Measurement is where good microwebsites get misjudged in both directions. A property that is genuinely working gets abandoned because the wrong numbers looked flat; a property that is quietly failing gets defended because the wrong numbers looked great. The problem is almost never a lack of data — it is measuring the things that are easy to move instead of the things that predict revenue, and stopping at the metrics that flatter a report rather than the ones that answer the only question that matters: did it pay.
This post lays out a framework that runs all the way to that question. It sits downstream of the economics guide, which forecasts the return before you build; measurement is how you verify what actually happened after launch. The two are halves of one discipline, and neither is complete alone.
Measuring the right thing
Every metric a microsite produces falls somewhere on a line from vanity to revenue. Vanity metrics move easily and prove little — total pageviews, impressions with no context, a ranking for a term nobody converts on. Revenue-predictive metrics move harder and mean more — qualified calls, booked work, cost per acquisition. The trap is that the vanity end is easier to report and always seems to be going up, which makes it comfortable and useless in equal measure.
The framework below is organized into three tiers precisely so the comfortable numbers stay in their place as early signals, and the decision always comes back to the tier that reaches revenue. Measurement only earns its keep when it gets there.
Tier one — visibility (leading)
Visibility metrics are the earliest signals: impressions, keyword rankings, and share of the local pack for the terms that matter. They tell you the property is being found, and they move first — often weeks before any revenue does, which makes them genuinely useful as an early read on whether the build is taking hold.
What they do not tell you is whether any of it is worth anything. A property can climb the rankings for terms that never convert and produce a chart that looks like success and a bank balance that disagrees. Treat visibility as the leading indicator it is: necessary, encouraging, and insufficient.
Tier two — engagement (diagnostic)
Engagement metrics — conversion rate, session quality, micro-conversions like a form started or a guide downloaded — are diagnostic. They sit between being found and being paid, and they are where you catch a property that is ranking but leaking. When visibility is strong and revenue is weak, the answer is almost always in this tier: the traffic is arriving and failing to convert, and engagement metrics tell you where.
Read this tier to fix, not to celebrate. Its job is to explain the gap between the tier above it and the tier below.
Tier three — business impact (outcome)
Business-impact metrics are the outcome the whole property exists to produce: qualified calls, booked work, and revenue attributed to the asset. These are the numbers that justify the build to anyone holding the budget, and they are the ones the first two tiers are ultimately in service of.
They also move last and hardest, which is why the discipline of the tiers matters — you watch visibility and engagement so you are not flying blind in the months before the outcome numbers accumulate. But when the time comes to judge the asset, this is the tier that decides.
Closing the loop: from metrics to ROI
A measurement framework that stops at rankings has not measured anything that matters. The entire point of the tiers is to reach the question an owner actually asks: did this asset return more than it cost. Answering it takes three moves.
First, attribution — connecting revenue back to the property. In local work this runs on dedicated contact paths for the asset, form-source tagging that records where a lead came from, and matching offline bookings back to the online source that produced them. Attribution is also where honesty is required: last-touch and first-touch models each tell a partial story, and some bookings will never trace cleanly. A good attribution setup does not pretend to perfect precision; it captures enough, consistently enough, to be decided on. Our World Cup Limo case study is an example of clean organic attribution — no paid media in parallel, so match-day bookings could be tied to the microsite and the visibility it earned.
Second, cost per acquisition. Attributed revenue and booked work turn into a real cost to acquire a customer through the asset — and that number only means something in comparison. Set it against your paid channels and your other marketing spend. A microsite whose cost per acquisition beats your paid channels and keeps falling as the build cost recedes is doing exactly what an asset is supposed to do.
Third, the decision. The ROI number exists to drive an action: keep the asset as it is, scale it because it is beating the alternatives, or retire it because it is not. Measurement without that decision at the end is just reporting. This is the moment the forecast from the economics guide gets checked against reality — the forecast said it would pay; the loop tells you whether it did.
Reading the tiers together
The tiers are not independent scorecards; they are a sequence with an expected shape. Visibility should move first, engagement should convert that visibility, and business impact should follow. When they move in that order, the property is healthy. When they diverge, the gap is the diagnosis: visibility without engagement is a conversion problem, engagement without impact is a lead-quality or fulfilment problem, and impact without visibility usually means you are measuring the wrong terms. Reading the tiers together lets you catch a problem in the weeks it takes to show up in the leading indicators, rather than the quarter it takes to show up in revenue.
Reporting cadence
Match the cadence to the tier. Visibility and engagement are worth a weekly glance, because they move early and give you time to react — but resist over-reacting to the noise of any single week. Business impact and the ROI loop are a monthly-to-quarterly review, because they accumulate slowly and a short window tells you little. The goal is a rhythm that catches real trends without mistaking normal variance for a signal.
A focused asset has to prove its return, not just its traffic. If you want measurement set up to reach that proof from day one — tracked lines, attribution, and a cost per acquisition you can actually compare — get a free audit, or see the framework applied in our case studies. And for the other half of the discipline, the forecast that this verifies, start with the economics guide and the hub playbook.