What’s Actually Inside a Monthly SEO Retainer

TL;DR: A retainer buys continuous compounding work — technical maintenance, content production, link acquisition, and the judgment calls in between. The problem is most retainer agreements describe hours or deliverables instead of outcomes, so you can be fully served and still not know whether it’s working. Here’s what should be in the scope, and what should be in writing.

Table of Contents

• What a retainer buys that a project doesn’t
• Hours, deliverables, or outcomes
• The line items clients never see
• What should be in writing
• Red flags in a retainer scope
• When a project beats a retainer

What a Retainer Buys That a Project Doesn’t

A project has an end. You buy a site rebuild, you get a site. You buy an audit, you get a document.

SEO doesn’t end, because the inputs don’t hold still. Google ships updates. Competitors publish. Your service mix changes. A page that ranked third in March sits eighth in July because two competitors rewrote theirs. Nothing broke. The ground moved.

That’s what a retainer is for: continuous work against a moving target. The compounding matters more than any single month. Twelve months of consistent publishing and technical hygiene produces results that twelve one-month sprints never will, because rankings respond to sustained signals, not bursts.

The honest version of the pitch is this: month one of a retainer is almost always the worst value you’ll get, and month fourteen is almost always the best. If you’re not planning to stay a year, a retainer is the wrong instrument.

Hours, Deliverables, or Outcomes

Every retainer is priced on one of three bases. Knowing which one you signed changes how you should manage it.

Hours. “Twenty hours a month of SEO work.” Transparent about cost, silent about value. You can burn twenty hours on anything. This model rewards agencies that work slowly and punishes ones that work fast, which is exactly backwards.

Deliverables. “Four blog posts, two optimized pages, one technical fix, monthly report.” This is the most common model and the easiest to verify. The weakness: deliverables get produced whether or not they’re the right deliverables. Four posts a month becomes a treadmill, and nobody stops to ask if post 37 should exist.

Outcomes. “Move these fifteen keywords into the top five; increase organic form fills by 40% over two quarters.” Rare, because it requires both sides to agree on measurement and both sides to accept risk. When it works, it works better than the other two. It only works when the agency controls enough of the variables — if your sales team doesn’t answer the phone, no agency can promise booked jobs.

The best retainers we’ve seen use deliverables as the floor and outcomes as the goal: here’s the minimum production you’re guaranteed, and here’s what we’re aiming those deliverables at this quarter.

The Line Items Clients Never See

Ask an agency to unbundle a retainer and you’ll usually get the visible half: content, page optimization, reporting. Here’s the rest of it, which is real work and real cost even though it doesn’t produce an artifact you can look at.

Monitoring. Someone checks whether your site is still indexed, still fast, still not serving a 500 error at 2am, and whether last week’s plugin update broke the contact form. Most months this produces nothing to show you. The month it catches a de-indexed service page, it pays for the year.

Competitive watching. Tracking who moved, what they published, and whether their gain came at your expense. This is where good strategy shifts come from, and it never appears on a deliverables list.

Algorithm response. When Google ships a broad update, someone has to determine whether your traffic change is you, the update, or seasonality, and then decide whether to act or wait. Deciding to wait is real work and produces no deliverable at all.

Internal linking and architecture upkeep. Every new page changes the shape of the site. Somebody has to keep the link graph coherent. This is quiet, unglamorous, and one of the highest-leverage things in the whole engagement.

Judgment. The decision not to chase a keyword. The decision to consolidate two posts instead of writing a third. The call that a competitor’s tactic is a short-term trick you shouldn’t copy. You are paying for this more than you’re paying for the posts, and it’s the one line nobody itemizes.

What Should Be in Writing

Insist on all six.

1. Production minimums. Specific counts of specific things per month. “Content as needed” is not a scope.
2. Response times. How fast someone replies when your site goes down or a ranking drops.
3. Reporting cadence and contents. What you get, when, and which metrics are considered primary.
4. Who does the work. Names and roles. Subcontracting is fine when disclosed.
5. Ownership. You own the content, the accounts, the tracking configuration, and the link data. If the agency owns your Google Business Profile or your Analytics property, fix that before you sign.
6. Exit terms. Notice period, what gets handed over, and in what format. A thirty-day notice with a full handoff is normal. A twelve-month lock with no handoff is not.

Red Flags in a Retainer Scope

“Guaranteed number one rankings.” Nobody controls the ranking. Guarantees on positions are either meaningless or attached to keywords nobody searches.

No named keywords or pages. If the scope never says which pages get worked on, there’s no way to tell whether the work happened.

Reporting as a major line item. The report is automated. The analysis is the value. If the line says “reporting: $400,” ask what the human does.

Link counts with no sources. “Ten links per month” without URLs means placements from a network. You’re buying risk.

The same scope for every client. Ask what’s different about your plan versus the last client they signed. If nothing is, you bought a template.

Everything is “ongoing.” Audits, keyword mapping, and citation cleanup are one-time. If they’re billed every month indefinitely, ask what’s genuinely being redone.

When a Project Beats a Retainer

We talk clients out of retainers regularly. A retainer is the wrong purchase when:

• Your site has a structural problem that has to be fixed before content matters. Buy the fix as a project first.
• You have fewer than about ten service-plus-city combinations. There isn’t twelve months of work.
• You can’t answer the phone. Fix operations before you buy demand.
• You want to bring it in-house within a year. Buy a project plus training instead.

Buying a retainer to solve a project problem is how businesses end up eighteen months in with a nice-looking report and no more jobs.

Quick Recap

• A retainer buys compounding continuous work; month one is the worst value and month fourteen the best.
• Hours-based retainers reward slowness. Deliverable-based retainers become treadmills. Outcome-based retainers require shared measurement.
• Roughly half the real work — monitoring, competitive watching, judgment — produces no artifact.
• Get production minimums, response times, ownership, and exit terms in writing.
• If the problem is structural or the site is too small, buy a project instead.

Frequently Asked Questions

What does a typical SEO retainer cost?
For a single-location service business in the US, most retainers land between $1,500 and $4,000 a month. Multi-location and competitive metro markets run higher. Below about $1,000 you’re generally buying maintenance rather than growth.

How long before a retainer shows results?
Expect early technical and local wins inside 60 to 90 days, and content-driven ranking movement between months four and eight. If nothing has moved by month six, the strategy is wrong, not the timeline.

Can I pause a retainer for a slow season?
You can, but understand the cost: publishing and link momentum decay, and restarting takes roughly as long as the pause. A better option is reducing production for the off-season while keeping technical monitoring live.

Should the agency or I own the Google Business Profile?
You. Always. Grant the agency manager access. Ownership of your profile, your domain, your Analytics, and your Search Console stays with the business.

What’s a fair notice period?
Thirty days is standard and reasonable. Sixty is defensible for large scopes. Anything requiring twelve months with no out clause is a financing arrangement, not a service agreement.