First 90 Days of Local SEO: Work, Evidence & Review

The first 90 days should leave your business with a clearer picture of which local searches matter, a set of visible improvements, and a record of who did what. You may also see early search or inquiry movement. But the first practical test of an SEO engagement is whether the work you bought can be inspected, understood and used.

A repair business paying for local SEO should be able to open an improved service page, see corrected information about its service area and know which types of inquiries are being counted. A practice with several offices should see a plan that reflects actual locations and appointment services. Neither should receive three months of colorful rankings charts with no way to tell what changed.

This guide is for owners evaluating a new retainer or preparing for a 90-day review. It is a way to accept the work as well as discuss the results. The schedule below is a useful framework, not a universal agency contract: a website with a serious indexing problem, a new location, or a complex approval process may need a different order. The deliverables and responsibilities should be agreed in writing.

The three checkpoints at a glance

A useful 90-day engagement moves from a shared baseline to implemented work and then to evidence-based decisions. Ask for documents and live changes that you can keep, rather than a report that only the provider can interpret.

Local SEO review checkpoints: day 30 links priorities to a baseline and approved scope; day 60 links delivered changes to live evidence and resolved blockers; day 90 uses comparable outcomes to continue, change or stop. — Each checkpoint connects work to evidence and a decision the owner can make. The detailed acceptance table below keeps the conditions and dependencies visible.

Three checkpoints for accepting agreed local SEO work
Checkpoint	What should be in your hands	What you can check yourself	What must be named if unfinished
Days 1–30: establish the assignment	Your priority services and actual service areas; current search and inquiry baseline; a ranked problem list; a page-by-page work plan; access and publication responsibilities	Can you identify the pages and services being prioritized? Can you see where the baseline numbers came from? Does each promised fix have an owner?	Missing access, unavailable conversion data, business-profile eligibility, content approval or developer capacity
Days 31–60: get changes live	A dated change register with URLs; completed or scheduled fixes; improved priority service content; necessary local-profile corrections; examples or previews of the next content items	Open the URLs, try the form on a phone, inspect the updated business details, and compare the completed work with the agreed plan	The blocker, who controls it, its business consequence, and the revised date
Days 61–90: review what happened	A list of shipped work, remaining gaps, comparable search and inquiry data, an explanation of what is still uncertain, and a specific next-quarter decision	Can the provider separate live changes from plans, search visibility from leads, and qualified inquiries from general contact attempts?	What will be stopped, changed, measured differently or escalated

If the contract only covers a narrow audit or a small Google Business Profile project, do not expect the same output as a full website and content retainer. Judge the written scope first. Then verify that the promised work exists.

Days 1–30: agree on what a useful local customer looks like

By day 30, you should have an agreed commercial assignment and a prioritized plan. That assignment needs more than a list of keywords. It should identify the services that make sense to sell, where the business can actually serve customers, what currently prevents discovery or contact, and who will approve the fixes.

Take a contractor who serves Brooklyn and parts of New Jersey. The valuable inquiry may be a large kitchen remodel within the travel radius, not every search mentioning a handyman. A dental practice may value a request for a particular appointment type, with a valid location and available capacity, more than general visits to a health article. If the agency cannot explain those distinctions, more traffic may not answer the business question.

What should the baseline include?

A sensible baseline uses the available sources and labels their limits:

Google Search Console: searches and pages generating impressions and clicks, with a defined date range and a useful distinction between branded and nonbranded demand. Impressions are not visits or leads.
Website analytics: actual measured visits and important page activity, if tagging and consent conditions make the figures reliable. A missing or duplicated event should be recorded, not quietly treated as an accurate baseline.
Inquiries: form submissions, answered calls or appointment requests that can genuinely be recorded. If no source attribution exists yet, keep the inquiries total separate rather than assigning it to SEO.
Local discovery: the eligible Google Business Profile's existing information, visible performance where available, and the relevant service/location searches actually checked. A single Maps screenshot is not a dependable market-wide ranking measurement.
Commercial context: which services have available capacity, what counts as a suitable inquiry, and who can respond quickly to the people who make contact.

