Blog · Board and management reporting
Market research KPIs compare usable responses, project cost, delivery, and documented business use. Work through a project-level example and the checks that keep quality and speed in balance.
A KPI market research teams can act on connects usable evidence, timely delivery, business use, and cost. Track those dimensions together. A cheaper completed questionnaire is not a better result when it fails the study's eligibility or quality rules.
The research lead owns the measurement definitions. Project managers maintain dates and status, finance supplies approved and actual costs, and the business sponsor confirms how findings were used. Give every measure an owner who can explain the underlying rows and act on the result.
A key performance indicator earns that status because it changes a decision. Keep descriptive counts available, but avoid making every count a target; the distinction between a KPI, metric, and measure helps keep the review focused.
| Management question | Measure | Calculation | Decision it supports |
|---|---|---|---|
| Are completed questionnaires usable? | Usable-response yield | Accepted responses / completed questionnaires | Review recruitment and quality exclusions |
| What did usable evidence cost? | Cost per usable response | Actual project cost / accepted responses | Review scope, methods, and supplier costs |
| Did agreed work arrive? | Project completion | Delivered projects / projects due | Resolve outstanding deliverables |
| Did it arrive by the agreed deadline? | On-time delivery | Projects delivered on time / projects due | Change scheduling or escalation |
| Was the evidence considered? | Documented business use | Projects with documented use / projects due for review | Improve briefing and sponsor follow-through |
These are management definitions for the example below. They are not a universal research-quality score. AAPOR's Standard Definitions cautions that no single number captures overall survey quality and that response-rate information alone cannot establish the amount of nonresponse error.
Use a project ID that appears in the research register, cost file, and results summary. Export the original agreed deadline, actual delivery date, approved budget, actual cost, method, audience, and business sponsor. Keep revised deadlines separately so a late project cannot become on time merely through an overwritten date.
For survey work, summarize completed questionnaires, accepted responses, and exclusions by reason. Establish eligibility and quality rules before reviewing results. Keep those rules attached to the project definition; changing them halfway through a field period changes the meaning of the yield.
The Census Bureau's Statistical Quality Standards require pretesting and refinement of new data-collection instruments within their scope. That is a useful discipline for commercial research too: investigate confusing questions and routing before trying to solve a weak yield by buying more responses.
For business use, record a review-due flag and a documented-use flag. A project whose sponsor review is still pending is not yet a failed-use project. Export project-level summaries for this review; respondent names and contact details are unnecessary.
The research team reviews three completed projects after their delivery deadlines. Actual cost includes supplier invoices and internal labor allocated under the same policy as the approved budget. All currency amounts are USD. These are illustrative management records, not external benchmarks.
| Project | Completed questionnaires | Excluded | Accepted responses | Budget (USD) | Actual cost (USD) |
|---|---|---|---|---|---|
| Concept test | 240 | 24 | 216 | 18,000 | 19,440 |
| Packaging test | 180 | 36 | 144 | 14,400 | 12,960 |
| Usage study | 300 | 15 | 285 | 21,600 | 22,800 |
| Total | 720 | 75 | 645 | 54,000 | 55,200 |
The response check is completed questionnaires minus exclusions equals accepted responses: 720 - 75 = 645. The cost file must reconcile to the $55,200 included in this review, with any unallocated costs listed separately.
Usable-response yield = Accepted responses / Completed questionnaires
Cost per usable response = Actual cost / Accepted responses
Budget variance = Actual cost - Approved budget
| Project | Yield calculation | Yield | Cost per usable response (USD) | Budget variance (USD) |
|---|---|---|---|---|
| Concept test | 216 / 240 | 90.0% | 19,440 / 216 = 90.00 | 1,440 |
| Packaging test | 144 / 180 | 80.0% | 12,960 / 144 = 90.00 | -1,440 |
| Usage study | 285 / 300 | 95.0% | 22,800 / 285 = 80.00 | 1,200 |
| Total | 645 / 720 | 89.6% | 55,200 / 645 = 85.58 | 1,200 |
Percentages round to one decimal place and unit costs to cents. Calculate totals from the summed inputs, retaining full precision until display. The total cost overrun is $55,200 - $54,000 = $1,200, or $1,200 / $54,000 = 2.2% of budget.
Delivery and use come from a separate project register. Each flag below means yes when it is 1 and no when it is 0. Review due means the agreed sponsor follow-up date has passed.
| Project | Due | Delivered | On time | Review due | Documented use |
|---|---|---|---|---|---|
| Concept test | 1 | 1 | 1 | 1 | 1 |
| Packaging test | 1 | 1 | 0 | 1 | 0 |
| Usage study | 1 | 1 | 1 | 0 | 0 |
| Total | 3 | 3 | 2 | 2 | 1 |
Project completion is 3 / 3 = 100.0%, but on-time delivery is 2 / 3 = 66.7%. Documented business use is 1 / 2 = 50.0%; the usage study remains outside that denominator until its review is due.
The packaging test spent less than budget but delivered late and had no documented use by its review date. Its lower spend does not establish success. Ask whether the delay missed a decision meeting, whether the sponsor received the findings, and whether a decision occurred without being logged.
The usage study has the strongest usable-response yield and lowest unit cost, but that does not prove its method is better for a different audience. Review method and scope before changing suppliers. Documented use also does not prove financial return: a sponsor can use a finding without a defensible estimate of incremental revenue.
The market research dashboard covers how to organize the recurring review. For evidence about continuing customer behavior rather than project execution, use the separate customer KPIs review.
Bring a project register, aggregated response-quality summary, and project-cost export. Covirage maps the columns; its tools compute and check the comparisons, and the external AI model explains the results with every figure traced to its rows. Create a free Covirage account, which takes about a minute, and bring a sample of your own data after setting up your business.
See self-service analytics for the supporting approach to answering business questions from your files.
Choose the measure tied to the research decision. Usable-response yield helps a survey operations manager, while delivery before a launch decision matters to the sponsor. Pair the primary measure with a quality check and cost measure so improving speed does not conceal weaker evidence.
No. A completion measure may divide completed questionnaires by starts, whereas a response rate uses defined eligible cases and treatments of unknown eligibility. Record the exact denominator. Neither a high completion percentage nor a high response rate alone establishes that the findings represent the intended population.
Keep a decision log with the project ID, review date, decision owner, finding considered, and resulting decision or documented reason for no change. Divide projects with documented use by projects whose review is due. A decision to retain an existing approach can count when the evidence informed it.
Compare projects with similar methods, audiences, scope, and cost definitions. A specialist business interview and a broad consumer survey buy different evidence. Track changes within a comparable group before using cost per response to judge suppliers or move the research budget.