A product team with ten flavor ideas and a budget for four faces a real problem. Pick the wrong four, and the launch reaches only a fraction of the market. TURF analysis solves exactly this problem by finding which combination of products, features, or messages reaches the largest possible share of an audience, not just the most popular single option.
Most teams default to ranking choices by popularity alone. That habit misses overlap. Two popular options can appeal to the same buyers, while a less popular one pulls in a completely different audience segment.
In this blog, we will break down how TURF analysis works, walk through real examples, and cover the exact steps to run one using your own survey data.
What is TURF analysis?
TURF analysis, short for Total Unduplicated Reach and Frequency analysis, is a market research method that finds the combination of items, such as products, flavors, features, or ad messages, that reaches the largest possible share of an audience with the least overlap.
Researchers first built the method for media planning, where the goal was reaching the most viewers with a limited number of ad placements. Teams later expanded it to product lines, menu planning, and messaging tests, anywhere a business has to pick a short list from a longer one.
As a market research technique, TURF analysis works alongside tools like surveys and conjoint studies to turn raw preference data into a short list a team can actually act on.
How does TURF analysis work?
TURF analysis works by testing every possible combination of items from a list respondents rated or selected, then ranking each combination by unduplicated reach. Unduplicated reach is the share of people who would choose at least one item in that set.
Frequency, the second half of the acronym, counts how many items in a set each respondent picked. It shows whether interest runs deep across a few items or spreads thin across many.
Analysts also track top-down reach: how much of the audience a brand would reach by offering only its single top-ranked item, before adding anything else to the mix.
Here is a simple illustration. A newsletter survey shows one topic chosen by 30 percent of subscribers. That looks like the obvious winner until a TURF run reveals a different set of three topics, none of them the single most popular choice, that together reaches 78 percent of the list. Reach, not popularity, is the number that actually matters.
Behind the scenes, the calculation is mostly brute force. The model checks every possible group of a given size, adds up how many respondents picked at least one item in that group, and sorts the results from highest reach to lowest. A list of ten items tested in groups of three already produces 120 combinations, which is why this calculation is normally left to software rather than done by hand.
When should you use TURF analysis?
TURF analysis earns its place whenever a team has to cut a long list down to a short one and needs to know the cut will still reach the most people. A few signals point directly to it.
- Limited shelf space, budget, or app real estate forces a choice among many options
- The decision is which three to five items to keep in a lineup, not which single item wins
- Two or more items look similar in survey scores and the team needs to know if they overlap
- The test involves messaging, ad placements, or content topics rather than physical products
- The cost of guessing wrong is high. Nielsen’s Breakthrough Innovation research, reported by Marketing Week, found that more than three in four new FMCG product launches fail within their first year
Not every preference question calls for TURF. When the goal is judging one concept at a time with no combination logic involved, monadic testing is usually the better fit.
TURF analysis vs conjoint analysis: What’s the difference?
TURF analysis and conjoint analysis both start from a preference survey, and that overlap causes real confusion about which one to run. Conjoint analysis measures how people trade off individual attributes like price, size, or color. TURF measures which whole combination of options reaches the most people. For a full breakdown of the trade-off method, see our guide to conjoint analysis.
| Aspect | TURF analysis | Conjoint analysis |
|---|---|---|
| What it measures | Unduplicated reach of a combination | How respondents trade off product attributes |
| Typical question format | Multiple choice, multiple answer | Ranked or rated attribute bundles |
| Best used for | Picking a short list from many options | Pricing and feature trade-off decisions |
| Main output | Ranked combinations by reach percentage | Attribute-level utility scores |
Real-world examples of TURF analysis
The logic behind TURF analysis stays the same across industries, even though the items being tested change completely. These three examples show how the same method applies to a physical product, a menu decision, and a software roadmap.
Launching a new beverage flavor line
A beverage brand preparing a sparkling water launch tests ten flavor concepts but has budget for only four. Surveying 1,000 consumers through QuestionPro Audience, the brand finds its two most popular flavors overlap heavily, appealing to the same drinkers. TURF analysis instead points to a four-flavor set that includes one lower-ranked flavor, because that flavor uniquely attracts a group the top two never reach.