The written baseline should say what was not measurable. That line is worth keeping. When conversion tracking is added in month two, the apparent increase in recorded leads might partly reflect better measurement rather than new customers.

What should the first audit actually decide?

Ask the provider for the highest-priority problems linked to the work you want. A business may have ten thousand minor crawl warnings and one important service page that cannot be reached from the website navigation. If the page is genuinely blocked from indexing, that may need to be fixed before adding articles. If the service page is indexed but leaves the most common customer question unanswered, rewriting it may be more useful than a technical redesign.

The audit should turn findings into decisions: keep, improve, create, combine or investigate, with a reason. It should also distinguish observations from guesses. A suspected conversion problem is not a proven lost-revenue figure.

By the end of the first month, the owner should be able to name the first two or three deliverables that will visibly change the business website. A large research deck is not a substitute for that answer.

Days 31–60: can the business open the work it paid for?

By day 60, the engagement should have moved from diagnosis to implementation, unless a specific dependency makes the agreed changes impossible. What goes live first depends on the website. For one company it may be correcting a broken contact form; for another, clarifying the business's principal service page; for another, fixing an incorrect location profile.

A useful month-two report should attach its claims to identifiable objects. "Improved the website" is hard to verify. "Updated the emergency repair service page on day 45 to clarify after-hours coverage and connect it to the booking route" can be checked. A provider can link to the live URL or show the version difference.

Website changes worth checking

For the services that matter most, look for clearer descriptions of what is offered, who qualifies, where work is performed, what customers need before contacting you and how they can ask for an estimate or appointment. Relevant related pages should help visitors move from a question to the appropriate service.

That sometimes means new content. It also frequently means improving what is already published. A local business with a vague primary service page can have a more urgent problem than a shortage of weekly blog posts. There is no useful rule that every industry needs a fixed number of city pages or articles.

When a provider says a technical fix has shipped, ask for the affected URLs and a simple explanation of the previous condition. Some changes require a developer, website administrator or hosting vendor. The work register should separate recommended, approved, implemented and verified, instead of marking all four as "complete."

Google Business Profile: only where the business is eligible

A local SEO proposal may involve category accuracy, services, hours, photos, the address or service area, and a legitimate process for requesting and responding to customer reviews. The work should reflect how the company actually operates. A traveling service business is not automatically entitled to create separate profiles for every place it serves.

Google's business representation rules distinguish staffed storefronts, service-area businesses and ineligible virtual-office arrangements. The owner should retain appropriate control of the profile. If a profile is ineligible, suspended or awaiting legitimate verification, the report must state that condition rather than claiming an optimization was completed.

For a multi-location organization, meaningful differences matter: real offices, relevant staff, available services, opening hours and a separate inquiry path when appropriate. Making identical location pages with different place names creates a larger website but does not resolve a customer decision.

What about reviews, listings and outside mentions?

These belong in the review when they belong in the agreed scope. For a listing correction, ask for the actual listing and the corrected detail. For review support, look at the request process and responses the provider was engaged to handle, rather than judging success solely by a promised number of new reviews. Genuine customer participation is not inventory an agency can manufacture.

Outside mentions need the same distinction. An outreach list, a message sent and an article published on another website are different outputs. Ask which one was contracted, where the published mention can be inspected, and whether a paid placement was involved. A submission awaiting someone else's approval should remain pending in the register. Counting every attempted contact as an acquired link conceals the work still left to do.

The acceptance question is specific: did the provider complete the activity it controls, report the actual external outcome, and explain the next decision? An impressive total without inspectable examples does not answer it. Nor does a small total, on its own, establish that the work was poor.

Who must supply the missing piece?

Business owners often have information no external agency can invent. A med spa must approve accurate procedure descriptions. A contractor has to confirm which projects and towns it will accept. A professional practice needs the right person to approve claims and booking language. If the agency can write but cannot publish, someone must own that final step.

