How to Set Content Cadence and Capacity Across a Full Client Roster Without Burning Out

You said yes to four posts a month for a new logo, and now you're doing the math at 9pm wondering who is actually going to write them. That feeling is common, and it is fixable. This guide shows you how to decide how much content each client should get, turn those promises into hours by role, and plan agency content capacity across the whole roster so your team stays on schedule instead of underwater.

You'll need one thing before you start: honest numbers about your team's real hours. Nothing fancy. A spreadsheet is fine.

Step 1: Audit the hours your team really has

Start here, not with the client. You cannot promise a cadence until you know what your team can deliver.

For each person, write down their role and skills, their available weekly hours, planned time off, internal meetings, admin, training, and any sales or account duties. Then add their current client and internal commitments, and their expected availability for the next four to twelve weeks.

Here's the part most agencies skip. A 40-hour week is not 40 hours of production. One agency capacity-planning example subtracts eight meeting hours and four admin hours from a 40-hour week, leaving 28 productive hours, and perhaps only 20 of those are available for billable client work.

Keep three ideas separate, because mixing them is what wrecks a plan:

• Capacity is the total hours the team could work in a period.
• Utilization is the share of available time spent on billable or client-facing work.
• Workload is the work actually assigned.

Someone can have capacity on paper and still be unavailable. They may lack the skill the job needs, be locked to another deadline, or owe time to internal work.

Two formulas do most of the lifting. Agency utilization = billable hours ÷ total availability × 100 tells you what already happened. Delivery capacity = (gross capacity − internal hours) × weekly utilization target estimates what you can safely sell, where gross capacity = working hours per week × weeks in the period.

You'll know this step is done when you have a role-by-role sheet showing real available hours, committed hours, remaining hours, required skills, planned absences, internal work, and a named owner for every big commitment. You should be able to see four to twelve weeks ahead, with a rolling three-to-six-month view for hiring calls.

Where people go wrong: booking a role above 100 percent, treating an empty calendar as availability when the skill is wrong, leaving revisions and client calls out of the estimate, and promising work before checking the role with the least room left.

Pro tip: Use your own history. If a task always runs longer than you estimate, replace the estimate with the actual. A capacity plan is only as good as the numbers inside it.

Step 2: Sort your clients into a few service tiers

Deciding how much content per client to promise starts here, and it does not start with a number. Resist the urge to give every account the same number of articles. Build a small set of service tiers instead, each describing a level of attention you can reliably deliver.

Three dimensions make a tier real:

1. Business priority. How much the client's growth, pipeline, visibility, or retention depends on content.
2. Content intensity. How many deliverables per period, and how complex they are.
3. Agency effort. The hours across strategy, research, writing, editing, client review, publishing, reporting, and distribution.

A tier is a package, not a post count. Spell out the primary pieces, the supporting distribution content, the reporting rhythm, the level of strategist involvement, the revision boundary, and what you need back from the client.

Here's a shape you can borrow and fill with your own numbers:

Those tier names and counts should come from your own history of effort and capacity. There is no industry table that tells you the right number.

This is also where per-client separation starts paying off. In DeepSmith, each client sits in its own workspace with isolated brand context, content, plan, and team access, so one account's voice, product facts, competitors, and tracked prompts never leak into another's. Separate workspaces solve context. They do not solve your roster math, which is the rest of this guide.

You'll know this step is done when every client has one assigned tier, a written deliverable list, a monthly or quarterly commitment, an effort estimate by role, and a clear line around what counts as new scope.

Where people go wrong: selling a high-volume package because the client asked, without costing it. Using one cadence for clients with different goals. Counting articles and ignoring distribution, reporting, meetings, revisions, and account management. Letting every client be urgent. Building so many custom packages that a small team cannot schedule them.

Step 3: Set each client's cadence from their goal, not a benchmark

Content cadence is more than a volume target. It covers how often you publish, plus the timing and the pattern behind it. The question is not only how much you can produce. It is what should go out, when, and why.

Work the decision in order.

1. Name one primary objective

Common ones: brand awareness and thought leadership, lead generation, search visibility, audience education, or support for a launch or season. Pick one to govern the cadence. Secondary goals can shape the mix, but they don't get to grow the workload forever.

2. Decide the real content mix

Long-form articles, short supporting pieces, case studies, decision-stage content, guest posts, video, newsletter copy, social. A client asking for "four posts per month" is often asking for four researched articles, four approval cycles, social posts, newsletter copy, internal linking, and reporting. Price and schedule the whole service, not the headline unit.

3. Check the audience and the channels

Frequency should change by channel. Social may reward publishing more often while the blog rewards fewer, stronger pieces. Use evidence you actually have: customer conversations, surveys, website analytics, search performance, email and social engagement, and conversion data. A generic frequency chart is not a reason to promise anything.

