Why Google Maps Rankings Change: Wave or Current?

Everyone who starts tracking their rank has the same second week: “Third yesterday, seventh today. What did I do?” Usually the answer is: nothing. A Google Maps ranking isn’t a fixed league table, it’s a live system, and learning to read it starts with learning what normal movement looks like.

Let’s set out the idea of this guide first:

Wave or current? A wave is the temporary movement a ranking makes within a day and between points: it’s on the surface, it comes back on its own and it needs no response. A current is a directional shift that persists across consecutive measurements with the same settings: it’s real, it has a cause and it needs a response. The quality of your local SEO decisions comes down to how quickly you can tell the two apart.

The cost of getting this wrong isn’t abstract. A business that mistakes a wave for a current breaks a setup that was working: it changes its category, rewrites its description, pays in a panic for a dubious “quick fix”, and when the rank comes back on its own the next period, it credits its wrong intervention with the success. A business that mistakes a current for a wave does the opposite: it watches a real decline for weeks thinking “it’ll pass”, misses the window to act and spends far more effort winning back the ground it lost. Both mistakes share one cause: an impression where a measurement should be.

First, the forces that create waves; then a method for catching currents.

Six forces that move rankings

1. Location: the biggest variable

The same search made at the same minute from two different streets returns two different lists, because distance is a structural input to local ranking. Some of the sentences that begin “my rank changed” are really sentences that begin “I looked from somewhere else”. That’s why tracking from a single point doesn’t work; a grid scan is the only approach that measures with the location variable held fixed.

Location doesn’t act at the same scale everywhere, either. In dense centres with many competitors, a few streets can reorder the list completely; in a sparse area the same business can hold a similar rank across a wide area. The practical consequence: “what’s my rank?” has no correct one-number answer. The correct answer is a surface: where you’re strong, where you weaken, and where the boundary runs. A report reduced to one number starts by throwing away the information you need most.

2. Device: mobile and desktop are separate tables

Mobile and desktop results can differ. Mixing the two into one series is like plotting the readings of two different thermometers on the same chart; that’s why our methodology measures devices separately and never blends them into one average.

The difference doesn’t come only from screen size: the length of the results list, the balance between the map and the list, and how people behave all diverge between the two surfaces. Because most local searching happens on mobile, a drop on the mobile table costs you differently from one on desktop.

3. Personalisation: the searcher’s glasses

The list a real user sees can be personalised by their search history and their interactions with profiles. Your phone shows “your” list, not your customer’s. That’s why measurement is done from a neutral user’s point of view; most of the gap between what you see yourself and what a scan shows is the difference those glasses make.

The sneakiest form of this force is an owner fooling themselves: because you search for your own business several times a day, that listing becomes familiar to you and easy to find in the list, and you read that as “I rank well”. A private window isn’t a full fix either: location, language and signals that don’t depend on your account stay in place. Your own phone isn’t a diagnostic tool; at most it’s a source of suspicion. Treat what you see as a question, not as data.

4. Competitor moves: a zero-sum game

Space in the top three is limited: a competitor rising means you falling, even if you did nothing wrong. The good thing about competitor-driven shifts is that they leave traces: a profile that has filled up, a faster flow of reviews, a changed category. When you see a drop, one of the first places to look is the competitor’s profile; the four-point check is in the competition branch of the diagnostic guide.

The second event that changes the competitive picture is a new player entering the field: a business opening in your area settles into the list as its listing is processed and pushes you down a place. That doesn’t mean your performance dropped; the number of seats on the surface stayed the same while the number of players grew. The distinction matters because the response is different: a performance drop needs repair, rising competition needs differentiation, through a precise primary category, a steady flow of reviews and a deeper profile.

5. Google updates: the ground shifting

Google updates its local ranking systems from time to time, and in those periods large-scale shifts happen. Nobody knows the exact calendar, and strategy isn’t changed on a rumour that “an update landed”. The distinguishing pattern: an update-driven shift usually shows up across several search terms, across the whole surface, in the same period; problems specific to you tend to break one profile’s picture.

The right reflex in those periods isn’t standing still, but not rushing. Big changes made while the ground is moving can’t be measured: you can no longer separate how much of the movement came from your change and how much from the update. If you see other businesses in your field moving in the same period, let things settle first; keep collecting your own data and decide afterwards.

