The media buying cost most agencies miss

Most agency owners know what they spend on media buyers. They know the salary, contractor invoices, platform fees, and perhaps the percentage of client retainers consumed by paid media work.

What tends to go unmeasured is the cost of the work wrapped around campaign execution.

A campaign doesn’t begin when someone clicks “publish.” It begins with a client brief, a planning call, audience research, asset chasing, spreadsheet checks, naming conventions, landing page reviews, budget approvals, tracking validation, and platform setup. Then there is daily monitoring, pacing checks, search term reviews, creative fatigue checks, client questions, monthly reporting, deck building, and the inevitable Slack message asking why cost per lead moved 18% on Tuesday.

For a marketing or creative agency spending 20 or more hours each week on media execution, manual work can become a quiet margin drain. It doesn’t always show up as a failed account. It shows up as account managers working late, specialists carrying fewer accounts than expected, and partners stepping into delivery because the team has no spare capacity.

The question isn’t simply, “What does AI automation cost?”

The useful question is, “What is manual media buying already costing us, and which parts should an agent own?”

For agencies in the $1 million to $25 million range, we usually see the broader operational leakage land somewhere between $60,000 and $180,000 each year. Media execution is often one of the first places to look because the tasks are repetitive, measurable, and tied directly to client margin.

If you want a broader view of where that leakage sits, see Omni for marketing and creative agencies. The rest of this article focuses on the numbers behind paid media work.

What manual media buying actually includes

“Managing paid media” sounds like a single role. In practice, it is a chain of small decisions and handoffs. Most are necessary. Many don’t need a human to initiate them from scratch every time.

Consider a typical client account with Google Ads, Meta, LinkedIn, or a mix of those platforms.

Campaign setup work

A new campaign can require 4 to 12 hours before it has spent a dollar. The range depends on the number of platforms, audience segments, creative variants, tracking requirements, and approval layers.

That time often includes:

• Translating a client brief into campaign structure
• Reviewing past results and competitor activity
• Building keyword, audience, or placement lists
• Creating naming conventions and campaign taxonomy
• Uploading creative and writing platform-specific copy
• Setting budgets, bid strategies, exclusions, and conversion events
• Checking pixels, UTMs, CRM routing, and landing pages
• Preparing an approval summary for the client
• Documenting what was launched and why

The manual cost rises when the agency is handling local variations, multiple product lines, seasonal pushes, or short campaign cycles. A specialist may be fast in the platform, yet still spend a surprising amount of time moving data between documents, email threads, project management tools, and ad accounts.

This is where automation has to be judged carefully. It won’t replace strategic decisions about a client’s offer, market, or creative direction. It can reduce the assembly work around those decisions.

An AI workflow can take an approved brief, campaign template, audience rules, and creative library, then draft campaign structures, build checklists, create UTM variants, identify missing inputs, and prepare a launch pack for human approval. The media buyer still owns the final decision. They no longer need to construct every repeatable element by hand.

The weekly optimization burden

Campaign setup gets attention because it is visible. Optimization is where time expands without anyone noticing.

A media buyer may check 10 to 20 data points per account across spend, pacing, leads, cost per lead, conversion rate, click-through rate, frequency, search terms, audience performance, creative performance, and pipeline quality. On a small account, this could take 20 minutes. On a complex account, it can take several hours each week.

The issue is not that all accounts need daily manual changes. The issue is that people still need to look at them to know which accounts need action.

That creates a monitoring tax.

For an agency with 15 active paid media accounts, a team can easily spend 20 to 40 hours a week across monitoring, optimization, internal notes, and client updates. Even if a specialist is only making changes on a subset of accounts, they are still reviewing the entire portfolio.