4. Check the season

Your schedule doesn't have to be identical every month. Bend it around launches, industry events, buying seasons, planning periods, and the weeks the client's experts are unavailable.

5. Choose the lowest cadence that still serves the goal

If your team can deliver one excellent, useful article a month but not four, then one is the honest answer. A schedule you keep beats a bigger promise that slips. Deciding content cadence per client is a service commitment, so treat it like one.

Setting content cadence per client this way keeps the promise honest, because it is built from their goal and your hours rather than from a chart. Benchmarks are useful context here, and only context. A 2025 Orbit Media survey of 808 bloggers found about half of marketers publish two to four times a month, and that more frequent publishing and more detailed articles were both associated with stronger reported results. Those are correlations across a survey population, not a quota for your client. The same survey put average blog-post creation time at just under three and a half hours, which describes blogging in general, not an agency package with research, interviews, fact checking, approvals, SEO, and distribution attached.

You'll know this step is done when you can answer, for each account: which goal this cadence serves, which channels and content types are included, how many primary pieces are planned, what the client owes you, how much of your capacity the package eats, and what evidence would change it.

Step 4: Turn every cadence into hours by role

Now convert promises into numbers. The unit of capacity planning is not "one article." It is the full service effort behind that article.

For each content unit, estimate time for the strategist or account lead, research, subject-matter expert, writer, editor, SEO or AEO review, design if any, client communication and approval, publishing and QA, distribution or repurposing, reporting, expected revisions, and a contingency for complexity or slow feedback.

Then two sums:

Total client effort = deliverable effort by role + account overhead + review and revision time

Roster demand for role = that role's hours across every client package

Compare each role's demand against that role's sustainable capacity from Step 1. Role-level hours are what agency content capacity really means. This is the heart of agency workload planning, and the reason a roster that looks fine in total can still be on fire.

Don't forget the client's own hours. A cadence isn't planned until you've documented who supplies expertise, how much time your contact can give, who approves drafts, how fast approvals come back, how many stakeholders review, and what happens when feedback is late.

That last one sets more cadences than agencies admit. One editorial-calendar example works it out plainly: if three subject-matter experts can each give six hours in a quarter, and each article needs one hour of interview plus one hour of review and approval, the client supports nine articles that quarter. Not because you can't write more, but because they can't feed more.

Automation changes some of these numbers, and it's worth being precise about which. DeepSmith produces finished, brand-grounded articles with research, internal and external links, a cover image, and publish-ready metadata already in place, and keyword coverage, heading structure, schema markup, and citation-ready formatting are part of the writing pipeline rather than a cleanup pass afterward. Deep IQ holds each client's positioning, products, personas, and brand voice, so drafts arrive on-brand without a re-brief. What that does is shrink the mechanical hours. It does not shrink strategy, client context, approvals, or quality control, and you still budget those in full.

Where people go wrong: estimating writing but not research, editing, approvals, publishing, reporting, or distribution. Using one average article time for every topic. Ignoring the strategist as a bottleneck. Treating client review as free. Totalling hours without splitting them by role. Assigning work by open calendar space instead of skill and deadline fit.

Step 5: Build one roster plan and protect a buffer

Agency workload planning only works at the roster level, so pull every client's demand into a single view. Planning each account on its own island is how small teams get surprised.

Your roster view should carry every client, every deliverable and due date, the role required, estimated hours by role, client priority, fixed and flexible deadlines, client dependencies, planned time off, remaining capacity, and a risk flag.

Compare demand with capacity per role, not just for the team. Agency content capacity is a set of role numbers, never one team number. A roster can have plenty of total hours and still break, because every client needs the same editor in the same week.

Common mistake: Setting cadence per client before checking role-level capacity across the whole roster. A package that fits one account can flatten the strategist, editor, or approver serving ten of them.

And please don't plan to 100 percent. Full utilization leaves nothing for client changes, revisions, urgent requests, sick days, meetings, sales work, creative recovery, cross-training, or a broken CMS on publish day. The official Kanban guidance makes the same point in flow terms: when resources are fully used, slack disappears, flow suffers, and context switching hurts more. The goal is a steady stream of finished work, not a full calendar.

The research supports having a buffer, but no source sets one universal percentage. Pick one from your own volatility and test it. Reserve time for unexpected requests, slots for urgent work, schedule room around fixed launches, and skills cover for your bottleneck role. Label the buffer out loud. Hidden buffer gets sold twice.

When demand runs past capacity, work a deliberate order: protect contractual and launch-critical work, then work tied to the client's main goal, then move internal work, then lower-priority content, then reassign where the skill fit is safe, then use qualified freelancers for a defined gap, then reduce cadence or change the package, then extend timelines, and only then hire.