6. Your own changes: the movement we want

A category correction, a fuller profile, a rhythm of reviews: these move the ranking too, which is the point. The problem is separating their effect from the noise of the other five forces. The tool that separates them is a measurement routine: change dates in a log, scans on the calendar with the same settings.

There’s a delay on top of that: the effect becomes visible as profile changes are processed, as new reviews accumulate and as behavioural signals build up, not on the day the change was made. That’s why changing five things in the same week is the most expensive habit in local SEO: whether the result is good or bad, “which one worked?” can no longer be answered. One meaningful change at a time, then a measurement: boring, but the only routine that works.

Three things that feed the six forces

The six forces push the picture around. But there are three movements that feed those forces, and much of the daily movement comes from them. They aren’t a seventh force; they are fuel for the existing ones.

The flow of reviews: speed is a signal, and so is stopping

Reviews aren’t a one-off target; they are an ongoing flow. When the flow speeds up the profile comes alive, and when it stops, that gets noticed too, and even if your count stays the same, your relative position weakens if your competitor’s count grows. How fresh and how continuous reviews are matters alongside the average rating. The trap here is starting a “review campaign” in a wave period and then dropping it: a sudden pile-up followed by total silence is a weaker pattern than a steady flow. Replying to the reviews you get is part of the same routine; the templates are in how to respond to Google reviews.

Profile changes: the most overlooked cause

When a ranking drops, outside causes come to mind first, yet the cause is often inside: a category was changed, opening hours were updated, an address was corrected, the service list was reworked, or an edit suggested by a user was applied to the profile. The last one matters most, because your profile can receive edits you didn’t make, and you only see them if you check the profile regularly. When you notice a drop, the first question shouldn’t be “what happened out there?” but “what changed on my listing?” The checklist is in Google Business Profile optimization.

Season and demand: the field changes, not the table

A season isn’t a ranking setting, but it changes demand, competition and search patterns. In season, more businesses enter the category, the number of searchers grows and the terms people use shift. Your rank can stay the same while your number of calls falls, or the opposite: your rank can drop while the phone rings more. That’s why reading rankings apart from business results misleads: the two are related but not the same thing. For seasonal businesses, the most valuable asset is last season’s measurement series; without a history to compare against, you can’t say whether this year’s movement is normal or not.

How to tell a wave from a current

The method has three rules, and all three rest on measurement discipline:

1. Look at the surface, not the point. Movement at one point is a wave; its diagnostic value is low. A shift in the same direction across the whole surface, with average rank and coverage moving together, is a candidate current.
2. Look at the series, not the scan. A single scan can’t decide that there’s a current. The rule is simple: one scan is an observation, two consecutive scans are a suspicion, three are a decision. A directional movement that persists for three consecutive periods with the same settings is no longer a wave.
3. Open a window on causes. When you spot a current, look at a two-week window in your log: your own change, a competitor’s trace, or a general shift? A current with no cause is rare; in a business that keeps a log, the cause is usually found within three lines.

All three rules share one condition: the measurements you compare must have been produced with the same definitions. A tool that calculates its average with penalties, or blends devices into one series, can make a wave look like a current; the five things to check in a tool are in the local rank tracker guide.

“The same settings” isn’t a loose wish here; it’s a list you can count:

• the same search terms: drop one term and add another, and the two periods can’t be compared;
• the same grid: the same centre, radius and number of points; making the surface bigger distorts the average on its own;
• the same device split: mobile against mobile, desktop against desktop;
• a similar time and day: don’t compare a weekday midday scan with a Sunday night one;
• the same calculation: how unfound points are counted and how the average is taken must not change between periods.

If those five don’t hold, what you have isn’t a series but a set of photographs that look alike and can’t be compared, and every “trend” drawn from them is made up.

How much movement is normal?

There’s no numerical answer that fits everyone, and anyone who gives you a number is guessing. But what makes the normal band of movement wider or narrower is clear:

The practical reading: a few positions of movement at the edge of the surface is not news; the same movement at the centre is. When you judge a drop, first ask “where did it move?”, because the same size of movement means different things in different parts of the surface.

