Agency KPIs: The Metrics That Tell You If Your Agency Is Healthy
The agency KPIs that matter are the ones that tell you whether the business is healthy: average client value, retention, client concentration, pipeline coverage, delivery capacity and margin. Last month's revenue and how busy the team feels are lagging signals. By the time they drop, the problem started 60 to 90 days earlier.
Our team has coached 700+ agencies in Canada, the US and Europe. Our firm BlueLock Consulting worked with Shopify's Plus team and was later acquired by Centro. The pattern is consistent. Healthy agencies review a short scorecard every month, and they treat each number as someone's job. This guide sits inside the operating system in the agency business and pairs with agency OKRs, which turn those numbers into goals the team can actually run.
Why Last Month's Revenue Is Not Enough
Revenue is the number everyone watches, and it is the last one to move. A client who is quietly unhappy still pays this month. A thin pipeline still looks fine while delivery is busy. Underpriced work still shows up as a full calendar.
Then one retainer ends, one proposal slips, or one senior person leaves, and the month looks like a surprise. It was not a surprise. Retention, concentration, pipeline and capacity were already telling the story.
If revenue is the only KPI on the wall, you are steering by the rear-view mirror. Use it. Just do not use it alone.
The Agency KPI Scorecard
You do not need twenty metrics. You need six that describe the health of the business, reviewed on a fixed cadence, with a named owner for each one.
KPIWhat it tells youHow to calculate itReviewAverage client valueWhether you are building a lean, high-value book or a crowded, low-margin oneMonthly revenue from clients ÷ number of active clientsMonthlyRetentionWhether your delivery and reporting earn the next termClients (or revenue) still with you at the end of the period ÷ clients (or revenue) at the start, excluding new wins if you want a pure retention viewMonthly, with a quarterly look backClient concentrationHow exposed you are to one account leavingRevenue from your largest client ÷ total monthly revenueMonthlyPipeline coverageWhether next quarter's revenue is already in motionQualified opportunities in the pipeline, compared with the new revenue you need after expected churnWeeklyDelivery capacityWhether the team can take new work without the founder absorbing itActive accounts and delivery load against your project manager's time, using the 70/30 ruleWeeklyGross marginWhether growth is actually profitableRevenue minus direct delivery cost (people, contractors and tools tied to client work), before overheadMonthly
These are business KPIs. Client-facing metrics (leads, cost per lead, conversion rate, rankings) belong in client reports, not on the agency scorecard. Mix the two and the team starts optimizing campaigns while the firm quietly gets weaker.
Book a call to review your scorecard
Average Client Value Beats Client Count
The fastest way to misread a growing agency is to celebrate the number of clients. Twenty accounts at about $5K a month is $100K in revenue and a heavy delivery load. Ten accounts at about $10K is the same top line with fewer relationships, clearer scope and more room for a lean team.
We walk through that math in the pricing and lean-team guides, and it is one of the first KPIs we look at with owners. If average client value is flat or falling while headcount rises, the agency is getting busier, not stronger.
Published examples of what a higher-value book looks like:
Apex Leads raised its minimum retainer from $1.5K to $4K a month as it rebuilt the offer around booked meetings, and grew from $15K to over $90K a month in 8 months.
Spring Media moved clients from smaller, single-service retainers to broader growth retainers and grew from $10K to $150K a month over 24 months.
PN Digital grew from $30K to over $200K a month in 8 months, including its first $40K-a-month enterprise contract.
Orizon grew from $2M to $5.3M in annual revenue in 12 months, and its average deal size more than doubled.
None of those stories start with "add more logos." They start with a clearer offer and a higher-quality book of business. If you need to reset the commercial side, start with how to price agency services.
Watch: How to Build a $100K/Month Agency With Just 4 People.
Retention Is a Growth Metric
New logos get the attention. Retention pays the bills. A client on a $5K-a-month retainer who leaves after 60 days brought in $10K. The same client kept for 12 months brought in $60K, for the same cost of winning them. That gap is why retention belongs on the scorecard next to sales.
In the published Apex Leads case study, client retention sat around 30%. Clearer onboarding, structured reporting and a more proactive account process lifted it to over 70%, while monthly revenue moved from $15K to over $90K in 8 months. The two numbers moved together because a leaky book of business makes every sales win temporary.
