Three Businesses Show Up. Yours Is Not One of Them.

On a Tuesday in February a woman sat in her car outside a school with eleven minutes to fill. Her sister had sold a house badly the year before, and that morning, in the way people actually decide things, she had decided she was going to get this right. She typed four words into her phone.

Three businesses came back, with a small map above them and a row of stars under each name. She read them for about as long as it takes to read three names, tapped the second one, and left a voicemail.

You were not on that screen. You have never known this happened and there is nowhere you could go to find out. It is not a lost lead, because it never became a lead. Nothing arrived in the CRM because nothing was sent. The most ordinary way a stranger picks an agent produced no record of you at all.

The part worth knowing is that the order she saw was neither an accident nor a secret. Google publishes what decides it, in a paragraph almost nobody in this industry has read, and one of the three things it names is a fact about you that no amount of money or effort will change.

The search that already happened, and why you cannot see it

Every other way a client finds you leaves a trace. A referral comes with a name attached. A portal inquiry arrives with a timestamp. Even the person who does nothing at all with your open house has stood in a room you were in.

Local search leaves nothing. A person types a phrase into a phone, three businesses appear, they choose one, and the two they did not choose are never told. There is no impression count on a search you lost, no notification, no weekly digest of the times your name was not in the list. This is the only meaningful channel in the business where the failures are completely silent, and silence is why it is so easy to believe nothing is happening.

Something is happening. What you can see of it is one screen: the Business Profile's own performance report, which tells you how many people rang from the listing, how many asked for directions, how many tapped through to the website, and which searches surfaced you. It is not the whole picture and it is not a ranking report. It is the only genuinely first-party number in this entire subject, it is free, and in most small businesses nobody has opened it in a year.

It is worth separating two things that get called the same name. There is the map pack, which is the short list of businesses with the map above it, and there is the ordinary run of blue links underneath. They are ranked by different machinery and they are won in different ways. This article is about the first one, because it sits above everything else and because the person in the car never scrolled far enough to reach the second.

What Google actually publishes about this

Most of what is written about local rankings is somebody's inference from watching results move. The document underneath it is short, public and free, and it is worth reading rather than reading about.

Google's page on improving your local ranking opens the section on ranking with a warning rather than a technique. There is no way to request or pay for a better local ranking on Google, it says, and it says the algorithm details are kept confidential to make the ranking system as fair as possible for everyone. Anybody offering to place you is either selling something that does not exist or selling something Google would suspend you for.

Then comes the sentence the whole industry is built on top of. Local results, the page says, are mainly based on relevance, distance and popularity. Underneath, the three subheadings read Relevance, Distance and Prominence, and the third one has quietly changed names between the summary and the detail. That is a small thing, but it is the kind of small thing worth noticing on a page this heavily quoted, because the word people repeat is prominence and the word in the sentence is popularity, and neither one is defined anywhere with a number.

Relevance is described as how well a Business Profile matches what someone is searching for, and the advice for improving it is to provide complete and detailed business information. Distance is how far each business is from the customer who is searching, and when the customer has not said where they are, Google uses what it already knows about their location. Prominence is how well known a business is, and the page says it is based on information like how many websites link to your business and how many reviews you have.

That is the whole published model. Three inputs, no weights, no thresholds, and one sentence saying the details are deliberately withheld.

The input you cannot do anything about

Distance is the one that changes how you should think about all of this, and it is the one nobody selling local search wants to dwell on.

Your presence on the map has a shape. It is roughly centred on the address you verified, it fades as you move away from it, and it is competing against a different set of businesses in every direction. The person in the car was ranked against whoever was near that school. Fifteen minutes north, the same search produces a different three, and one of them may well be somebody you have beaten on service every time you have met them.

Google's guidelines for representing your business put a number on the outer edge of this, and it is the only number in the whole subject that is stated plainly. The boundaries of a profile's overall service area, the guidelines say, should not extend farther than about two hours of driving time from where the business is based. That is generous, and the very next sentence softens it further by allowing that larger areas may be appropriate for some businesses, so it is a boundary rather than a wall. What it is not is a marketing decision. A service area is a description of where you are.

So there is a limit built into this work, and it is honest to say it early. Local search will help you win the ground around you. It will not put you on the map in a town twenty-five minutes away where somebody else's office actually sits, because the thing being ranked is a business with an address and the address is one of the three inputs. That is a real gap, and the answer to it is a different surface entirely, which is an indexable page for each area you genuinely serve. That has its own rules, its own risks and its own article.

