The agency that stopped hiring to keep up

At a glance

• Retainer clients supported — 100

• New sites checked — ~10 per month, pre- and post-launch

• Monthly consumption — ~300 hours

• In-house SEO hires required — 0

• Cost vs. two equivalent US hires — 28%

The client

A full-service digital agency in the United States. They sell websites and they sell SEO retainers, and the two are wired together: a new site is the front door, the retainer is the relationship. Their reputation rests on both being right.

The challenge

Their SEO book had grown to around 100 retainer clients. On top of that, roughly ten new websites went live every month — each needing a pre-launch technical check and a post-launch verification pass before it could be handed over with a straight face.

That is not one job. It is three jobs braided together, and they compete:

• Pre-launch checks are deadline-bound. A site launches on a date the client has already been told. The check either happens before that date or it doesn't happen.

• Post-launch verification is urgent but invisible. Nobody asks for it. But a redirect map that silently drops a third of a site's equity turns into a furious phone call six weeks later, and by then the damage compounds daily.

• Monthly retainer work is relentless. A hundred clients, each expecting movement and a report, every month.

Whenever a launch week got heavy, retainer execution slipped. Whenever retainer work was protected, launch checks got compressed into a rushed skim. The team was consistently choosing which client to underserve. Nobody was doing bad work — there simply wasn't enough of the right kind of hour in the month.

The obvious fix was hiring. It was also the wrong fix. The workload wasn't flat — it spiked around launch clusters and quarterly retainer pushes, then subsided. Hiring two specialists to cover the peak meant carrying two salaries through the troughs, plus recruitment, plus onboarding, plus the risk of getting the hire wrong.

The structural problem: their capacity was fixed and their demand was lumpy. Every agency owner reading this knows that shape.

It started with 50 hours

Like every ACX agency relationship, this one began with the minimum commitment — one month, one block of hours, no restructuring required to find out whether it worked.

That matters more than any argument we could have made. An agency that has been burned by a vendor doesn't want a pitch; it wants a cheap way to test the claim. Fifty hours is small enough to approve without a board conversation and large enough to prove something.

It worked, and it grew. Consumption now runs at around 300 hours a month.

What we do

Turned the checks into a standard

Before anything scaled, we made the work repeatable. Working from the agency's own practices, we built two master checklists.

Pre-launch technical SEO check — run on staging, before the site is public:

• Crawlability and indexation: robots.txt, meta robots, canonical logic, staging noindex removal — the single most common launch-day catastrophe

• URL architecture: structure, consistency, trailing slashes, parameter handling

• Redirect map: full old-to-new mapping, validated end to end, chains and loops removed

• Metadata across every template and every published page

• Heading hierarchy and semantic structure

• Structured data: Organization, LocalBusiness, Service, Breadcrumb, FAQ as applicable

• Internal linking and orphan-page detection

• Core Web Vitals baseline — LCP, CLS, INP — with named remediation items

• Image optimisation: format, compression, dimensions, alt text

• XML sitemap and Search Console readiness

• Mobile rendering and responsive behaviour

• Analytics and tag verification — confirmed firing, not merely installed

Post-launch verification — run within the first days of going live:

• Live crawl against the pre-launch expectation, differences flagged

• Every redirect re-tested against the live server

• Indexation monitoring and manual sitemap submission

• Search Console and analytics confirmed clean, coverage errors triaged

• Live Core Web Vitals against the staging baseline

• 404 sweep and recovery

• Rankings and traffic watched against the pre-migration baseline

Each checklist carries a severity model — blocker, should-fix, nice-to-have — so the agency's project manager can read one line and know whether the launch goes ahead. That single decision, made reliably, is most of the value.

Built a queue that absorbs spikes

Ten launches a month don't arrive evenly. They cluster. So the engagement runs as a shared queue rather than a fixed roster:

• Launch dates are visible to the ACX team as far ahead as the agency knows them

• Pre-launch checks are scheduled backwards from the launch date, not forwards from the request

• Post-launch verification is scheduled automatically the moment a launch goes live — it is never something someone has to remember

• Retainer execution fills the space around the launch calendar and deliberately absorbs the quiet weeks

Because unused hours roll forward, a light month isn't wasted. It becomes budget for the next cluster.

