Do this on your phone, not the office desktop, because that computer knows who you are and will flatter you. Search your trade and your town. Above every website there is a small map and three businesses.
That box is the map pack. For a homeowner who wants a floor measured this week it is most of the page, and if you are not one of the three you are not fourth. You are behind a tap almost nobody makes.
The useful part, buried under everything sold as local SEO, is that Google publishes how the three get chosen. Almost nobody sells against those factors, because two are slow and the third cannot be sold at all.
The three things the box is actually judging
Proximity, relevance and prominence, with the column that matters most: how much of each you can move.
| Factor | What it means in plain words | How much you control it |
|---|---|---|
| Proximity | How far your address is from the person searching, or from the place they typed | Almost none. It is geography |
| Relevance | How well what your listing says you do matches what they typed | Nearly all of it, and most of it in one afternoon |
| Prominence | How well known the business looks. Reviews, the rest of the web, your own site | All of it, slowly. This is the one that compounds |
That table reframes the job. You are not trying to rank in your city. You are trying to rank in the neighborhoods around your address, and to earn the ones further out with pages and prominence.
How far the box reaches depends on how crowded the market is. Eleven flooring companies inside a short drive and it pulls from a tight circle. One shop for forty minutes and it reaches a long way, because there is nothing closer. Anybody quoting a mileage number is guessing.
Pick one primary category, then stop editing it
The primary category is the biggest relevance lever on a profile, and the field contractors flip back and forth on for years. Two agencies will tell you two opposite things. The rule, stated once:
The primary category describes what your address actually is, not the range of work you sell.
- A homeowner can walk in and stand on samples? You are a flooring store. Pick the store category.
- Your address is a shop, a yard or your own house, and nobody visits? You are a flooring contractor. Pick the contractor category. Claiming a storefront you cannot show anybody invites a hard look you do not need.
Secondary categories carry the rest of what is true. Hardwood refinishing, tile, carpet, cabinets, bathroom remodeling if you genuinely do it. Do not add what you do not do, because a category you cannot service books appointments you have to cancel.
Then leave it alone. Every change resets how the listing gets read. The site side of this argument is in the flooring SEO guide.
Address and service area, and the setting installers get wrong
This is where a profile breaks quietly, and the failure is worse than a bad ranking. It is a suspension mid season.
- Storefront. Keep the address showing and the hours accurate, including the holiday ones nobody updates. Let it be a place. That is the advantage a showroom has.
- Installer with no showroom. Hide the address and set service areas instead. Google's guidelines say a business that does not serve customers at its address should hide it. A home address with no sign and a van in the driveway is the most common reason a contractor listing goes down, and reinstatement takes weeks you cannot afford in July.
The service area list is not a wish list. Name the towns you will actually drive to. Covering half the state does not make you rank in half the state. Proximity is still doing that work, and a long list only makes the profile less specific.
One listing per real location, meaning a sign, a phone number and staff. While you are in there, check whose account owns the profile. If it is your marketing company, fix that this week, for the same reason you own your ad account.
Reviews are the only asset here that compounds
Everything above is setup. Reviews are the part that keeps paying, and the part most contractors do by accident. Three things get mashed together under the word:
- Count. The total. Slowest to move, and the hardest for a competitor to catch up on.
- Velocity. The pace they arrive. Twenty in one week after two silent years looks exactly like what it usually is.
- Recency. A wall of five stars, all from three years ago, tells a homeowner the good crew left. Readers see that in two seconds.
An illustration, using round invented numbers. Not our results and not a projection. Put your own job count and your own ask rate in the same slots.
- You finish 12 jobs a month, and you ask all 12.
- Say 4 in 10 of the people you ask actually write one. Call it 5 a month.
- That is 60 in a year, arriving steadily, from the neighborhoods you worked in.
Now put that next to an imagined competitor, same invented numbers, sitting on 90 reviews with nothing since last spring. Inside two years the count is gone, and long before that the homeowner reading both profiles knows which company is still busy.
How to ask without getting the reviews deleted
Three rules, enforced unevenly, which is why people get away with breaking them until they do not.
- No incentives. No discount, no gift card, no draw entry, not for a review. This is the one that removes a year of reviews in a single batch.
- No gating. Do not ask how it went and then send the link only to the people who said great. Same ask, everybody.
- Nobody writes them but the customer. Not you, not the office, not a vendor selling packs. Those get found, and the profile pays for it, not the vendor.
What works is embarrassingly simple. Ask in person, at the moment. The installer standing on a finished floor while the homeowner says it looks amazing is the best moment there is, and most companies waste it by having the office email four days later. He asks right there, then sends the link from the driveway.