Why the top of a very short list is worth more than it should be

There is an assumption underneath everything written about ranking, which is that being first is worth more than being third for the sensible reason that people compare the options and the first one is usually the best. It is worth knowing that somebody tested that, and that it is only partly true.

A group at Cornell put 22 people in front of Google with an eye tracker running, and put a proxy between them and the search engine that could quietly rewrite the results page. Sixteen of them produced usable data. Some got Google's ordering untouched, some got the top two results swapped, and some got the whole page reversed. Nobody was told, and the paper records that when asked afterwards, none of the subjects had suspected any manipulation. A separate panel of judges then rated the results by how promising each one looked, without knowing what anybody had clicked.

The paper, published in 2005, reports the result as counts because the counts are small, and they are worth reading in that form. When the reader clicked exactly one of the top two links and the judges had rated the first one better, the click went to the top link nineteen times out of twenty, which is what you would hope. When the judges had rated the second one better, the click still went to the top link five times out of seven. And in the group where the pair had been secretly swapped, so the link on top was there for no reason at all, the click still went to the top link ten times out of seventeen.

The authors call this a trust bias and their conclusion is a single sentence: users have substantial trust in the search engine's ability to estimate the relevance of a page, which influences their clicking behaviour. Read that beside a list of three businesses on a phone. Some share of the first business's calls are arriving because it is first, and that share is not earned in any sense a person would recognise as merit. It is an inheritance.

What prominence is made of, and what it is not

Here the industry consensus and the document disagree with each other, and there is no polite way to put it, particularly since this website was on the wrong side of the disagreement until this article was written.

Ask anybody what moves the map pack and you will get three things: a complete profile, consistent name, address and phone details everywhere you appear online, and recent reviews. The middle item is the one that funds an entire category of software. It is not in Google's document. What the document names under prominence is how many websites link to your business and how many reviews you have.

That does not make directory tidying worthless. A phone number that is wrong in four places is wrong for the people who ring it, which is reason enough, and links and directory entries are not always separable. It does mean that a plan built mostly on submitting your details to ninety directories is a plan built on something the ranking document does not mention, while the two things it does mention are the two that are hardest to buy: somebody else choosing to link to you, and clients choosing to write about you.

That is an uncomfortable answer for a service page and it is the true one. The prominence half of local search is mostly a consequence of being genuinely present somewhere, and the work that produces it looks like sponsoring the thing your town does in September, being the person a local reporter calls, and asking every single client for a review rather than the pleased ones. Software can keep all of that current. It cannot manufacture any of it.

The profile rules that decide whether you can have one at all

Before ranking is worth thinking about, there is a shorter question that most agents have never checked, which is whether the profile they have is one they are allowed to have.

The same guidelines document names this business explicitly. Doctors, dentists, lawyers, financial planners and insurance or real estate agents are all listed together as individual practitioners, which is the category that gets a profile of its own. In the next breath the guidelines say that sales associates or lead generation agents for corporations are not individual practitioners and are not eligible for a Business Profile at all. Where you sit between those two sentences is a question about how you actually work rather than about what your card says.

The rest of the rules are the sort that get broken by somebody being clever. A practitioner is told not to hold several profiles to cover different specialisations. Where several public-facing practitioners share a location, the organisation gets its own profile and the practitioner's profile should be titled with only the practitioner's name, not the brokerage's. A solo practitioner at a branded location is told it is best to share the organisation's profile rather than start a competing one. And an address has to be an address: a rented mailing address you do not operate from is a virtual office and is not eligible, while a desk in a co-working space needs signage, staff during business hours and the ability to receive customers.

None of this is exotic and all of it is enforced by suspension rather than by a warning letter. A profile that disappears takes its reviews and its position with it, and the appeal is a form.

What local SEO actually does, week to week

Under the category name it is four unglamorous jobs, and the AI part is not the clever part. It is the part that does not get bored.

The first job is finishing the profile, which almost nobody has. Categories, every service written out, the service area drawn honestly, the hours including the strange ones, the address, the attributes, real photographs of a real place. Google's own advice for the relevance half is simply to provide complete and detailed information, which is an unsatisfying instruction precisely because there is no trick in it.

The second is keeping it true, which is where this decays. Businesses change quietly. A number moves, a service stops, an office is left, a holiday changes the hours, and none of those events tell the profile about themselves. A profile is at its most accurate on the day somebody fills it in and gets worse every day after that.