Took on monthly retainer execution

Alongside the launch work, the ACX team runs recurring SEO execution across the retainer book — on-page optimisation, technical monitoring and triage, rank and visibility tracking, and monthly client-ready reporting in the agency's own template and branding.

Made the agency's people more valuable

The point was never to replace the agency's strategists. It was to stop them spending their week on crawl reports. With execution and verification handled, their senior people moved up the value chain — to client strategy, to the conversations that renew and expand a retainer, and to selling.

How we run it

The team. One account coordinator owns the relationship end to end. Behind them, in-house SEO specialists do the work, under an SEO manager who owns the standard. The agency briefs one person and gets a department.

Two QA gates, not one.

1. 1.

   Service-line QA. Every deliverable is reviewed by the QA officer inside the SEO department, against a standard owned by that department's head.

2. 2.

   Coordinator sign-off. The work then returns to the coordinator, whose check is deliberately different in kind: not is this good SEO? but is this what we told the client we were delivering?

That second gate is the one most providers skip, and it prevents the most common failure in outsourced delivery — competent work that isn't the work somebody asked for.

Service levels. First-round delivery within 48–72 hours of a request. The internal benchmark is one or two minor revisions per delivery; anything beyond that is treated as a brief failure or a QA failure and reviewed as one, not quietly absorbed.

Communication. Everything runs through Dash, ACX's client management portal — requests, delivery, revisions, hours drawn and hours remaining, all in one place with a history. No work lives in someone's inbox.

Confidentiality. ACX works behind the agency's brand. Their end clients see one agency and one point of contact. Nothing carries our name.

Results

Capacity

• 100 retainer clients supported on a steady monthly cycle

• Around 10 sites a month cleared through both pre- and post-launch checks

• Zero SEO hires added to cover the growth

• Grown from a 50-hour starting block to around 300 hours a month

Quality

• Every launch ships with a documented, twice-reviewed check behind it — no more judgement calls made under deadline pressure

• 48–72 hour first-round delivery, held consistently

• One to two minor revisions per delivery as the working benchmark

Cost

At 300 hours a month, the agency pays $4,500 a month — $54,000 a year.

Two US SEO specialists covering the same hours cost approximately $16,250 a month — $195,000 a year, using a $75,000 median base loaded at 1.30× for employer taxes and benefits.

• ACX Hourly Retainer, 300 hrs/month — $54,000

• Two US SEO specialists, loaded — $195,000

• Difference — $141,000 — ACX is 28% of the cost

Per hour that's $46.88 loaded in-house versus $15 with ACX. And the comparison flatters the in-house option: $46.88 buys one individual contributor with no manager above them and no QA behind them. The $15 includes the coordinator, the SEO manager, the specialists and two QA gates.

Loading multiplier of 1.30× is deliberately conservative — US Bureau of Labor Statistics data supports 1.43×. Excludes recruitment, equipment, software, HR administration and severance risk, all of which fall on the employer in the in-house scenario. Salary data February 2026.

What it unlocked

The most important outcome isn't in the numbers. It's that the agency stopped treating new business as a capacity risk.

Before, a big new client or a busy launch quarter provoked a real internal debate: can we actually deliver this, and if we win it, who does the work? That hesitation is expensive in ways that never show up on a P&L — deals slow-walked, upsells not offered, referrals not chased.

Now the answer to "can we take this on?" is yes by default. Delivery capacity became something they buy in the amount they need, in the month they need it, rather than something they hire for and hope to fill.

They also gained something quieter: consistency. A hundred clients getting the same standard of work in a heavy month as in a light one. For a retainer business, where churn is the whole game, that consistency compounds.

The engagement

• Model — Hourly Retainer

• Started — Q4 2025 — as a 50-hour block

• Now — ~300 hours per month

• Team — Account coordinator · in-house SEO specialists · SEO manager · departmental QA

• Delivery SLA — 48–72 hours, first round

• Status — Active

Why this model fit: lumpy demand, mixed work types, and a need for capacity that expands and contracts without a hiring decision attached to it.