Manual work typically includes:

• Checking budget pacing against monthly targets
• Investigating spend spikes and delivery drops
• Reviewing keyword and search query waste
• Watching frequency and creative fatigue
• Comparing current performance with the prior period
• Checking lead quality in the CRM
• Writing internal notes on actions taken
• Answering account manager questions
• Preparing recommendations for the client

The cost isn’t only labour. Delayed action carries a second cost. If a campaign exhausts budget too quickly, sends poor leads into a sales team, or misses a clear performance change for three days, the client experiences the impact before the monthly report explains it.

An Account Health Agent in Omni ops changes the operating model. It watches connected accounts daily, measures performance against the rules that matter for that client, flags risk and opportunity, and drafts the next-step message before the account manager has to ask.

That doesn’t mean it should autonomously increase spend by $20,000. It means a person receives a useful exception list instead of opening every account in search of one.

For example, the agent might flag that a campaign is 14% ahead of pacing, has a rising cost per qualified lead, and is drawing most spend from one audience segment. It can attach the relevant metrics, suggest actions based on the account’s playbook, and prepare a client-safe explanation. The media buyer reviews the recommendation and acts.

That is a better use of specialist time than routine checking.

Reporting is often the hidden margin killer

Agency owners frequently underprice reporting because it gets bundled into account management. A monthly report can look straightforward from the outside. The actual process is not.

Someone exports platform data. Someone cleans it. Someone compares it against the previous period. Someone adds CRM numbers. Someone explains variance. Someone updates slides. Someone asks the media buyer for context. Someone drafts the email. Then there are the client questions that arrive after the report is sent.

Account managers can spend 30% to 50% of their time on reporting, decks, follow-up emails, and Slack updates in agencies where reporting is heavily manual. That leaves less time for client strategy, retention conversations, and revenue expansion.

The Reporting Agent is designed for precisely this type of work. It pulls performance data from each connected platform, drafts the monthly report, and prepares the account manager’s email summary ready to send. The team reviews the narrative, adds strategic judgment, and sends a report that doesn’t consume a full day of production.

A good reporting workflow should do more than create charts. It should answer four client questions:

1. What happened?
2. Why did it happen?
3. What did the agency do about it?
4. What happens next?

The first question is data. The remaining three need context. An agent can draft the first version by connecting performance movement with known campaign actions, budget shifts, and account notes. A human can then correct the story, add commercial insight, and make a recommendation.

You can see how these capabilities fit inside the wider Omni platform, rather than treating reporting as a standalone dashboard problem.

A practical manual versus automation cost model

The right cost model uses your real team cost, not a generic software price comparison.

Start with fully loaded hourly cost. This includes salary or contractor cost, payroll burden, benefits, management time, software, and a reasonable allowance for non-billable time. For experienced media talent, agencies commonly find their fully loaded cost falls somewhere around $55 to $110 per hour, depending on location and seniority.

Now apply that number to the work that can be reduced, not eliminated.

Activity	Manual weekly time	Time with agent support	Weekly time released
Campaign setup and QA	6 to 10 hours	3 to 5 hours	3 to 5 hours
Monitoring and optimization prep	8 to 16 hours	3 to 7 hours	5 to 9 hours
Reporting and client updates	6 to 12 hours	2 to 5 hours	4 to 7 hours
Internal status and handoffs	3 to 6 hours	1 to 3 hours	2 to 3 hours
Total	23 to 44 hours	9 to 20 hours	14 to 24 hours

This is not a promise that every agency saves 24 hours from day one. Account complexity, data quality, platform connections, and approval processes all matter. It is a useful planning range for an agency that has already documented a repeatable paid media process.

Take a conservative example.

An agency releases 14 hours a week from manual media operations. Its loaded cost for that work is $75 an hour.

• 14 hours x $75 = $1,050 each week
• $1,050 x 48 working weeks = $50,400 in annual delivery capacity

That is only the direct labour view. If the released time allows a media buyer and account manager to carry two more accounts without hiring, the commercial impact can be materially higher. If it prevents a client from leaving because poor pacing was caught earlier, the impact is higher again.

