What a $2,000 SEO retainer should actually buy

A $2,000 SEO retainer is not cheap enough to be casual and not large enough to do everything. That is what makes it a useful test of an agency: can it choose the work that matters most, ship it, and show you the evidence without hiding behind volume?

The retainer should not buy “SEO activity.” It should buy a small operating system for search: a diagnosis, a prioritized roadmap, implementation, measurement, and clear monthly decisions.

This guide describes a healthy version for a small or mid-sized business site. Ecommerce catalogs, multi-location companies, migrations, international sites, and regulated markets can require more scope. The operating principles do not change.

The first month should buy clarity

Before publishing ten posts, the agency should be able to answer five questions:

1. What does the business sell, to whom, and where?
2. Which searches show buying or problem-solving intent?
3. Which existing pages should satisfy that intent?
4. What prevents those pages from being found, trusted, or acted on?
5. What is the highest-value constraint this quarter?

That requires access to the site, Search Console, analytics, the content management system, and any relevant local or ecommerce data. These accounts should belong to the business. The agency should work through delegated access, not create an analytics history you lose when the relationship ends.

A solid first month usually produces:

• a technical and content audit in writing;
• a query-to-page map for the priority demand;
• a list of indexation, canonical, redirect, internal-link, performance, and structured-data problems;
• a baseline of important pages and queries;
• a 90-day worklist, ordered by impact and dependency;
• at least one meaningful fix or page shipped.

The audit does not have to be a 90-page deck. It has to tell you what is wrong, why it matters, what will be done first, and what evidence will tell you whether that choice was good.

Every later month should buy shipped work

SEO compounds because assets and improvements remain after the invoice. Each month should add or improve something you can point to.

Depending on the constraint, that may include:

• a new service, category, comparison, or location page;
• a field guide that resolves a recurring buying question;
• rewritten titles, descriptions, headings, and copy on high-impression pages;
• internal links that connect useful answers to commercial destinations;
• consolidation of competing or thin pages;
• redirects and canonical repairs;
• crawl, sitemap, rendering, or indexation fixes;
• product or service structured data;
• image, template, or performance improvements;
• legitimate promotion to relevant partners, publications, or communities.

The exact count should vary. A technically difficult template repair can be more valuable than a batch of articles. A month with no visible pages can still be healthy if code shipped and the agency can show the before, after, and reason.

What should not vary is the ability to name the work.

The retainer should fund prioritization

A $2,000 budget has an opportunity cost. If the agency spends it on four generic posts, it cannot also repair the broken service-page template, merge six pages competing for the same search, and rebuild the conversion path.

So the roadmap should explain what was deliberately not done.

A useful monthly planning note might say:

• We improved the high-impression service pages before writing new guides because the site already earns discovery but loses the click.
• We fixed duplicate category URLs before outreach because new authority would have been split across competing versions.
• We deferred the blog calendar to build two commercial pages that the query map showed were missing entirely.

Those are decisions. “Ongoing optimization” is not.

The content should have a business job

The phrase “four blogs per month” is easy to price and easy to waste.

Before a piece is commissioned, it should have a defined role:

• satisfy a commercial search directly;
• answer an objection before a sale;
• support a category or service page with depth;
• earn internal or external links because it contains useful evidence;
• consolidate scattered information into one maintained answer;
• help existing customers use or choose the product.

The brief should name the search intent, the reader, the next useful action, the existing pages it should link to, and why this page deserves to exist separately.

AI can make producing undifferentiated copy cheaper. That raises the value of original evidence, operating knowledge, strong editing, and a page structure built around a real decision. The retainer should pay for judgment, not word count.

Technical SEO should include implementation

An audit that finds the same problems every month is a subscription to diagnosis.

The engagement should define who can implement each class of change. If the agency has repository or CMS access, it should ship agreed fixes. If development stays with your team, the agency should provide tickets precise enough to estimate and test, then verify the result after release.

At minimum, technical work should keep an eye on:

• whether important URLs are crawlable and indexable;
• duplicate and canonical versions;
• redirects and broken internal links;
• sitemap quality;
• rendered content and metadata;
• mobile usability and meaningful performance problems;
• structured data that accurately describes visible content;
• site changes that could erase traffic during a redesign or migration.

Google’s SEO Starter Guide is refreshingly plain about the job: help search engines understand the content and help users find the site and decide whether to visit. The technical checklist serves that job; it is not the job by itself.

Reporting should be one page before it is a dashboard

You are entitled to the underlying accounts. You should also receive a short interpretation.

Every month, the agency should answer:

What changed

Name the pages, fixes, internal links, tests, and promotions that shipped. Link to them.

What it did

Show the relevant evidence: discovery by query and page, qualified organic sessions, conversions, leads, revenue where measurement allows, and any losses that require a response.

What happens next

State the next priorities and why they follow from the evidence.

Impressions can rise before clicks. Clicks can rise before leads. Leads can rise while quality falls. A good note explains where in that chain the current constraint sits instead of choosing the biggest green percentage.

What the retainer probably cannot include

At this budget, a single engagement cannot credibly promise all of these at senior quality:

• daily publishing;
• national digital PR and media relations;
• a large recurring link quota;
• full conversion redesign and development;
• management of dozens of local profiles;
• international market and language expansion;
• deep analytics engineering;
• unlimited development;
• instant rankings.

Some may be added as projects or become the focus for a quarter. The red flag is not a narrow scope. It is a broad promise with no explanation of what receives the actual hours.

A healthy 90-day example

That is not a universal plan. It is what a retainer looks like when work, evidence, and decisions connect.

The five questions to ask before signing

1. Who will do the work? Ask to meet the person making weekly decisions, not only the salesperson.
2. What will exist after 90 days? Listen for pages, fixes, systems, and account ownership.
3. Who implements technical recommendations? An audit without a path to production is unfinished.
4. Which accounts do we own? The correct answer is all of the important ones.
5. What would make you tell us not to buy SEO? A useful partner can name the conditions where ads, conversion repair, product work, or nothing at all should come first.

The retainer earns its invoice when you can see the judgment, inspect the work, and understand the evidence. Rankings matter. The operating discipline that makes them more likely matters first.