A 35-person AI agency in Seattle installed three dashboards, subscribed to a business intelligence platform, and started tracking 47 different metrics. Six months later, the founders admitted that nobody actually looked at most of them. The dashboards were overwhelming, the metrics were disconnected from decisions, and the team could not distinguish between numbers that mattered and numbers that were just noise. Paradoxically, the agency had more data visibility than ever but worse decision-making because nobody could focus on what was important.
The solution is not more metrics. It is the right metrics — a curated set of numbers that provide genuine insight into the health and trajectory of your agency, connected to the decisions you actually make. This playbook defines the metrics that matter for AI agencies, explains what each one tells you and what to do when it moves, and provides a framework for building a measurement system that drives action rather than analysis paralysis.
The Metric Hierarchy
Not all metrics are created equal. Organize them into a hierarchy based on how they are used.
Level 1: North Star Metrics (3-5 Metrics)
These are the numbers that define the overall health and trajectory of your agency. Leadership reviews them weekly. Every other metric exists to explain or improve these.
Revenue growth rate: Year-over-year revenue growth. This tells you whether the business is growing. Target depends on stage — early agencies should target 30-100%+, mature agencies 15-30%.
Net profit margin: Net income divided by revenue. This tells you whether growth is profitable. Target: 15-25% for a healthy agency.
Client retention rate: Percentage of clients retained year-over-year. This tells you whether you are building lasting relationships or churning through clients. Target: 85%+ annually.
Team utilization rate: Billable hours divided by available hours. This tells you whether your team is deployed against revenue-generating work. Target: 70-78% blended.
Employee retention rate: Inverse of voluntary turnover. This tells you whether you can keep the talent that makes everything else possible. Target: 85%+ annually.
Level 2: Functional Metrics (10-15 Metrics)
These metrics provide deeper insight into each business function. Functional leaders review them monthly.
Sales metrics:
- Pipeline coverage ratio: Weighted pipeline value divided by revenue target. Target: 3-4x.
- Win rate: Deals won divided by total opportunities. Track by deal size and service type.
- Average deal size: Average contract value of won deals. Track trends.
- Sales cycle length: Average days from opportunity creation to close.
- New client acquisition cost: Total sales and marketing cost divided by number of new clients acquired.
Delivery metrics:
- On-time delivery rate: Percentage of projects delivered within the agreed timeline. Target: 80%+.
- Gross margin: Revenue minus direct delivery costs, divided by revenue. Target: 50-65%.
- Client satisfaction score: Average satisfaction rating across active clients. Target: 8+/10.
Financial metrics:
- Days sales outstanding (DSO): Average collection period. Target: under 45 days.
- Revenue per employee: Total revenue divided by headcount. Target: $150,000-250,000.
- Cash reserve: Months of operating expenses in cash. Target: 2-3 months.
People metrics:
- Time to hire: Days from job posting to accepted offer. Target: under 45 days.
- 90-day retention: New hires still employed at 90 days. Target: 90%+.
- Employee satisfaction: Regular survey score. Target: 7.5+/10.
Level 3: Diagnostic Metrics (As Needed)
These metrics are used to investigate when Level 1 or Level 2 metrics show problems. You do not review them on a regular cadence — you dig into them when something needs explaining.
Examples:
- Rework rate by project (investigate when margins are low)
- Time allocation by category (investigate when utilization is low)
- AR aging breakdown (investigate when DSO is high)
- Client health scores by segment (investigate when retention is declining)
- Billable versus effective rates (investigate when revenue per hour is declining)
- Win rate by lead source (investigate when pipeline conversion is weak)
Defining Your Metrics
For each metric you track, define:
Name: Clear, unambiguous name Definition: Exactly how it is calculated (formula, data sources, inclusions and exclusions) Owner: Who is responsible for this metric? Target: What is the goal? Frequency: How often is it measured and reviewed? Action threshold: At what point does a change require investigation or intervention? Data source: Where does the data come from?
Why precise definitions matter: "Revenue" means different things to different people. Is it billed revenue, recognized revenue, or collected revenue? Is it gross or net of discounts? If two people calculate the same metric differently, your discussions will be about the number rather than the insight.
Building Your Metrics Dashboard
Dashboard Design Principles
One page: Your primary dashboard should fit on one screen. If people have to scroll or click through tabs, they will not use it.
Current and trended: Show the current value and the trend over time (typically 6-12 months). A utilization rate of 73% in isolation is less useful than knowing it was 78% three months ago and has been declining.
Compared to target: Every metric should show the target alongside the actual. A green/yellow/red color system works well for quick comprehension.
Actionable: Every metric on the dashboard should connect to a decision. If you cannot articulate what you would do differently if the metric changed, it does not belong on the dashboard.
