Google Business Profile Optimization: A Working Order

A locksmith spends a Sunday on his business profile. He uploads photographs, writes a description, adds every service he can think of, publishes an update. Three months later the phone rings about as often as it did before.

Nothing he did was wrong. It was done in an order that made most of it invisible. His primary category was still set to something broader than the work he actually does, so every field underneath it was answering a question nobody was asking of him.

That is the failure mode this guide is built around. Most advice on the subject is a list of fields, and a list has no opinion about what to do first. This one is an order, and after each step it asks the question the lists leave out: did it work, and how would you know?

Google Business Profile optimization — often shortened to GBP optimization — is the work of completing and correcting the fields on a business profile so they match how customers actually search, and then verifying whether the change made a difference. It is not a completeness score, and it is not a task you finish.

Profile work is the part of local SEO that sits almost entirely inside your own control. Your competitors’ review counts, the density of your district and the distance between you and the searcher are facts you inherit; the profile is the one asset you can rewrite this afternoon. That is why it comes first — and also why it gets overworked long after it has stopped being the constraint. Where it fits among the other ranking factors is set out in the Google Maps SEO guide; this page assumes that ground and gets on with the doing.

Why order beats completeness

Some fields are gates, others are shopfronts. The primary category is a gate: get it wrong and your profile is not considered for that family of searches at all, so nothing inside the shop is seen. Photographs are a shopfront — they help someone choose you after they have found you. Rearranging the shopfront while the gate is shut is work you will never be paid for.

And your time is the scarce input. Most owners can give the profile a couple of hours a week; where those hours go is the real decision, and a ladder makes it simple: always work on the lowest unfinished rung.

The rungs also differ in how fast they can possibly answer. Corrections near the top can surface relatively quickly, because they change what you are eligible for. The bottom rung speaks in months, because it is built from other people’s behaviour rather than yours.

The question this guide keeps asking

After each step below there is a short “Did it work?” block. It exists because the commonest way to waste a year on a profile is to keep changing things without ever establishing whether any of it moved.

Evidence comes in three shapes, and mixing them up is where self-assessment usually goes wrong:

1. Field-level. Did the edit save, and is it still there a week later? Some edits are reviewed before publishing, and some quietly revert.
2. Eligibility-level. Do you now appear at all for a wording you previously did not? Presence comes before position, and it is the earliest honest signal.
3. Area-level. Has your position across the area you serve changed? This is the only shape that answers did we gain anything — and it needs a reading taken before you started.

Two disciplines make all three usable: change one thing at a time, and write down the date you changed it. Correct a category and launch a review push in the same week and you have an outcome with two candidate causes and no way to separate them. How to take a comparable reading is covered in the Google Maps SEO guide; everything below assumes you have a baseline.

Rung 1 — The primary category

The primary category is the heaviest field on the profile, and the one most often set once, in a hurry, on the day the listing was created.

It works as a filter rather than a boost. It decides which searches you enter at all, which is why a close-but-broader category does not cost you a few positions — it enters you in a different race, where you look weak for reasons that have nothing to do with your business.

Choosing it, in this order:

• Name the job that pays the bills. Not the widest description of your trade, and not the ambition. If most of your income comes from one thing, that is the primary category; everything else is secondary.
• Use the customer’s word, not the trade’s word. Businesses describe themselves in the language of their profession; customers search in the language of the problem. Where the two differ, the profile follows the customer.
• Look at who is already there. Run the search you want to win, from inside your own area, and read the primary category the listings shown are using. It is the most direct evidence available about which category that search is attached to, and it costs nothing to check.

When nothing fits. Some trades genuinely have no exact category. Pick the nearest accurate one and carry the precision into the service list, which exists for exactly that gap. What you must not do is pick a busier neighbouring category on the theory that more traffic is better: you become a poor match in a crowded room instead of a good match in the right one.

Secondary categories are for services you genuinely provide at this address, today. The instinct to add more, on the grounds that each is another door, works backwards: a profile that keeps turning up for searches it cannot satisfy is a weaker candidate for the ones it can.

