Strategy
Part of Marketing strategy, sorted out: the 2027 view
Marketing strategy benchmarks: how to vet an outside survey
How to vet an outside survey before using it as a marketing strategy benchmark: who was asked, when and how, and whether the result fits your business.
What to take away
- Treat an outside survey as context for a question, never as a target.
- Check who was asked, how, and when before using any figure.
- Compare the survey with your own customer and economic evidence, and record the decision each benchmark supports.
A survey figure can look like a marketing strategy benchmark, but it describes someone else's population. Universal conversion, acquisition-cost, retention, or budget ratios ignore price, margin, lifecycle, channel role, capacity, category, measurement scope, and maturity. Vet the source first, then decide what the figure can inform.
Ask who was surveyed and how
Survey transparency decides whether a benchmark can be interpreted. The AAPOR Disclosure Standards call for information such as the population, sponsor, sample, collection mode and dates, weighting, processing, quality procedures, and limitations.
Record those details beside the figure, or mark the benchmark as insufficient for the decision. A number with no stated population or field dates cannot be compared with anything your own business does. Keep the record short enough that a colleague will actually read it.
- The population is defined.
- The sponsor is named.
- The sample and collection mode are described.
- Field dates are stated.
- Weighting and processing are explained.
- Limitations are written down.
Read an executive survey with care
The 2025 CMO Survey results report aggregate executive responses about spending, priorities, organization, and capabilities. That makes it context for what executives report, and a weak guide to what your own company should do.
Check the sample, field period, measure wording, subgroup, and compatibility with your business before using a figure as context. A respondent group made up of different kinds of firms answers a different question from yours, even when the headline measure sounds identical.
Use external surveys to frame questions, not to replace the business's own economics and customer evidence. If a figure raises a question, answer it with your own data before it influences a plan.
Audit the market evidence behind the figure
Track research freshness, source coverage, direct-customer contact, response bias, sample adequacy, known limitations, competitor updates, and assumptions awaiting validation. Official demographic and business data can provide context, but local and segment decisions may require direct research.
Market Evidence Benchmarks
- Research freshness
- Source coverage
- Direct-customer contact
- Response bias
- Sample adequacy
- Known limitations
- Competitor updates
- Assumptions awaiting validation
Turn each figure into a question about your own data
Averages can hide a segment receiving slower service or worse outcomes. Use cohorts and distributions when you compare a survey figure with your own customer progress, from awareness through renewal, referral, complaint, refund, and exit.
Review revenue quality, gross margin, contribution, discount, return, service cost, sales time, retention, cash timing, capacity, and full marketing cost. Separate acquisition from retention and new from existing demand. Compare expected, high-use, and downside scenarios before deciding that a peer figure is relevant.
Keep benchmarks apart from cause
A benchmark reports what was observed. It does not show that a change caused a result. When a team needs that answer, the NIST handbook's section on choosing an experimental design describes the process of choosing a design to obtain the results you need.
Keep benchmark reporting and controlled effect estimates as separate work. The first frames a question and the second tests it. Turning observation into causation is how a plausible survey average becomes an unfounded target.
Vetting record
Use one row per survey figure, and attach the record to the decision it informs. If a detail cannot be found, say so in the third column rather than filling the gap with an assumption.
| Vetting question | Detail to record | If it is missing |
|---|---|---|
| Who was asked? | Population, sample, sponsor | Treat as background only |
| When and how? | Collection mode, field dates | Do not compare with current results |
| How were answers shaped? | Weighting, processing, wording | Do not use the figure as a target |
| Does it fit us? | Subgroup, type of business | Use it to frame a question, not a number |
Give every benchmark an owner and an end date
Set a benchmark owner, source, population, date, comparison method, acceptable range, alert threshold, and retirement condition. Preserve the original forecast and benchmark when assumptions change. Realistic planning improves when teams learn why estimates were wrong instead of rewriting them.
Record the decision that each benchmark supports. The numbers then feed back into marketing strategy development, where the positioning and economics they test were chosen. The common marketing strategy questions look at why pitched results and real results differ.
Common questions
What is a good marketing benchmark?
One whose population, definition, method, date, limitations, and business context are close enough to inform a specific decision.
Should an industry average become a target?
No. An average is context, while a target expresses intent and must reflect the company's economics, capacity, customer promise, and risk.
What if a survey does not disclose its method?
Mark the benchmark as insufficient for the decision. At most, use it to frame a question that your own data can then answer.