Build the ask into the job, not into somebody's memory, because memory loses to a busy Friday. Then reply to every review, short and specific, inside a couple of days. Replies show a reader that somebody is home, and they are the only place a bad review stops being the last word.
Name, address and phone, without buying the citation packages
Same name, same address, same phone, written the same way everywhere. Suite number formatted identically, Street or St picked once and kept, one number that rings the same desk. The reason is mechanical: Google is matching records from many places to satisfy itself they describe one business, and contradictory records make the match weaker.
Where it matters, in order:
- Your own site. Footer and contact page, marked up so a machine can read it and not only a person.
- The Google profile itself.
- The big data aggregators. The handful of records other directories quietly copy from.
- Apple Maps. It is what an iPhone opens by default, and half your market is holding one.
- Two or three trade listings you belong on. Your manufacturer's dealer locator beats fifty generic directories, because a homeowner shopping that brand opens it.
Not worth it: the cheap package that promises to blast you into hundreds of sites. Most are dead pages, and a year later you are paying somebody to clean up records you paid to create. If you change numbers, update the short list the same day and keep the old one forwarding for a year.
Photos and products, and the myth attached to them
Photos, and not stock ones. Job sites, before and after from the same angle so the after reads as an after. Your crew, your trucks, your shop from the road as a driver sees it.
Cadence beats volume, and the reasoning is plain. A profile touched last week reads as an open business. One touched last March reads as a company that might not answer. Ten to twenty photos a month is plenty if they are real, which is about one job's worth. Location data inside the files does nothing, so ignore anybody selling geotagging.
The services list, or the products list if your profile shows that instead, is free relevance almost nobody fills in. Write the real lines out: hardwood installation, engineered plank, luxury vinyl, tile and stone, carpet, stair treads, subfloor repair, cabinets and bathroom remodeling if you do them. Two plain sentences each. An hour once, and the listing knows what to match.
The competitor whose listings do not exist
Every market has one. A flooring listing at a mailbox counter. A name that reads like a search query. A lead generation company holding a dozen listings across the metro so it can sell the calls to whoever pays.
They break the guidelines and they will outrank you until somebody says something. Do it without losing a week:
- Document first. Screenshot the listing, pull the address up on street view, and write the name down exactly as it appears, with the date.
- Use the edit control on the listing itself for the simple cases: a name stuffed with keywords, an address that is a shipping counter, a business that closed.
- Escalate the volume cases. Google runs a separate complaint channel for owners reporting several at once. It wants evidence attached, which is why step one exists.
- Expect slow and partial. Some come down in days, some never do. Log what you filed.
Honest sizing: an afternoon a quarter. The competitors beating you fairly are where the work is.
What a month of honest local work actually looks like
Not a ninety day roadmap with twelve workstreams. Four weeks, four jobs, then the same four forever.
| Week | The work | Why it sits here |
|---|---|---|
| One. The setup pass | One primary category. Secondaries that are true. Hours including holidays. Address shown or hidden correctly. Services written out. Site details matched | Almost all of your relevance, done once. The pass nobody does properly |
| Two. The review habit | Write the ask script. Put the link in a text template. Teach every installer the moment to use it. Reply to everything unanswered | The only compounding asset here, and a habit not a task |
| Three. The pages | One page per real service and one per real town you serve, written by somebody who has been there | The map handles the neighborhoods near you. Pages are how you show up in the ones that are not |
| Four. Photos and the check | Photos from this month's jobs. Answer the questions on the profile. Report obvious fakes. Read the profile's call and direction counts | Small, repeatable, and the week that gets skipped first when things get busy |
The expectation, as a range with the reasoning attached rather than a number pulled from the air: 60 to 90 days. Profile edits get re-read over weeks, not days. Review signals build at the pace you finish jobs. And a three slot box shuffles on its own, so one good Tuesday is noise. A month of this is not a test. A quarter of it is.
The shortest version
One primary category, chosen from what your address really is, then left alone. Address hidden if nobody visits it. A review asked for on every job by the person standing on the finished floor. Name, address and phone identical on the short list that matters. Photos from this month.
Keep it in proportion. The map pack is the cheapest inquiry you will ever get and the least controllable thing you own. Run it because it is nearly free, and do not let it be the reason nothing else is running in April. If you need estimates booked this month, that is a paid channel and a phone that gets answered, and the wider menu is in how to generate flooring leads. Judge what you spend here by what a booked estimate ended up costing, not by rankings, which is the argument in cost per booked estimate.