The third is feeding the two inputs the document actually names, which means a steady flow of reviews and the slow accumulation of other people's pages mentioning yours. The fourth is reading the report: how many calls, how many direction requests, which searches, this month against last. That last one is the difference between a service you can judge and a service you have to trust.

What renting the same attention costs

The obvious alternative to all of this is to buy the position, and the honest comparison is not that ads are bad. It is that almost nobody has measured what their ads are actually adding, and the one organisation that ran the experiment properly got an answer that surprised everybody.

In March 2012 eBay stopped bidding on search queries containing its own name on two search engines, while continuing to buy exactly the same terms on a third. That third one is the control, which is the whole point: without it you are comparing this month against last month and calling seasonality a result. Three economists then wrote up what happened in a working paper that was later published in Econometrica.

The naive comparison, before and after, said click volume was 5.6% lower once the ads stopped. Measured against the platform where the ads kept running, only 0.529% of the click traffic was actually lost, so 99.5% of it was retained. The people were still coming. They were simply arriving through the unpaid result sitting directly underneath the advert instead of through the advert.

The paper then does the same thing to the return on investment, and this is the number worth carrying out of it. Using ordinary regression on the observational data, the return came out at over 4,100% without controls and over 1,400% with time and geographic controls. Using the experiment, it came out at negative 63%, with a confidence interval that rejects the possibility of the channel yielding positive returns at all.

Be careful with this. It is eBay, a name tens of millions of people type deliberately, and the queries were ones containing that name. Nobody is typing your name, which is exactly why the brand-keyword half does not transfer. What does transfer is the shape of the mistake: the traffic that substitutes most cleanly for paid clicks is the traffic that was coming anyway, and no amount of dashboard staring will reveal that, because the dashboard is built from the observational data that produced the 4,100%.

What it costs, and how long it takes

The first month is mostly repair, and repair is priced by how wrong things currently are. A single profile that is two thirds filled in, one address, one set of hours and a handful of reviews is a short piece of work. Three agents, an office profile and two practitioner profiles that disagree with each other, a phone number that changed in 2023 and a service area drawn around a wish is a longer one, and most of the time in it goes on finding out what is true rather than on typing it in.

After that the cost is a monthly rhythm rather than a project, and what drives it is how many surfaces have to stay current and how much of the review asking you want handled rather than remembered. There is no software licence here that dwarfs the rest. The recurring number tracks attention.

The honest answer on time is that the two halves move at different speeds. Fixing a profile changes what people see immediately, because the profile is a record rather than a ranking, and an accurate record with real photographs and current hours converts better on the day it goes up. Position is slower and it is not promised by anybody, including us, for the reason printed at the top of Google's own page.

There is one cost that never appears on a quote and it is the one that decides the outcome. Reviews arrive because somebody asks, every time, including on the deals that went sideways. If nobody in the business is willing to do that, the plan is missing one of the two inputs Google actually names, and no amount of profile maintenance replaces it.

What it does not do, and should not pretend to

It does not buy a position, and this is not a disclaimer, it is a quotation. Google's page says in bold that there is no way to request or pay for a better local ranking. Any proposal that includes a promised position is either untrue or is describing something that will get the profile suspended.

It does not move you closer to anybody. Distance is one of the three published inputs and it is a physical fact. A business on the eastern edge of a county will lose searches on the western edge to somebody who is simply standing there, and the correct response to that is a different tactic rather than a better profile.

It does not survive a business that will not ask for reviews. Reviews are one of two things named in the ranking document, they cannot be bought without breaking policy, and the asking has to be systematic to produce a recent history rather than a cluster from 2023.

It does not fix a business that is hard to reach. A profile is a promise about hours, a phone that gets answered and an address a person can arrive at. Winning the search and then missing the call is an expensive way to fail, and the guidelines are explicit that a practitioner should be contactable at the verified location during the hours stated.

And it does not stay done. Everything above decays, quietly, in the direction of being wrong, and a profile nobody maintains is a profile slowly becoming a liability rather than an asset.

How to find out where you actually stand, in ten minutes

All of this is checkable tonight, for nothing, and the checking is more useful than any report anybody will sell you.

Open a browser you are not signed into, on a phone that is not yours if you can borrow one, and search the phrase a stranger would use for your service in the town you want to work in. Not your name. Write down the three businesses that come back. Do it again standing somewhere else in your market and write down the three that come back there. The difference between the two lists is distance doing its work, and it is the single most useful thing you will learn all week.

