and watching their AI recommendation rate slowly drop.
"Maintenance work, not content marketing. Almost nobody is doing it."
Why the math gets worse than you'd expect
The four decay curves don't run independently. They feed each other in a way that makes the math worse than the sum of the parts.
A drop in review velocity drops AI's confidence in your business, which lowers your impressions, which means fewer customers reach you, which means fewer reviews next month. Stale photos drop your GBP impressions, which cuts the number of people who could have left reviews in the first place. Stale schema lowers your AI citation rate, which lowers traffic. Stale content drops search ranking, which compounds into everything above.
Each curve alone is a slow leak. All four together is an accelerating drift.
The business that fixed everything once and walked away in 2024 is not where they were in 2024. They're measurably further behind. Six months from now they'll be more behind. Twelve months from now their competitor who runs the maintenance is going to be hard to catch.
Signal strength over time without maintenance
"Each curve alone is a slow leak. All four together is an accelerating drift."
Why owners don't maintain it themselves
Here's the part most marketing content skips. None of the work above is technically hard. A motivated owner could learn each piece in a few hours. The reason it doesn't get done is not skill. It's the calendar.
The actual time math, roughly:
All in, you're looking at roughly two billable hours a month of attention from someone who knows where to look. The work itself isn't hard or technical. It just has to be consistent, on a calendar, on someone's responsibility list.
The thing service business owners run out of is not skill. It's someone whose job it is to keep that calendar.
"The thing service business owners run out of is not skill. It's someone whose job it is to keep the calendar."
A specific picture
Picture a plumber whose website has someone running the maintenance on it.
Every two weeks a new photo gets posted to GBP. A real phone photo from a real job that week. The customer who just had their burst pipe fixed gets a text the same evening with a one-click review link, and most weeks at least one new review shows up. The service page descriptions get updated once a quarter when pricing shifts or a new neighborhood gets added. The schema gets a check after any site change. Nobody is doing heroic work. Someone is just doing the small consistent work.
Twelve months in, that plumber's GBP impressions are higher than the year before, his AI citation rate has gone from intermittent to regular, and his phone is ringing more than the competitor across town who's still running on a 2024 setup.
Nothing dramatic happened in any single month. The compounding happened because nothing decayed.
"Nothing dramatic happened in any single month. The compounding happened because nothing decayed."
The question you're probably asking
If you're already paying a marketing agency or a freelancer, the natural question is why this isn't already happening. Here's the honest answer.
Most agencies still operate on the 2019 mental model. They set up the website, run some Google Ads, send a monthly report. The work is real but it isn't the work that maintains the four decay curves. They're not lazy. They're working from a different playbook than the one Google and AI search now use.
Some agencies have updated. They run review velocity systems, GBP cadence, schema audits, content refresh on a calendar. Most haven't, because the shift happened faster than agency tooling caught up to it.
The gap between "we do marketing for you" and "we maintain the four signals for you" is the gap that's costing service businesses the most right now. It's not the agency's fault. But the gap is yours.
"The gap between 'we do marketing for you' and 'we maintain the four signals for you' is the gap costing service businesses the most right now."
What to do next
The audit at weblingo-quiz.pages.dev runs through your business and shows you where each of your four signals currently sits. Which ones are still fresh. Which ones have started decaying. Which one is the highest-leverage fix.
It's free. Takes two minutes. No call.
From there you have a clear picture and you can decide what you want to do about it. Some owners take the report and run the maintenance themselves, some figure out what to ask their existing agency to add, some hire someone to run the four curves on a calendar so they can go back to running the business. All three options are better than running on 2024 signals into 2027.
Free. Two minutes. No call required.
weblingo-quiz.pages.dev
Start the 2-minute auditThe audit shows where each of your four signals currently sits on the decay curve, and which one is the highest-leverage fix right now. Free while we're testing v2 of the audit software.
P.S. The four decay curves aren't theoretical. They're already running on your business right now, including the months between when you fixed the signals and right now. The audit tells you how far each one has slipped. Whether you decide to fix it yourself, ask your agency, or hire someone to keep the calendar, knowing where you actually stand is the first move. Two minutes. Free.