Most marketing reports are inventories: every number the tools can export, arranged in charts, signifying nothing in particular. The fix isn’t more metrics — it’s mapping the right handful to the funnel stage each one governs. When every KPI has a stage, an owner, and a decision it informs, reporting shrinks and improves simultaneously. Here’s the stage-by-stage map, plus the discipline that keeps it honest.
First, the Rule That Organizes Everything
A KPI earns its place by answering one question: if this number moves, what would we do differently? If nothing changes regardless of the value, it’s trivia. Every metric below is paired with the decision it drives — copy that pairing into your own dashboard, and delete anything on it that has no answer.
One more distinction worth keeping sharp: leading indicators (traffic, engagement, pipeline) predict outcomes and give you time to react; lagging indicators (revenue, CAC, retention) confirm outcomes after the fact. Healthy dashboards pair them, because leading-only dashboards celebrate motion and lagging-only dashboards diagnose too late.
Stage 1: Awareness — Is the Right Audience Discovering You?
- Reach and impressions within target segments — not total reach; a million impressions to the wrong audience is a vanity spike. Decision: which channels and content earn more investment.
- Branded search volume — the cleanest free signal that awareness is converting into memory. Decision: whether brand-building efforts (content, social, PR) are compounding.
- Organic impressions in search console — the leading indicator of SEO health, visible months before clicks arrive. Decision: which topic clusters to expand.
- Share of voice — your mentions and visibility relative to named competitors. Decision: whether you’re gaining or losing ground in the conversation that matters.
What to ignore at this stage: follower counts as a goal (they’re a byproduct), and total pageviews without source or audience context.
Stage 2: Engagement — Does the Audience Care Once It Arrives?
- Engagement rate per post (interactions ÷ reach) — compared against your own trailing 90-day average, not industry folklore. Decision: which content pillars and formats get more slots in the calendar.
- Time and depth on content — scroll depth or engaged time on key pages. Decision: which pieces deserve a rewrite versus a promotion push.
- Email engagement — click rate above open rate (opens are inflated by privacy proxies). Decision: subject lines, send cadence, and list pruning.
- Returning visitor share — are people coming back, or is every visit a first date? Decision: whether the content is building an audience or just collecting drive-bys.
Stage 3: Conversion — Does Attention Become Leads and Customers?
This is the stage leadership should see first, and where most funnels leak silently.
- Conversion rate by step — visitor→subscriber, subscriber→trial or lead, lead→customer, each measured separately. A single blended rate hides which step is broken. Decision: where optimization effort goes this quarter.
- Qualified leads, not raw leads — define “qualified” in writing with sales, or the number is negotiable fiction. Decision: channel budget allocation.
- Cost per acquisition by channel — with honest cost accounting (include agency fees and tools, not just ad spend, when comparing strategically). Decision: scale, fix, or kill each channel.
- Sales cycle length and win rate from marketing-sourced leads — the numbers that reveal lead quality after the handoff. Decision: whether marketing is filling the pipe or clogging it.
Stage 4: Retention and Revenue — Does It Compound?
- Customer lifetime value (LTV) — computed conservatively on a 12–24 month margin basis. Decision: how much you can afford to spend acquiring each segment.
- CAC payback period — months until a customer’s margin repays acquisition cost. Under ~12 months, growth funds itself; beyond, growth consumes cash. Decision: the pace at which you can responsibly scale spend.
- Cohort retention curves — do customers acquired in a given month keep buying or using? Flattening curves justify aggressive acquisition; decaying ones make it arson. Decision: whether the next dollar goes to acquisition or retention.
- Referral and expansion rates — the share of new revenue arriving free from existing customers. Decision: whether word-of-mouth mechanics deserve systematic investment.
Assembling the Dashboard: The 12-Number Rule
Pick at most twelve numbers across the four stages — typically three per stage — and assign each an owner and a target or baseline. Two views usually suffice: a channel-owner view (weekly, granular, diagnostic) and an executive view (monthly, dominated by stages three and four). Present bottom-up: money first, conversions second, engagement only where something needs explaining. The classic failure is the reverse — forty charts of reach and likes, with revenue as a footnote.
The Cadence That Turns Numbers Into Decisions
Weekly: a fifteen-minute anomaly scan — is anything broken or spiking? No strategy changes. Monthly: performance versus targets, ending with a mandatory section titled Decisions and changes; if it’s empty two months running, the meeting is theater. Quarterly: the economic review — CAC, LTV, payback by channel — plus kill/scale calls and next quarter’s targets. This rhythm matters more than metric selection; a mediocre KPI set reviewed with discipline beats a perfect one admired passively. (For the full measurement system — tracking setup, attribution, experiment design, and the traps of reading data wrong — see our complete guide to marketing analytics.)
Stage-Mismatch: The Error Behind Most KPI Arguments
When teams fight about metrics, it’s usually a stage mismatch in disguise: judging a brand-awareness campaign by last-click conversions (stage 1 effort, stage 3 yardstick), or celebrating engagement on content that never converts (stage 2 success, stage 3 silence). Every campaign should declare its stage before launch and be judged by that stage’s KPIs, with one leading indicator from the next stage tracked as the bridge. That single habit dissolves most reporting arguments — and most bad budget decisions — before they start.
The goal was never to measure everything. It’s to know, at each stage of the funnel, the two or three numbers that tell you where the constraint is — and to have already decided what you’ll do when they move.
