Reversing an 18-Month Decline in Google Business Profile Visibility

The problem: sixteen consecutive months below the prior year

Local search visibility had been eroding for a year and a half before anyone had a number for how much.

Read in sequence, the scheduled performance reports told one story. Total Google Business Profile impressions came in below the prior year in sixteen consecutive 30-day windows, bottoming at 22.5 percent down. In absolute terms the network fell from roughly 193,000 impressions per 30 days at its 2022 peak to roughly 113,000 by June 2024. That is about 41 percent of its visible presence in Google Search and Maps, gone.

Review acquisition told the same story from a different angle. Across the entire network the locations were collecting somewhere between ten and twenty new Google reviews a month between them, and nearly every month ran behind the year before.

How we approached the turnaround

Fix the measurement first, then run everything else as a test with a scheduled read-out.

The account was generating plenty of reporting and almost no decisions. Our first move was to make the numbers usable: one consistent comparison window, one definition per metric, and a baseline for every location rather than a single blended figure that hid which markets were actually failing.

From there, every change ran as a documented test. Each one carried a written hypothesis, a named validation method, a defined data source, and a fixed read at 30, 60 and 90 days. Tests that earned their place got rolled out across the network. Tests that did not were stopped and recorded as such, which matters more than it sounds: the reason this program compounded is that we were not quietly carrying failed work forward.

The work itself ran across four fronts, applied to the whole estate rather than to a handful of flagship locations:

• Listing-level optimization across every profile in the network, including the categories, attributes and content fields that actually feed local relevance.
• Review generation, rebuilt as a repeatable process rather than an occasional ask.
• Location content and structured data, so each market had genuinely location-specific pages and machine-readable detail behind them.
• Competitive grid monitoring, to see ranking movement in a market before it showed up as a drop in impressions.

The specific tests, the order we ran them in and the thresholds we used to call a winner are the part we keep for clients.

Results: twenty-four months without a negative reading

The trend turned in August 2024 and has not turned back.

The first positive reading is the window ending 4 September 2024, at 13.7 percent up. Nothing measured after it is negative. Across the two years since, the reporting moved to overlapping three-month windows, which strip out most of the month-to-month noise in Google’s data. Eighteen of those windows have now been measured. Mobile Search impressions rose in all eighteen. Direction requests, the highest-intent action Google reports on a business listing, rose in seventeen of eighteen and have been accelerating, reaching 25.3 percent in the most recent window.

Growth was not carried by one strong market. In that most recent window, direction requests rose at every single location in the network, and mobile Search impressions rose at all but one. That breadth is the signature of work applied consistently across an estate, not of a single market getting lucky.

The same recovery as an absolute level

Percentages hide how far the account had actually fallen. Because every comparison report states both its own period and the period a year earlier, the two sides can be stitched into a continuous level series. It puts the low point in June 2024 at roughly 113,000 impressions per 30 days, against about 193,000 at the 2022 high and about 163,000 by July 2026.

Local pack rankings improved 27 percent on the tracked query

Ranking coverage across a two-mile grid, measured monthly against five named local competitors.

Average Total Grid Rank measures the mean position across every point of a 13 by 13 grid, so it captures coverage across a whole trade area rather than a single point-blank ranking. Lower is better. On the tracked programme query it improved from 20.86 to 15.22, a 27 percent gain, and has held between 14.5 and 15.2 for eight consecutive months. The best single reading was 14.48.

Review generation went from a trickle to a program

The one metric here with a clean step change rather than a trend.

Through 2023 the network collected roughly ten to twenty Google reviews a month across all locations combined, and nearly every month ran behind the year before. That inverts in mid-2024. The 30 days to 31 July 2024 brought in 111 reviews against 16 a year earlier. The quarter ending 31 December 2024 brought 164 against 33. The most recent monthly comparison available shows 102 against 12.

The local search turnaround in brief

A network in freefall, a measurement rebuild, and two years of growth that has not broken yet.

This account arrived in an eighteen month slide. Sixteen consecutive year over year windows came in below the prior year, the worst of them 22.5 percent down, and roughly 41 percent of the network’s visible presence in Google Search and Maps had already gone.

We rebuilt the measurement so the numbers could actually drive decisions, then ran every change as a documented test with a hypothesis, a validation method and a fixed read at 30, 60 and 90 days. Winners were rolled out across the whole estate. Losers were stopped and written down.

Two years on, the record reads: twenty four months with no negative reading, total impressions up 26.9 percent and then a further 14.1 percent, direction requests up 25.3 percent with every location in the network contributing, review volume up nearly seven times in the best month, and map pack grid coverage 27 percent better and holding. Every figure came from the client’s own scheduled reporting, cross checked against the location level detail behind it.