You'll know this step is done when the plan makes overload visible before the deadline, naming the client, deliverable, role, and week that collide.

Step 6: Cap work in progress and forecast on a rolling basis

Work in progress is simply how many items are in flight. A WIP limit is the ceiling on how many are allowed in a stage, lane, person, or work type.

For a multi-client content team you can cap research items, drafts in progress, pieces awaiting internal review, pieces awaiting client approval, pieces awaiting publish, work assigned to one strategist or editor, work for one tier, or total active pieces across the roster.

Why bother? WIP limits cut context switching, make blockers visible, shrink the pile of nearly-finished work, push the team to finish before starting, and stop new work entering a full stage. A team that looks busy can still deliver slowly when ten accounts pull at once.

Setting them is not complicated. Map your real delivery stages. Count the average number of items sitting in each stage over a few cycles. Agree an opening limit per stage or role. Watch blocked work, cycle time, and completed work. Adjust after a few weeks.

Common mistake: Raising a WIP limit every time you hit it. Hitting the limit is the limit doing its job. Find the blocked stage first, or you'll just create more unfinished work.

Then keep three horizons running. This week, protect deadlines and clear blockers. The next four to twelve weeks, compare planned client demand with role-level capacity. The next three to six months, look for recurring overload, likely pipeline demand, freelancer needs, and hiring signals. Update weekly or monthly, revisiting actual versus estimated time, scope changes, time off, new opportunities, client pauses or expansions, bottleneck roles, and anything that keeps getting carried forward.

You'll know this step is done when you can say why a new request was accepted, delayed, reassigned, outsourced, or declined, and no new commitment enters a full role without a trade-off.

Step 7: Review actuals, rebalance, and know when to hire

Monthly or quarterly, review output and sustainability together.

Per client, compare planned versus published content, planned versus actual hours, revision volume, approval delays, distribution effort, results against the primary goal, margin where you can see it, and whether the tier still fits.

Across the agency, look at utilization by role, remaining capacity by role, carried-over work, WIP breaches, urgent requests, overtime and missed deadlines, contractor usage, the role that keeps becoming the bottleneck, and any client you're quietly overservicing.

Use role-specific utilization targets, never one agency-wide number. One agency-utilization reference lists 75 to 80 percent for producers and freelancers, 35 to 50 percent for managers, and more than 10 percent for sales and administration. Another reports 75 to 90 percent for weekly delivery-oriented roles, while noting annual organizational targets often land far lower once time off and non-delivery work are counted. Reference points, not mandates.

Here's the arithmetic in one line. Say a role has 28 productive hours a week and you set a 75 percent delivery target: planned capacity is 21 hours. If the roster needs 24 hours from that role, you have a three-hour weekly gap. Don't close that gap by assigning the work and hoping. That's the illustration, not a target for your agency.

One capacity source treats 32 to 35 billable hours a week as full capacity for a discipline, and suggests looking at hiring when that level holds for three months with demand still coming. Before you hire, check whether you're overservicing, whether your estimates are wrong, whether legacy clients need repricing, whether internal work can wait, whether a specialist contractor covers the skill gap, and whether the demand is really recurring. Hiring is rarely the first answer.

Choosing where the next free slot goes is its own decision, and it should follow evidence. DeepSmith's Content Map turns each client's site and their competitors' sites into one topic map with coverage gaps and untapped topics, and Opportunity Agents read that data plus the client's AI-visibility results and return ideas with the justifying data point attached. That turns "which client gets the next article" into a call you can defend in a renewal conversation.

The figures shown are demo data from a sample workspace.

You'll know this step is done when every client has one decision attached: keep the cadence, reduce it, raise it because results and capacity support it, or change the mix or price. Every recurring bottleneck gets one too: better estimates, a workflow or WIP change, reassignment, a contractor, repricing, or a hire.

What to do next

Pick one thing this week. Build the role-by-role capacity sheet from Step 1, using real hours and not calendar hours. Everything else in this guide leans on that number, and most agencies have never written it down.

Then assign tiers, cost each cadence by role, reserve your buffer, set WIP limits, and review actuals before you change a single commitment. That loop is the whole content capacity planning agency owners can actually run week after week. You already know your clients. This just makes the math visible.

If mechanical production hours are what's eating your margin, that part is worth attacking directly. DeepSmith runs AI-search visibility tracking and content production in one place, with a separate workspace per client, so your strategists spend their hours on judgment instead of briefs, linking, and metadata. Pro is $99 a month for 20 articles, 50 tracked prompts, and 5 seats, and Grow is $199 for 40 articles, 100 prompts, and 7 seats. There's a 7-day free trial with no long-term contract, so you can test it against one account before you touch the roster plan.