Treat a category change as an event. Note the date and change nothing else that week. A category change made alongside a name change and a batch of new photos is an experiment with no control.

Did it work? Look for eligibility before position. The honest signs: you begin appearing at all for searches you were absent from, and the competitors shown alongside you change — you are now being compared against businesses like you. If nothing on the map has moved after a fair interval on the same settings, the useful conclusion is that the category was not your constraint. That is a result, not a failure; it tells you to spend the next hour further down the ladder.

One check that proves nothing: searching your own business name and finding yourself. Branded searches find you almost regardless of category. The test has to be the service wording.

Rung 2 — Name, address, phone, hours

This rung rarely produces a dramatic jump. It prevents losses, and the losses it prevents are the kind nobody ever attributes to a profile.

The name is the name. The one on the sign, the invoice and the door. Adding a service or a city to it is a guideline violation and a bad trade: the upside is a position or two, the downside is the listing itself. The full argument is in the main guide; the operational point here is consistency — the same spelling on the profile, the website, the invoice and every social account. Where two versions exist, something has to decide which is correct, and you would rather that decision be yours.

Check the pin, not just the address text. A pin on the wrong side of a building, at the back of an industrial estate, or on the road rather than the entrance sends customers somewhere you are not, and each of them blames you rather than the map. Add the unit or floor if a stranger needs it. If customers never visit you — trades, mobile repair, cleaning, delivery — the address is hidden and a service area is defined instead, which changes the shape of the whole measurement problem and not just this field.

The phone number should be one that gets answered, in the same form everywhere you are listed. A separate tracking number means a second version of your own number now exists in the world; that has to be managed deliberately rather than discovered later.

Hours are the field that costs money directly. With no hours, a customer cannot tell whether the trip is worth making, and most will pick the competitor whose hours are visible. Worse than blank is wrong: showing open while the door is locked produces a wasted journey and often the review that comes with it. Two habits close this off — set special hours before public holidays rather than after, and update seasonal changes the same day.

Consistency elsewhere is an inventory job, not a technique. Search your own name, find everywhere you are listed, bring each into line with today’s details. Decide in advance which copy is the reference — your own site and the profile — so corrections have a direction instead of becoming a negotiation between old records.

Did it work? Not on a ranking chart; looking there will only tell you nothing happened. This rung is checked by inspection: read your listing as a stranger would, call your own number, follow your own pin to see where it lands. The evidence of success is the absence of a problem — nobody at the wrong door, nobody finding you closed — which is exactly why the work gets skipped. Do it anyway, and do it before spending money further down.

Rung 3 — Services, products, attributes, description

The category says what kind of business you are. This rung says what you specifically do, and on most profiles it sits empty.

Services. List the jobs the way customers name them, one line each. A garage that enters “car repair” and stops has turned a dozen possible matches into one; “brake replacement”, “clutch repair” and “diagnostics” are three separate things people search for. Two good sources for the wording: what customers say on the phone, and what your existing reviews already call the work. Both are your customers’ vocabulary, written down for free.

Products. If you sell things rather than only doing things, this is the only field that reaches brand- and model-level searches, which no category covers.

Attributes. Parking, step-free access, card payment, appointment-only, outdoor seating. Not ranking levers in any direct sense, but they decide whether you survive a filtered search — and being filtered out is a loss that never appears in a ranking report, because ranking never came up.

The description. One free-text field with a character limit, which rules out both usual temptations: neither a keyword dump nor an advertisement. The structure that works is plain — what you do and where, in the first sentence; the main services in ordinary language; then one concrete detail that separates you from the business next door. Claims like “the best in town” consume characters and convince nobody, because a reader cannot check them.

One caution saves a lot of disappointment: adding a service does not make you a candidate for a category you do not hold. This rung refines a match; it cannot create one.