Track both logo retention (did the client stay) and revenue retention (did the account grow, shrink or leave). A book that keeps every client but quietly discounts every renewal is not healthy.
For the operating habits behind those numbers, read why agencies fail at retention and how to fix it and agency client reporting. Reports that only list activity do not help a client defend the retainer.
Watch Client Concentration
If losing your largest client would drop you back toward the $30K plateau, you have a concentration problem, even if this month's revenue looks fine.
Concentration is not automatically a bad client. A strong enterprise account can be a good account. The risk is when that one relationship is also the founder's relationship, the team's main source of work, and the only reason payroll clears. Then a delayed renewal becomes a business crisis.
Review the share of revenue from your largest client every month. If it is high, the work is not "win anything." It is raise average client value across a wider base, lengthen other contracts, and make sure delivery quality is not being sacrificed to protect a single account.
Orizon is a useful contrast: it moved from referral-based, founder-dependent deals to a six-stage enterprise sales process with pipeline tracking, and grew average deal size without making the firm more fragile.
Pipeline Coverage: What's Coming in, Not Just What's Billed
Pipeline is the leading sales KPI. It answers a simple question: if no new inbound appeared tomorrow, would you still hit the next 90 days of new-revenue targets after expected churn?
In agency OKRs we use a straightforward example. If you want to add $30K a month in recurring revenue this quarter and your average retainer is $5K, you need six new clients, plus whatever you lose to churn. Your pipeline has to be large enough, and far enough along, to make that math realistic. A list of "maybe" conversations is not coverage.
Give pipeline a weekly owner and a weekly review. Count only qualified opportunities you would actually take at your current minimum. Orizon built pipeline tracking into a six-stage process so leadership could see every stage, not just closed deals. That is the standard: visible stages, next actions and a close date, not a spreadsheet of names.
Book a call to pressure-test your pipeline math
Delivery Capacity and the 70/30 Rule
Sales KPIs without delivery KPIs create a familiar failure: the team wins work the agency cannot deliver well. Quality drops, retention follows, and the founder is pulled back into every account.
A healthy project manager spends roughly 70% of their time on client work and 30% on internal work: SOPs, templates, training, capacity planning and fixing the process issues that caused last month's fires. When the 30% disappears, the system stops improving. We cover that rule in agency project management.
Capacity is not "how many clients can we stack." A $2K social account and a $40K enterprise build are not the same load. Measure active accounts, escalation volume and whether new work still requires the founder in the thread.
The published AgilizeIQ case study is the capacity version of this KPI in practice. The firm was winning enterprise work faster than its project management system could absorb it. With senior PMs, standardized onboarding and SOPs, it scaled project management capacity 4X in 90 days. New PMs were embedded in accounts within two to three weeks because the playbooks existed before they started. Capacity came from process, not from hoping the next hire would invent the system.
Margin Tells You If Growth Is Real
Revenue can rise while the business gets weaker. That happens when you add clients at prices that do not cover delivery, then hire to keep up. The calendar fills. Profit does not.
Gross margin, as we use it here, is revenue minus the direct cost of delivering the work: delivery salaries, contractors and tools tied to client accounts. Overhead (rent, software for the firm, the founder's non-delivery time) sits below that line. You need both views, but margin on delivery tells you first whether the work itself is worth doing.
We do not publish a single "correct" margin target, because a two-person studio selling large projects and a media shop on smaller retainers will not look the same. What we do see, and already walk through in the lean $100K model, is that underpricing is usually the start of the problem. Many agencies run $5K to $7K retainers for work that should sit at two or three times that. That forces more clients, more hires and thinner margins.
In that same model, $100K a month with a four-person core team often has payroll and direct cost around $40K to $50K a month, leaving roughly $50K to $60K in gross profit before overhead. Use those figures as a planning example, then run your own numbers. If revenue is up and margin is down, stop hiring and fix pricing first.
Ricochet is a published example of the commercial shift: the studio moved from hourly work to fixed-scope projects, grew from about $5K to $30K a month, and was later acquired by a larger agency. Predictable, scoped work is easier to margin than open-ended hours.
Founder Hours in Delivery
This is the KPI most owners feel and few measure. If you do not know how many hours a week you still spend in delivery and client firefighting, you do not know whether the agency is becoming a business or remaining a job with staff.