On the other hand, if an agency buys automation software but doesn’t connect its data, define operating rules, or change reporting habits, the return will be poor. Automation doesn’t create margin by existing. It creates margin when it takes ownership of a real workflow.

The ROI timeline should be measured in stages

Agency owners often ask how quickly AI media buying automation pays back. The honest answer depends on implementation discipline, not just software capability.

A realistic timeline has three stages.

Weeks 1 to 2: Map the work and clean the inputs

Start by identifying the recurring work that happens across accounts. Pull a sample of recent campaign launches, optimization notes, reports, and client emails. Look for repeatable decisions and recurring data sources.

At this stage, you may find basic issues that need fixing first. Naming conventions may be inconsistent. Conversion events may not align with CRM stages. Client reports may be built from three different spreadsheet versions. That discovery is useful. It tells you where automation will fail unless the operating process is tightened.

Weeks 3 to 6: Automate preparation and monitoring

The first return usually comes from work preparation. Build launch checklists, campaign templates, exception alerts, report drafts, and internal summaries.

This is where the Account Health Agent and Reporting Agent start returning time to the team. They don’t need to make autonomous buying decisions to create value. They need to reduce the time your people spend gathering data, spotting obvious issues, and drafting routine communications.

Most agencies can see visible time savings in this stage because the work is frequent and easy to compare against the old process.

Weeks 7 to 12: Redesign capacity and pricing

The bigger return appears only when the agency changes how it deploys its people.

If you simply use the saved hours to do the same work more slowly, there is no financial gain. Use the capacity to increase the account load per team member, improve strategic work, reduce contractor dependence, or sell a higher-value service tier.

This is also where content intersects with paid media. The Content Production Agent can create first-pass ad copy, variants, and supporting content from approved briefs. Your creative team edits from an on-brand draft instead of beginning with an empty document. That shortens the gap between performance insight and fresh creative testing.

For owners, the aim isn’t to remove people from the process. It’s to stop using skilled people as data movers, report assemblers, and message drafters.

What should remain human

Media buying automation works best when roles are clear.

Humans should still own client strategy, commercial judgment, creative direction, market positioning, major budget decisions, and the interpretation of unusual context. A client launching into a new market, facing a sales capacity issue, or dealing with a product problem needs experienced advice, not an automated bid adjustment.

Agents should own repeatable observation, preparation, aggregation, first drafts, and structured follow-through.

That division matters because it protects client trust. The account manager should be more available to have a useful conversation, not disappear behind a machine.

The best agencies use automation to make their expertise more visible. Rather than emailing a client a late report full of charts, they can proactively explain a risk, show the evidence, and recommend a next action while there is still time to act.

If you are weighing options across your business, our AI resources and guides can help you frame the questions. But the cost comparison needs to be done against your own account mix, labour model, and delivery process.

How to assess the opportunity in 60 minutes

You don’t need a six-month transformation project to find out where the margin sits. You need a direct look at the workflow.

An Omni Audit is a 60-minute working session, not a deck presentation. We map the work consuming your team’s time, identify the specific agent opportunities, and quantify the capacity and financial upside using your operating numbers.

You leave with three outputs:

• A workflow map showing where manual media execution is consuming time
• A prioritised list of AI agent opportunities, including reporting and account health
• A practical view of likely capacity release, implementation effort, and next steps

Book a 60-min Omni Audit if your team is spending 20 or more hours a week on media execution and you want to see the numbers before making another hire.

The real comparison is not manual media buying versus fully autonomous media buying. It is manual monitoring, reporting, setup, and follow-up versus a better operating model where agents handle the repetitive work and your people handle the decisions clients pay for.

For a closer look at the operating model, the AI audit for marketing and creative agencies lays out the areas we assess. When your team can manage more accounts without lowering the quality of client attention, the economics change quickly.

Book my Omni Audit and we can work through your media buying workflow, account by account.