Ask for the dependency next to the task, not hidden in email. A practical record might say:

Illustrative work register: delivery is not the same as verification
Item	Status	Evidence	Dependency	Next decision
Primary service-page revision	Implemented	Live URL and dated before/after copy	None	Verify the inquiry path on mobile
Booking-form fix	Approved, not live	Issue with reproducible failure	Owner's developer, scheduled date	Confirm completion after release
Business Profile category correction	Pending eligibility review	Current profile state	Business owner verification	Do not present as completed
Two proposed city pages	Not approved	Duplicate-purpose concern	Real local demand and service evidence	Improve existing service-area explanation instead

This is an illustrative acceptance register, not a claim about SEOTEC clients. Its value is simple: the owner can approve the invoice against the agreed work without having to become an SEO technician.

Days 61–90: review the business evidence, including inconvenient evidence

A day-90 review should answer whether the plan was implemented, what the available results show and what should happen next. A rise in visibility can be encouraging; an increase in qualified inquiries is closer to the real business goal. When neither has improved, the response should be an explanation and a revised hypothesis, not a convenient new chart.

Start with the shipped-work register. Which agreed changes are live? Which have not been approved? Which were removed because new evidence made them unnecessary? If most of the fee paid for an audit and implementation, those deliverables are measurable independently of search movement.

Then compare like with like: the same pages or query groups, compatible date windows and appropriate country/location context. A newly published page may have no useful historical baseline. Branded searches might increase because of offline advertising. Seasonal demand can change even when website quality stays constant. State those differences rather than mixing them into one victory percentage.

The SEOTEC guide to measuring SEO results explains how impressions, clicks, analytics sessions and inquiries fit together. Google's Search Console help describes the underlying click and impression metrics.

What if the rankings have barely moved?

First ask what was implemented and when. A site with important fixes live only ten days before the review has a different observation window from one improved throughout the quarter. Search response times vary, and a strong traffic result by a fixed day cannot be promised. Google explains that changes can take from hours to months to appear in search, and that some produce no noticeable effect. Its timing guidance supports separating the delivery review from the search-outcome review.

Second, inspect the relevance of what was pursued. More impressions for questions that will never produce local customers may be less useful than modest exposure for a high-value service. The agency should explain its page and query choices rather than only showing the largest numbers.

Third, assess whether the customer path is working. If people reach the service page but the phone number is hard to use or the form fails, additional search visibility might not solve the next problem. That is a conversion and operations issue, and it belongs in the review.

What if the reports look excellent but inquiries do not?

Ask to separate four layers: search appearances → search clicks → website actions → genuinely useful business inquiries. A percentage increase at the first layer does not prove the fourth increased. Not every call comes through a tracked website click; not every call is qualified; and a form event may count even when no customer reaches the business.

For an appointment-based practice, compare relevant appointment requests and how many were bookable. For a home-services company, separate jobs within its service area and work categories from general calls. A provider cannot manufacture those business records. It can agree a careful way to count them with the owner's consent.

If attribution is incomplete, document the gap and fix the measurement process. Do not assign untracked business growth to SEO as a fact.

Turn the day-90 review into a decision

We would leave the meeting with one of three decisions, and a reason attached to it. Continue when the agreed work exists, the evidence can be inspected, and the next proposed tasks address a useful opportunity the business can afford. Continuing does not require pretending that every search change has already become a customer.

Change the assignment when execution is visible but the original priority no longer fits. Perhaps the new service page attracts the wrong kind of inquiry, the business cannot take more work in that area, or a technical dependency has delayed the original plan. Name the replacement task, the displaced commitment and the next review point. More of the same is a decision only when someone can explain why it remains worthwhile.

Pause or arrange a handover when important work remains unexplained, required access cannot be resolved, or the business no longer has a useful assignment for the engagement. Check the actual notice and handover terms before acting. Preserve the working pages, accounts and records rather than removing useful assets to mark the end of a relationship.

An unresolved result sometimes calls for a smaller investigation, not a verdict on the whole agency. If contact totals rose after tracking was repaired, for example, review the measurement change before treating the increase as acquisition growth. Put that question and its owner in the next plan. The meeting has then produced something more useful than either automatic renewal or an argument about a single ranking.