A quick-service restaurant testing new sauce options for a single menu slot runs the same logic. Rather than picking the three highest-rated sauces, TURF shows which three sauces together bring in the largest share of customers who would order at least one item featuring them.
Bundling a SaaS feature release
A software company planning its next release has twelve candidate features but room to ship four. TURF analysis run on customer survey data identifies the four-feature bundle that satisfies the widest slice of the customer base, instead of the four features that simply scored highest one by one.
Step-by-step guide to running a TURF analysis
Running a TURF analysis follows a consistent process regardless of what is being tested. Here is the sequence most research teams follow.
- Define the objective and audience. Decide what needs narrowing down, whether that is products, features, or messages, and who the target respondents are.
- Build a multiple-answer question. Set up a single survey software question that lets respondents select every option they would consider, not just their favorite.
- Collect enough responses. Aim for a sample large enough to represent the target market, since too few responses make small differences between combinations unreliable.
- Run the TURF simulation. Let the tool test every combination at a chosen set size, such as three, four, or five items, and calculate reach and frequency for each.
- Review the ranked combinations. Compare the top few results rather than only the single highest one, since results often cluster close together.
- Make the call and validate. Choose the combination that fits both the reach data and practical constraints like cost or production capacity.
How much does TURF analysis cost?
TURF analysis itself has no fixed price. What it costs depends on how it gets run. Building the model manually in a spreadsheet or statistics package takes analyst time, while running it inside survey software with a built-in simulator adds no extra cost beyond an existing research license.
QuestionPro’s TURF simulator runs directly on multiple-answer survey questions, so a team already using QuestionPro Market Research Software for data collection does not need separate software or a specialist to calculate reach and frequency by hand.
Common mistakes to avoid in TURF analysis
A few recurring mistakes quietly weaken TURF analysis results. Watch for these before trusting the output.
- Testing too many items at once, which dilutes the signal and makes most combinations look similar to each other
- Treating the single most popular item as the automatic top pick without first checking how much it overlaps with other choices
- Ignoring frequency data and looking only at reach, which hides how deep interest actually runs within the audience
- Running the survey on a sample too small or too skewed to represent the real target market; see our guide on survey bias for ways to keep sampling clean
- Locking in a combination without checking whether the runner-up options were nearly identical in reach
- Skipping a re-test after launch, even though preferences shift and last year’s winning combination may not hold up today
How do you measure and evaluate TURF analysis results?
Evaluating TURF results means comparing three numbers side by side for each combination: unduplicated reach, the share of people covered by at least one item; frequency, how many items per person on average; and top-down reach, what the single best item alone would deliver. A combination is worth choosing when its reach clears top-down reach by a meaningful margin, not by a rounding error.
It also helps to check how close the top three or four combinations sit to each other. When several combinations land within a point or two of one another, practical factors such as production cost, brand fit, or timeline should decide the tie.
A simpler way to think about TURF
TURF analysis will not tell a team which single idea is best loved. It tells them which set of ideas, taken together, leaves the fewest people out.
That shift, from picking a favorite to covering a market, is what makes the method worth the extra step over a simple popularity count.
Frequently Asked Questions (FAQs)
No. TURF analysis works equally well for messaging, ad placements, software features, and service bundles. Any scenario built on a multiple-answer survey question, where respondents pick every option they would accept, can be analyzed with TURF logic.
Most teams test between eight and fifteen items. Fewer rarely justifies running the analysis, while more than twenty makes the number of possible combinations too large to interpret clearly without narrowing the list first.
Plan for at least 200 to 300 completed responses for a typical study, and more if testing many items or breaking results out by customer segment. Small samples make close combinations impossible to tell apart with confidence.
Yes, using a spreadsheet or a statistics package such as R, though it requires manually building every combination and calculating reach by hand. Survey platforms with a built-in TURF simulator handle that calculation automatically.
Not on its own. TURF measures reach based on stated preference, not price sensitivity. Teams that need to weigh cost trade-offs alongside reach typically pair TURF with conjoint analysis or a separate pricing study.
Consumer packaged goods, food and beverage, and media planning teams use it most, since they routinely pick a limited lineup from a longer list of flavors, SKUs, or ad placements. Retail and SaaS teams have adopted it too.
