PPC KPIs: Compare Cost, Conversion, and Revenue

PPC KPIs connect paid-search cost to clicks, conversions, acquired customers, and attributed revenue. Start with the business outcome, then use the traffic and conversion measures to explain why it changed. A cheaper click is useful only when the downstream result still meets the company's requirements.

Choose the outcome before building the scorecard

A KPI needs an owner, a definition, and an action when performance misses its target. Impressions and clicks describe activity; customer acquisition cost and attributed revenue bring that activity closer to a business decision. The distinction between a KPI, metric, and measure helps keep the scorecard focused.

For paid search, retain these measures together:

Google Ads describes these search campaign measures, including CPC as cost divided by clicks. Its conversion-rate definition uses interactions that can be tracked to a conversion, so retain that denominator when reproducing a platform figure. The example below assumes every reported click is eligible.

Bring matching campaign and customer rows

Export campaign ID, campaign name, reporting dates, currency, impressions, clicks, eligible interactions, cost, conversion action, conversions, and conversion value. Record the attribution rule, counting setting, and export date alongside the file. Use campaign IDs to preserve continuity when names change.

Bring a second export containing unique lead IDs, campaign attribution, qualification status, customer status, and the sales value assigned under the company's reporting policy. Reduce this file to the campaign level before combining it with campaign cost. Otherwise, repeating a campaign's spend beside every lead multiplies the numerator.

Check whether conversion value represents actual sales or an assigned estimate for an action. Google explains that conversion values and related reporting columns depend on the conversion actions and values supplied. An estimated value for a lead is not booked revenue. The same documentation distinguishes reporting by interaction time from reporting by conversion time; select and label the basis that answers your question.

A worked paid-search scorecard

Fictional Ohio distributor Summit Industrial Supply compares September 2026 campaign activity, with outcomes frozen on October 23, 2026. Its selected conversion is a unique sales inquiry. New customers and sales value are assigned to those inquiries using one documented attribution rule; the figures are illustrative company inputs, not benchmarks.

Divide the corresponding fields to obtain the scorecard. Percentages and dollar ratios below are rounded to two decimal places; ROAS is shown to three decimal places.

For category search, 175 / 7,000 = 2.50% and 14,000 / 175 = $80.00 per inquiry. Acquisition cost is a different calculation: 14,000 / 35 = $400.00 per new customer. That difference is why a lead target must not silently become a customer target.

The total ROAS is 82,500 / 20,000 = 4.125x. It measures attributed sales relative to advertising cost, without deducting product costs, sales labor, or overhead. It therefore does not establish profit or incremental return.

Prove the total before interpreting it

Sum the input columns, then divide. The blended cost per inquiry is 20,000 / 325 = $61.54; do not average campaign cost ratios. Carry full precision through calculations and round only the displayed results.

There is a useful identity:

Cost per inquiry = Cost per click / Inquiry conversion rate

For the total, 2 / 0.0325 = $61.54. Reconstructing spend gives the same answer: 10,000 clicks multiplied by $2.00 = $20,000. Reconcile that spend to the campaign export for the selected period, and explain any difference from invoices caused by credits or billing dates.

Separately reconcile the inquiry count to unique lead IDs, the customer count to customer records, and attributed sales to the assigned sales rows. Keep unmatched records visible. A complete-looking report built by dropping unmatched customers is harder to challenge and less useful.

Assign decisions and owners

The paid-search manager owns campaign cost, traffic definitions, and the first explanation of a conversion-rate change. Sales operations owns inquiry acceptance and customer matching; finance agrees the revenue and cost scope. For the next stage of lead quality, use the separate MQL KPI scorecard, rather than expanding this report into every sales measure.

Brand search has lower acquisition cost in this example, but that does not automatically justify moving category-search spend. The audiences have different intent, and existing awareness can influence brand demand. Review capacity to spend, customer quality, margin, and evidence from controlled measurement before changing the budget.

Where PPC KPI reporting goes wrong

• Mixing conversion actions. Adding inquiries and purchases produces a count with no consistent economic meaning. Report each action separately.
• Comparing immature periods. Recent clicks have had less time to produce customers. Keep a common observation policy and label provisional outcomes.
• Joining spend to individual leads. Repeated campaign cost inflates totals. Aggregate each source to the same level first.
• Calling attributed revenue incremental revenue. Channel credit records an allocation rule; it does not establish what sales would have occurred without advertising.
• Treating missing outcomes as zero. Missing customer matches require investigation. They do not prove that the campaign produced no customers.

Calculate PPC KPIs on your own file in Covirage

Bring a campaign export and a sample of your own matched lead or sales data; Covirage maps the columns, and its tools compute and check the figures. The external AI model explains the results with every figure traced to its rows. Create a free Covirage account, which takes about a minute, then set up your business and bring your file.

See self-service analytics for the supporting approach.

Covirage is not affiliated with Google.