Did it work? This is the sharpest test on the ladder, because you can aim it. Take one service wording you just added and search that exact phrase from a point inside the area you serve — not from your own counter. What you are looking for is presence: appearing anywhere for a phrase that previously returned nothing of yours is the change, even if the position is unremarkable. If the phrase still returns nothing after a fair interval, it probably belongs to a category you do not hold, and no amount of rewriting the description will change that.

Rung 4 — Photos

Photographs are the rung done first and understood least. Be honest about what they do: they mostly change whether you are chosen, not whether you are found. That is not a demotion — being chosen is the point of being found — but do not expect the map to move because you uploaded a gallery.

Aim for coverage, not a count. The set that does the work is short:

• The exterior, as someone arriving on foot would see it. The most useful photograph on the profile and the one most often missing: a customer across the street is trying to recognise a doorway, not admire your reception desk.
• The interior, so the visit is not a gamble.
• The team — a business with faces reads as a business that answers the phone.
• The work you produce — the finished job, the plated dish, the repaired part.
• Products or menu, where you have them.

Real photographs of the actual place, not stock. Stock is recognisable, and a customer who spots it starts wondering what else on the profile is decorative. Replacing inherited stock imagery is a better hour than adding ten more shots of the same room.

Freshness matters more than volume. A profile whose newest photograph is years old raises a question no image can answer. It is also worth checking the photographs customers have added — you cannot control them, but you can make sure your own set is not outnumbered and outdated by them.

Did it work? Not on a rank map; looking there produces a false conclusion in either direction. The instrument for this rung is behaviour — the profile’s own performance reporting shows how many people called, asked for directions or clicked through to the site. Compare an equal window before and after, and treat the result as directional rather than precise. And do not credit photographs with a ranking change just because they were the last thing you touched: that is the exact error a change log exists to prevent.

Rung 5 — Reviews

Reviews are the slowest rung and the hardest to copy, which are two sides of the same fact. The fundamentals — steady flow rather than campaigns, recency mattering as much as the average, asking for a review rather than for a rating, and why buying them backfires — are in the main guide. What belongs here is the operational side, because this rung fails for want of a routine rather than for want of understanding.

Decide who asks, when, and how. A request that depends on someone remembering stops within a month. A request attached to a moment that always happens — the invoice, the handover, the follow-up message — keeps running by itself. Pick the moment, give the team one sentence to say, and make the link short enough to be usable.

Reply to every review, and write the reply for the next reader. The customer who left it has moved on; the person reading your answer is a stranger deciding whether to call. Short, specific and human beats an obviously pasted template, and a run of identical replies is its own kind of signal. A calm, factual reply to a bad review does more good than the review did harm; an argumentative one does more harm than the review ever could.

Read your reviews as source material. Customers describe your work in the words other customers will type, and those words belong in your service list. This is the quiet loop between rungs three and five that most profiles never close.

Did it work? Judge this rung by rate, not by total. The question is not how many reviews do we have but did any arrive this month, and the month before. A total rises forever and so always looks like progress; a rate tells you whether the routine is alive. Keep this on a separate line from your ranking work — the two speak in different units of time, and mixing months-long prominence movement into a weekly comparison makes both unreadable.

Rung 6 — Posts, Q&A and the rest

The last rung is worth doing and worth doing last.

Posts. There is no reliable public evidence that publishing posts lifts ranking; treat any confident claim otherwise as a hypothesis rather than a fact. Their value is for the person already looking at the profile: a current offer, a seasonal note, proof that somebody is home. Posting weekly to an otherwise unfinished profile is decorating a shop with the gate still shut.

Questions and answers. This section is public, and if you leave it empty strangers will fill it — sometimes helpfully, sometimes not, always without your knowledge. Seeding it with the questions you genuinely get asked every week is both a service to the customer and a way to be the author of your own answers.

Category-specific extras — booking links, menus, ordering, appointment URLs — exist only for some kinds of business, and where they exist they are usually worth more than a post, because they shorten the distance between finding you and buying from you.