Then open your own Business Profile and read the performance screen. Calls, direction requests, website taps, and the searches that surfaced you, this month against a year ago. If you have never seen it, the number will either be higher than you expected, which means something is already working, or close to nothing, which means the listing is not finished.

Then go through the profile as if you were an inspector rather than the owner. Is the primary category the one a stranger would choose. Are all of your services written out. Are the hours right this week. Is the address one you actually sit in. Is the most recent review from this quarter or from two years ago.

Finally, count the reviews and count the closings you had last year, and put the two numbers beside each other. Most agents in this business are asking a small fraction of their clients, usually the ones who were already delighted, and the gap between those two numbers is the clearest picture of the prominence half you will ever get.

Common questions, answered honestly

What is local SEO for a real estate agent, in plain terms?

It is the work of being the business that comes up when somebody nearby searches for what you do. Most of it is not writing. It is a Google Business Profile that is complete, correctly categorised and actually true, a steady flow of recent reviews, other people's websites mentioning yours, and real pages for the places and services you cover. The AI part is maintenance rather than magic: keeping the profile current, keeping the review requests going out, and watching what the profile's own report says, which are the three things that stop happening the moment a human gets busy.

How is this different from getting my website to rank?

They are two different competitions and they are decided by different things. The map pack ranks businesses, and the inputs Google publishes for it are relevance, distance and popularity, one of which is where the searcher is standing. The blue links below it rank pages, and distance is not one of the inputs in the same way. Practically, that means your website can be excellent and you can still be missing from the short list at the top, and your profile can be perfect and your website can still be invisible for everything except your own name. Most agents need both. Finishing the listing is normally the smaller job of the two.

How do I get into the map pack?

Nobody can tell you how, because Google says out loud that it keeps the details confidential and that there is no way to request or pay for a place. What is known is the list of inputs it publishes. Finish the profile so relevance has something to work with, accept that distance is fixed, and work on the two things named under prominence, which are reviews and other websites linking to yours. Anybody giving you a more specific recipe than that is describing their own inference and should say so.

How long does it take?

The accuracy half changes the day you do it: a finished profile with the right categories, real photographs and current hours is immediately more convincing to the person reading it, whatever it does to position. The position half is slow and is not guaranteed by anybody. The honest framing is that you are not buying a date, you are removing the reasons you are currently not eligible for a place, and then continuing to be the sort of business that accumulates reviews and mentions.

Can I rank in a town my office is not in?

In the map pack, not really, and the reason is the distance input rather than any failing on your part. The other route is a normal web page, on your own site, covering what you have genuinely done in that town, because a page is judged by what is written on it. That is a separate piece of work carrying a separate risk, since a set of thin pages with the town name swapped is something Google's spam policy names specifically. Our article on area pages covers it properly instead of glossing it here.

Do citations and directory listings still matter?

They matter for being correct and they are not what Google's ranking page names. That page lists links from other websites and reviews under prominence, and does not mention directory consistency at all. Keeping your details right everywhere is still worth doing, because a wrong phone number is a wrong phone number, and because some of those entries are links. What is not defensible is a plan whose main activity is submitting your details to a long list of directories while the two published inputs go untouched.

Is this better than running Google ads?

They do different jobs and the honest comparison needs a measurement most people have never made. Ads deliver traffic the day you turn them on and nothing the day you turn them off. A ranking arrives slowly and keeps working. What the eBay experiment shows is not that ads do not work, it is that the returns most businesses believe they are getting come from observational data that overstates them by an enormous factor, and the only way to find out what yours are worth is to turn them off in one place and leave them on in another.

Can anybody guarantee me a position?

No, and the sentence to quote back is Google's own: there is no way to request or pay for a better local ranking on Google. What can be guaranteed is the work. A finished profile, a review request that actually goes out every time, the mentions that come from being present somewhere, and a report you can read every month. Position is the outcome of those, not a product anybody can sell you.

What to do about it

Go and be the woman in the car for ten minutes.

Borrow a phone, stand somewhere in your market that is not your office, and run the search a stranger would run. Look at the three names that come back and at how little information she had to choose between them. Then open your own profile beside it and see whether the thing she would have read is finished, current and honest.

If your name is in the three, the work is to stay there, which is duller and more important than getting there. If it is not, you now know something you did not know this morning, and it is not that you are worse than the second business she called. It is that a screen you have never opened is answering a question about you, over and over, to people you will never meet, and nobody has been checking the answer.