What a Local SEO Retainer Actually Includes Each Month

A local SEO retainer worth paying for contains four recurring workstreams — Google Business Profile and review generation, location and service pages, citation and listing consistency, and technical health — plus reporting that ties back to calls and booked jobs rather than keyword counts. Everything else is either a project or a rebrand of one of those four. Google’s own local ranking guidance names relevance, distance and prominence as the three things the Maps pack weighs (Google Business Profile Help), and only two of those are things an agency can move.

That is the useful frame for reading any proposal you are handed. Distance is fixed by where your business sits. Relevance and prominence are the products you are actually buying, and a retainer that does not visibly work on both is priced for something it is not delivering.

What should be in the monthly scope?

Four things, every month: Google Business Profile management and a review generation process, new or rewritten service and city pages, directory and citation cleanup, and technical maintenance. A fifth — reporting on leads rather than rankings — is what lets you judge the other four. If a scope of work does not name all five, ask what replaced them.

The reason to insist on all four rather than the fashionable one is that they gate each other. Content on a service page cannot rank if the site is slow and half-indexed; a perfectly maintained profile will not convert if the page it sends people to does not mention the city they searched. In practice the sequencing matters more than the list. Prominence work — reviews, listings, mentions — is the slowest to move and sets the ceiling, so it should start in month one even though it reports last.

Here is the uncomfortable version of that claim, from our own property. Over the 28 days to 21 September 2026 our Provo page drew 936 Search Console impressions at an average position of 22.8, up twelve positions in a single window, and our Salt Lake City page drew 1,196 at 23.1. Both improved sharply. Both produced zero clicks, because positions 16 to 23 are the top of page two and page two does not get clicked. On-page work took us that far and stopped, exactly where the prominence ceiling says it should.

How do you tell a real retainer from a thin one?

Look for deliverables that name a quantity and an owner. “Ongoing SEO optimisation” is not a deliverable; “two service pages rewritten, fifteen citations corrected, a review request sequence running on every completed job, reported monthly against booked calls” is. Vagueness in a scope of work almost always survives into the invoice.

The second test is whether the agency will tell you something you do not want to hear. A good diagnostic call ends with at least one finding that reduces the scope — that your real problem is a missing phone number on your profile, or forty reviews you never asked for, rather than a content programme. When we sampled ChatGPT on 21 September 2026 with a Utah buyer question, the model volunteered a version of this test unprompted, advising the buyer to demand “examples where they’ve actually increased AI visibility for a service business, not just an ecommerce brand.”

The third test is measurement. Fewer than one in three Google searches still sends a click to the open web, according to SparkToro’s 2026 analysis, which means a report built on sessions alone will increasingly understate work that is going well and hide work that is going badly. Calls, form fills and booked jobs are the only figures that survive that shift.

Line item	What good looks like	What thin looks like
Google Business Profile	Weekly posts, Q&A, category and service audits, photo cadence	”Profile optimised” once, in month one
Reviews	A request sent on every completed job, replies within days	A one-off email blast to an old list
Pages	Named service and city pages, rewritten on a schedule	A monthly blog post on a general topic
Citations	An audited list, corrections tracked to completion	A bulk submission to 300 directories
Reporting	Calls, forms and booked jobs by source and city	A rank-tracker screenshot

Why do reviews keep coming up first?

Because reviews are now the entry gate for two separate surfaces at once — Google’s Maps three-pack and the business panels AI assistants put at the top of local answers — and the entry bar is far lower than most owners assume. A handful of genuine reviews buys eligibility; volume after that buys frequency.

Both halves of that are measurable, and we measure them on ourselves weekly. On 21 September 2026, Google’s live local pack for “seo salt lake city” gave its top seat to a firm carrying a single five-star review, ahead of competitors holding 74 and 27. The same day, ChatGPT’s answer to a Salt Lake and Provo buyer question opened with a fifteen-business panel drawn from Google profiles, and the smallest firm in it had five reviews. Flownomic has none, which is why we were absent from that panel in the same answer that named us in its written shortlist — a gap we publish rather than hide.

The volume half is documented outside our log. Trustpilot research reported by TechRadar Pro (2026) found that businesses with large review counts appeared in the great majority of relevant AI answers, against a tiny fraction for businesses with none. Read the two findings together and you get a sequence rather than a target. The first few reviews change your eligibility; the next fifty change how often you are chosen.

What a retainer must not do is manufacture them. Buying, incentivising or gating reviews breaches Google’s review content policy and gets profiles stripped, which costs more than the reviews were worth. The work is asking real customers, at the right moment, every time — which is unglamorous, which is why it is usually the line item that quietly disappears from a cheap proposal.

What about AI visibility — is that a separate retainer?

It should not be. The inputs that make an AI assistant name your business are mostly the same inputs that win the map pack: a complete profile, real reviews, consistent listings, and pages that state plainly what you do and where. Sold as a standalone product, AI visibility is usually the local SEO you already needed, priced as something exotic.

We hold that view against our own commercial interest, and the assistants agree. Asking four of them the same Utah buyer question on 21 September 2026, one told the buyer that “for a small business, hiring an agency that handles both classic Google organic optimization and generative search is generally more cost-effective than hiring a standalone GEO shop.” Another named Flownomic and then filed us as the specialist you would hire only if AI visibility were the main reason — advice we think is right in general and wrong about us, which is a useful reminder that an assistant’s summary of your business comes from what your own pages say.

There is a real technical layer on top, and it is worth naming so you can check whether you are being sold it or sold around it. Structured content, entity-clear copy, and evidence density genuinely change citation likelihood: the Princeton study that named generative engine optimisation found that adding statistics, quotations and cited sources lifted visibility in generated answers by up to 40% (Aggarwal et al., 2023). That is a content-quality intervention, not a separate discipline, and it belongs inside the retainer you are already paying for.

What can a retainer not fix?

It cannot fix distance, it cannot buy authority quickly, and it cannot make a thin market bigger. Those three limits explain most of the gap between what a proposal promises in month one and what shows up in month nine.

Authority is the one that surprises people, because it is the constraint that binds after on-page work is done. For the Provo agency query family, page-one results carry an average of roughly 7.9 referring domains in Google’s keyword data; the Salt Lake City equivalents need 22 to 25. Those are very different mountains, and an agency that quotes you the same timeline for both has not looked. Market size is the other: some city-plus-service combinations genuinely have ten searches a month, and the honest recommendation there is to spend the money on reviews and referrals instead of a page.

Google’s own guidance is blunt about the direction of travel here — its Search Essentials put people-first, genuinely useful content at the centre and treat scaled, low-value pages as a liability. A retainer that responds to a small market by producing more thin city pages is not just wasting your money; it is accumulating risk across your whole site.

What we’d actually do

If we were buying this for our own service business, we would ask for a scope naming quantities against all five workstreams, insist that the review process start in week one regardless of what else is in the plan, and require the monthly report to show calls and booked jobs by city rather than positions. Then we would hold the agency to one number for the first ninety days: how many genuine new reviews landed. It is the cheapest lever, it gates both the map pack and the AI panel, and it is the one an agency cannot fake on a slide.

If you want to see the rest of our first-party record before deciding anything, it is on our results page, and our own local work is described on the local SEO service page and the Utah hub. For what shapes the price, see what AI visibility costs and our published pricing. If you serve more than one city, Provo SEO vs Salt Lake City SEO covers how we handle service-area intent.

Ready to have that conversation? Book a 30-minute call — no charge, no obligation.