Before you hit the panic button: a 60-second check

When your rank moves, in order:

1. Question the source. Is the observation from one point (your phone) or from a measurement (a grid)? If it’s your phone: note the observation, wait for data.
2. Check the surface. Is the shift general or regional? If regional, which neighbourhoods?
3. Check the series. How many periods has it lasted? If this is the first: note it, wait.
4. Check the profile. Is there a warning, a suspension or a verification problem in the dashboard? If so, the issue isn’t ranking; go to Branch A of the diagnostic guide.
5. Check the log. Is there a change or an event in the window? If not, look at competitors’ profiles.

Those five steps kill most unnecessary local SEO decisions, such as changing a working category in a panic or signing up for an “urgent fix”, before they’re born.

If after the fifth step you still can’t find a tangible cause, what to do is clear: nothing. Measure for one more period. Intervening against a movement with no cause pollutes the next period’s data too; after that you can know neither the cause of the drop nor the cause of the “recovery”.

Signs of a real decline

A current, a real decline, has characteristic traces. If several of these appear at the same time, it’s time to drop the wave hypothesis:

• It holds a direction. Consecutive periods move the same way, with no bounce back.
• It spreads across the surface. Not just a few edge points; points near the centre are being lost too.
• Coverage shrinks. The boundary of the area where you appear in the top positions is drawn inwards; it gives an earlier warning than average rank.
• It happens across several search terms. Movement on one term can be specific to that term; movement on all of them points to a structural cause.
• It shows on both devices. Movement only on mobile or only on desktop is another story.
• Competitors have a matching gain. If a particular competitor is taking the places you’re losing, the cause is in the field.
• It lines up with an event on the profile. It coincides with a category, address or hours change, or a warning notice.

By contrast, these are not proof of a decline on their own: one search from your phone, a change at one point, a one-day deviation, or an acquaintance saying “I can’t find you”. All of them can be explained by personalisation and location; they trigger a measurement, not a decision.

When to act, and when to wait

Not every signal calls for the same response. The decision table:

When you do decide to act, the order matters too: first the basics (listing accuracy, category, address, hours), then depth (services, description, photos, posts), then flow (the rhythm of reviews and replies). Skipping the basics to work on depth is rowing faster in a leaking boat. The whole order is in the Google Maps SEO guide.

Why the measurement rhythm matters so much

Everything above rests on one assumption: that you have a comparable history. Without it, none of the rules in this guide work, because “wave or current?” is by definition a question about a series. A single measurement has no answer.

The rhythm does three jobs:

1. It creates a reference. You only know whether today’s picture is good or bad by comparing it with yesterday’s. Starting to measure on the day you see a drop means having no reference at exactly the moment you need one most.
2. It makes changes assessable. A piece of work with no measurement before and after teaches nothing, whatever the result; “I think it worked” can’t feed the next decision.
3. It gives early warning. Currents usually start at the edges. A business that measures regularly sees the loss at the edge points and acts before it reaches the centre; one that doesn’t notices the same loss when the phone goes quiet, at the most expensive point.

More important than how often you measure is how consistently. The same search terms, the same grid, the same device split, a similar time of day. If you want to see quickly where you stand, set up a first scan as described on the rank checker page; building a series, comparing period by period and following coverage is the job of regular scans, which is what Auto Tracking runs on a schedule.

One last warning: be careful with services that sell fluctuation itself as “the problem”. An offer that promises to freeze your rank either doesn’t know how the system works or knows and isn’t telling you.

Summary

• Google Maps ranking is a live system; movement isn’t a fault, it’s normal behaviour.
• Six forces create the movement: location, device, personalisation, competitor moves, Google updates and your own changes.
• The flow of reviews, profile changes and seasonal demand feed those forces.
• A wave is temporary and needs no response; a current is directional, persists and needs a response.
• The three rules for telling them apart: look at the surface not the point, the series not the scan, and open a window on causes.
• The normal band of movement depends on how dense the competition is, where on the surface you look and how broad the search term is; there’s no number that fits everyone.
• The decision to act comes from three periods of directional movement; the exceptions are profile warnings and basic data errors.
• All of this rests on one habit: measuring with the same settings, at regular intervals.