Implementation Options
Spreadsheet dashboard (simplest):
- Google Sheets or Excel with manual data entry
- Works for small agencies (under 15 people)
- Pro: Free, flexible, no setup time
- Con: Manual data entry, prone to errors, no real-time updates
BI tool dashboard:
- Metabase, Looker, Tableau, or Power BI
- Connects to your data sources for automated updates
- Works for mid-size and larger agencies
- Pro: Automated, visual, shareable
- Con: Setup time, cost, requires data to be in queryable systems
All-in-one agency tools:
- Productive.io, Scoro, or Kantata
- Built-in metrics specific to agency operations
- Pro: Purpose-built, integrated with project and financial data
- Con: May require migrating from current tools, cost, less customization
Recommended Dashboard Layout
Top row — North Star Metrics: Revenue (monthly and YTD versus plan) | Net margin | Client retention | Utilization | Employee retention
Second row — Pipeline and Revenue: Pipeline value | Pipeline coverage | Win rate | Revenue by service line | Recurring revenue percentage
Third row — Delivery and Operations: Active projects | On-time delivery rate | Average project margin | DSO | Cash position
Bottom row — People: Headcount | Open positions | Time to hire | Employee satisfaction trend
Metrics Cadence
Weekly (Leadership Team)
Review North Star Metrics in your weekly leadership meeting (10 minutes):
- Revenue: On track for the month?
- Utilization: Where is it this week? Trend?
- Pipeline: Any changes that affect the forecast?
- Delivery: Any projects at risk?
- Cash: Any concerns?
Monthly (Leadership Team + Functional Leaders)
Deep review of Level 1 and Level 2 metrics (60 minutes):
- Financial review: P&L actual versus budget, cash flow, AR aging
- Sales review: Pipeline health, win rate, forecast accuracy
- Delivery review: Project margins, on-time delivery, client satisfaction
- People review: Utilization, headcount, retention, satisfaction
- Action items: What needs to change based on what the numbers show?
Quarterly (Leadership Team + Full Company)
Strategic review against annual goals:
- Progress toward annual revenue and profit targets
- Client portfolio health (retention, concentration, growth)
- Operational improvement against targets
- People and culture metrics
- Adjust strategy and priorities based on data
Annually (Leadership Team)
Full year review and planning for next year:
- Year-over-year performance comparison
- Metric target setting for the coming year
- Dashboard and metric refinement
Common Metrics Mistakes
Mistake 1: Vanity Metrics
Tracking numbers that look good but do not drive decisions. Total revenue sounds impressive but does not tell you if you are profitable. Number of proposals sent does not tell you if you are winning.
Fix: For every metric, ask "What would I do differently if this number changed?" If the answer is nothing, it is a vanity metric.
Mistake 2: Lagging Indicators Only
Metrics that tell you what already happened (revenue, profit, retention) without metrics that predict what will happen (pipeline coverage, client health scores, employee satisfaction).
Fix: Balance lagging indicators with leading indicators. Pipeline coverage predicts revenue. Employee satisfaction predicts retention. Client health scores predict client retention.
Mistake 3: Measurement Without Action
Tracking metrics religiously but never acting on what they show. The dashboard becomes a decoration rather than a decision tool.
Fix: Every metric review must produce action items. If utilization dropped, what are you doing about it? If DSO increased, what is the collection plan?
Mistake 4: Individual Metrics Without Context
Evaluating a single metric in isolation. A utilization rate of 82% looks great until you realize it is driven by overtime that is causing burnout and quality problems.
Fix: Always interpret metrics in combination. Utilization plus employee satisfaction plus quality metrics together tell a more complete story than any one of them alone.
Mistake 5: Too Many Metrics
Tracking 50 metrics and reviewing 5. The noise obscures the signal, and the effort of maintaining unused metrics wastes time.
Fix: Ruthlessly prune. Start with the North Star metrics and add Level 2 metrics only as they prove necessary. Remove any metric that has not been used to make a decision in the past quarter.
Benchmarking
Agency Industry Benchmarks
Compare your metrics against industry benchmarks, but use them as directional guides rather than absolute targets:
Revenue:
- Revenue per employee: $150,000-250,000
- Year-over-year growth: 15-30% (mature), 30-100% (early stage)
- Recurring revenue percentage: 20-40% (good), 40%+ (excellent)
Profitability:
- Gross margin: 50-65%
- Operating margin: 15-25%
- Net margin: 10-20%
Delivery:
- On-time delivery: 80-90%
- Client satisfaction: 8-9/10
- Utilization: 70-78%
Sales:
- Win rate: 20-35%
- Sales cycle: 30-90 days
- Pipeline coverage: 3-4x revenue target
People:
- Voluntary turnover: 10-15% annually
- Time to hire: 30-45 days
- Employee satisfaction: 7.5-8.5/10
Your Next Step
This week:
- Identify your current North Star Metrics. Can you state your current values for revenue growth, profit margin, client retention, utilization, and employee retention? If not, that is your starting point.
- Calculate the metrics you are missing. Even rough estimates are better than no visibility.
- Review your existing dashboards or reports. What is useful? What is noise?
This month:
- Build or refine your primary dashboard with the North Star and Level 2 metrics.
- Establish precise definitions for each metric (formula, data source, owner, target).
- Implement a monthly metrics review cadence with your leadership team.
This quarter:
- Automate data collection for your most important metrics (connect dashboards to source systems).
- Establish benchmarks and targets for all Level 1 and Level 2 metrics.
- Conduct a metrics audit: remove any metric that has not informed a decision.
- Train the team on the metrics that matter and how to interpret them.
Metrics are the language of business performance. They transform opinions into facts, assumptions into evidence, and arguments into analysis. But only if you measure the right things, review them at the right cadence, and — most importantly — act on what they tell you.