What should you keep when you change SEO providers?

You should retain practical control over the website, content, data and local business presence your company relies on. That includes relevant administrative ownership, published files and content you have the rights to use, a record of changes and reporting access. The contract should define any exceptional ownership or licensing arrangements clearly.

A handover should include the page inventory and work register, key approved content files, analytics/Search Console and Business Profile access arrangements, current unresolved technical problems, and the next-quarter priorities. If the departing provider has a proprietary dashboard, ask for the underlying reports you are entitled to receive rather than depending on the continued life of that dashboard.

Be specific about external directory accounts and review-management subscriptions. Not every third-party tool account or advertising platform belongs to the business by default. Confirm what was created on the company's behalf and what can be transferred under the actual provider terms.

The SEOTEC guide to comparing SEO proposals discusses the ownership and exit terms to settle before a contract begins. A 90-day review is the point to confirm they work in practice.

Two local businesses can require different 90-day plans

Consider an established home-service company with strong reviews but weak service descriptions. If it already has a legitimate Business Profile and a usable website, a good early purchase may be service-page work, repairs to the inquiry route and better evidence of the jobs it wants. Spending the entire quarter creating dozens of generic neighborhood articles could delay that work.

Now consider a professional practice opening a genuinely staffed second office. The new location may need its own accurate operating details, appointment path and eligibility review. That is different work from adding more explanatory blog posts to the original location. The agency needs to know which practice areas or appointments each office actually provides before writing location-specific content.

In both situations the owner should receive comparable accountability, not identical deliverable quantities. The valuable common denominator is a plan that responds to the business, produces inspectable changes and adapts when the initial assumptions prove wrong.

Five questions to ask at the day-90 meeting

Show me the changes. Which work items are live? Which URLs or profiles changed, and what exactly was corrected?
Show me what the numbers measure. Are these impressions, clicks, sessions, calls, forms, qualified inquiries or actual booked work? Which periods and locations were compared?
Show me what did not happen. Which deliverables missed their dates, who controls the blocker and how will we resolve it?
Show me what you learned. Which services, pages or customer questions deserve more attention? Which parts of the original plan would you stop or revise?
Show me the next 90 days. What will be done, what decisions do you need from us, and how will we check completion?

An agency that can answer these questions with documents and real pages is easier to evaluate than one that asks for trust in a dashboard. That remains true when short-term search traffic is flat. It also remains true when numbers improve, because the business still needs to understand which work is worth continuing.

When a local SEO engagement is the wrong next purchase

An ongoing SEO retainer is not automatically the right response to every customer-acquisition problem. If the website's existing inquiries go unanswered, capacity is full, a booking form is broken, or the business urgently needs customers tomorrow, a different first investment may be more sensible.

Sometimes the company needs a discrete repair or a website improvement before ongoing content work. Sometimes paid advertising can serve an immediate demand while organic visibility develops. Sometimes the right project is to clarify its service offer. Our comparison of SEO, Google Ads and Local Services Ads covers that channel decision.

A sensible provider can say when it is not yet time to sign a full retainer. That judgment is part of what the business is paying for.

Use the 90-day review worksheet

Bring the agreed scope and the latest work record to the same meeting. For each item below, choose its actual status and record where the evidence lives. A public page address, a report name with its date range, or a short description of an agreed document is more useful than "done."

Keep delivered separate from verified. The provider may have supplied a page, while the owner has not yet checked the offer or contact route. Mark work outside the contract as not in scope; it is not an overdue commitment. For a blocker, name the person or team that controls the next step and what needs to happen.

The sheet is a working record, not an agency rating. Six completed boxes do not establish six useful improvements, and a blocked task may be explained by a real dependency. Use the notes to preserve that context. Save a copy for the next discussion so the original commitments do not quietly change between meetings.

Your working review sheet

No account or email is required. These entries stay in this page's memory and are not submitted to SEOTEC. Download a copy before leaving or refreshing. Do not enter passwords, personal customer information or confidential documents.

Day 30: scope and priorities

Priority services, real service areas and the first agreed pages.

Current status

Evidence: page, report or document