Did it work? Do not measure this rung on a map. The attribution risk outweighs the information: a post published the same week your position moved will take the credit for something further up the ladder, and you will then spend months publishing posts instead of fixing what actually worked. Judge these on whether the person on your profile got something useful.

Where optimization stops

A complete, accurate, well-worded profile guarantees nothing, and pretending otherwise sets up a disappointment that gets blamed on the wrong thing.

Profile work strengthens how well you match a search. It does not shorten the distance between you and the searcher, and it cannot manufacture prominence a competitor accumulated over years. So a well-optimized profile can still sit behind a nearer, older business in part of your area — and that is not evidence the work was wasted. It is the ladder telling you that you have finished the rungs you control.

A competitor with a thinner profile ahead of you proves nothing about profile work. They may be closer to the searcher, older in the area, or carrying more reviews. What their position does tell you is where not to spend: contesting the middle of somebody’s home ground is expensive, while the band just outside your current strength — where you appear, but not well — pays back the same effort faster.

Being in the top three is a question about places, not about your profile. You can hold a top slot in one neighbourhood and miss it entirely in the next, at the same minute, with the same profile. What determines that block of three is covered in the map pack guide.

If you run several locations, every rung repeats per address rather than per company, and pushing one set of categories, hours and photographs to all of them is exactly the mistake to avoid — see multi-location local SEO.

When nothing happens: a short diagnostic

Work through these in order; the answer usually turns up before the end.

1. Did the edit save? Open the profile as a visitor and confirm. Some edits are reviewed before publishing, and some revert.
2. Is the profile verified and in good standing? A suspended profile does not rank badly — it is absent, and no field work reaches it.
3. Are you comparing like with like? Same wording, same points, same device type, same spacing. Change a setting and the two readings answer two different questions.
4. Did you change one thing, or several? If several, you have an outcome without a cause, and the fix is procedural rather than technical.
5. Was the field you fixed the constraint? Rewriting a description will not rescue a wrong primary category. Go back to the lowest unfinished rung.
6. Has enough time passed for the rung you touched? Judging a review routine after three weeks tells you only that three weeks is short.
7. Where are you checking from? Your own address is the friendliest point on the map, and your own device has visited your profile before. A private window does not fix that: the window changed, the location did not.

Keep a change log — the cheapest tool here

One page, four columns, and it beats any amount of remembering: date · what changed · old value · new value. It costs a minute per edit and it answers, months later, the only question that matters — which of the things we did was worth doing? Without it you have a profile worked on for a year and no idea which hour of that year produced anything, so next year you repeat all of it, including the parts that did nothing.

Pair the log with a fixed measurement routine and the loop closes: read the area, make one change, note the date, read again on the same settings. The rule of the loop is that the settings never move. Everything else is negotiable.

A short audit list

• Is the primary category the narrowest accurate description of your main line of work, and do the secondary ones describe things you genuinely do at this address?
• Is the name the real-world name, spelled identically everywhere?
• Does the pin land where the door is, and are the hours current — including special hours set before the holiday?
• Are the services listed one job per line, in the customer’s words, with the attributes filled in?
• Is there a current exterior photograph a stranger could recognise you by?
• Does a review request happen automatically at a moment that always occurs, and is every review answered?
• Do you have a baseline reading of your area from before you started?

Summary

• Optimization is an order, not a checklist: the primary category first, extras last.
• A gate field decides whether you are seen at all; a shopfront field decides whether you are chosen. Do not decorate a shopfront behind a shut gate.
• Every rung needs its own kind of evidence — some are checked by inspection, some by presence in results, some by behaviour, and only some by a map.
• Presence comes before position. Appearing at all for a wording you were absent from is the first honest sign that something changed.
• Change one thing, write down the date, and keep the measurement settings fixed. Without a baseline, “did it work?” has no answer three months later.
• A complete profile makes you eligible to compete. It does not shorten distance and it cannot buy years of prominence.

The businesses that get somewhere with this are rarely the ones who worked hardest on the profile. They are the ones who worked on it in the right order, and kept a record of what they touched.