In the Apex Leads case study, the founder went from more than 60 hours a week in delivery to under 10, while revenue and retention rose. That is the point of the metric. Founder hours should fall as systems, a project manager and a delivery lead take the work, not stay high because "it's faster if I just do it."
Track it for four weeks in 30-minute blocks, the same way we describe in time management for agency owners. If the number is not falling as revenue rises, you still have an agency bottleneck.
How to Review the Scorecard Each Month
A scorecard that no one reviews is decoration. Build the review into the calendar the same way you time-block growth work.
CadenceWhat you look atWho is in the roomWeeklyPipeline (qualified opportunities, next actions, close dates) and delivery load against the 70/30 ruleFounder plus the sales owner and project managerMonthlyThe full six-KPI scorecard: average client value, retention, concentration, pipeline, capacity and margin. One decision per weak number.Founder plus delivery, PM and whoever owns salesQuarterlyTrends, not snapshots. Set or reset the OKRs that sit on top of the scorecard. Look at revenue retention, offer mix and whether founder hours in delivery actually fell.Leadership
Keep the monthly meeting to an hour. The output is not a longer dashboard. It is one or two decisions: raise a minimum, hire or hold on hiring, repair an at-risk account, or put more time into pipeline. Then the Friday review that already sits in a time-blocked week compares plan with actual.
Book a call to map your monthly operating review
Five KPI Mistakes We See
Tracking only revenue. It is lagging. Pair it with retention, pipeline and margin or you will always be late.
Too many metrics. If the team cannot recite the six numbers, the dashboard is too big. Client-channel metrics stay in client reports.
Confusing utilization with health. Billable hours can look strong when you are underpriced or the founder is still doing the work. Pair any utilization number with realization (what you actually collect against what you sold) and margin.
No owner for each number. A metric without a name next to it does not change. Pipeline needs a sales owner. Capacity needs a project manager. Retention needs a delivery or client-success lead. The founder owns the scorecard, not every line on it.
Reviewing only after a bad month. The point of the weekly and monthly rhythm is to see a weak number while you can still do something about it.
Frequently Asked Questions
What KPIs should a small marketing agency track?
Start with six: average client value, retention, client concentration, pipeline coverage, delivery capacity and gross margin. Add founder hours in delivery until that number is consistently low. That set is enough for most agencies under $100K a month. You can layer more detail later. You cannot skip the basics.
How often should we review agency KPIs?
Look at pipeline and delivery load every week. Review the full scorecard every month, with one decision attached to each weak number. Use the quarterly review for trends and to set OKRs. Daily dashboards tempt people to react to noise.
What is a good client retention rate for an agency?
There is no single number that fits every model. Contract length, offer type and whether you sell projects or retainers all change the picture. What we can point to is a published before-and-after: Apex Leads moved from about 30% retention to over 70% by fixing onboarding, reporting and account management, and revenue followed. If you do not know your current rate, that is the first gap to close.
Should agencies track utilization and billable hours?
They can be useful as a capacity signal, especially once you have a delivery team. They are a poor standalone health metric. High utilization can hide underpricing, scope creep or a founder who has not left delivery. If you track utilization, also track what you actually collect and the margin on the work. Price the offer on outcomes, not hours; see how to price agency services.
How many KPIs is too many?
If you cannot review them thoroughly in a one-hour monthly meeting, you have too many. Six business KPIs on one scorecard is plenty. Marketing and campaign metrics belong in client reports, where they help the client see the work, not on the agency's health dashboard.
What should we fix first if revenue is uneven?
Read the scorecard, not the top line. If retention is weak, fix delivery, onboarding and reporting before you spend more on sales. If retention is solid but pipeline is thin, the problem is acquisition. If both look fine and margin is poor, the problem is pricing or delivery cost. Uneven revenue is usually a lagging symptom of one of those three.
A healthy agency is not the one with the busiest dashboard. It is the one that can show, every month, that client value is holding or rising, clients are staying, no single account carries the firm, the pipeline can replace what leaves, the team can deliver without the founder, and the work is actually profitable.
If you want a second set of eyes on your numbers, book your free strategy call. Our team will help you put a simple scorecard in place and map the next 90 days around the KPI that is holding you back.