What Is TURF Data Analysis?
TURF data analysis is the process of interpreting Total Unduplicated Reach and Frequency results to identify the combination of items (products, flavors, features, or messages) that reaches the maximum number of unique consumers with the fewest options. TURF starts from binary or scaled preference data (which items would each respondent accept, buy, or consider) and algorithmically tests combinations to find the set that maximizes unduplicated reach, meaning the percentage of respondents who find at least one item in the set appealing. The output isn't a simple ranking of individual items by popularity. It's an optimization of the portfolio as a whole, accounting for overlap between items that appeal to the same people.
Why TURF Data Analysis Matters
Offering the top 5 most popular items individually doesn't guarantee you're reaching the most people. Two items might be loved by the same 30% of consumers while a less popular item uniquely appeals to a segment nobody else covers. TURF finds these non-obvious combinations. A CPG study by IRI found that TURF-optimized product lines reached 15-20% more unique consumers than lines built by simply choosing the highest-rated individual items. For any business with shelf-space constraints, menu limitations, or portfolio budgets, TURF directly translates into more efficient assortment decisions.
How TURF Data Analysis Works
The Input Data
TURF requires a matrix showing which respondents would accept (or purchase, or consider) which items. For a flavor study, you might ask 500 respondents to check all flavors they'd be willing to buy from a list of 15. Each respondent's row has a 1 or 0 for each flavor. This binary acceptance data is the standard input. Some TURF implementations also work with scaled data (acceptance threshold set at, say, top-2 box on a 5-point scale), which you convert to binary by coding responses above the threshold as 1 and below as 0.
How the Algorithm Works
TURF evaluates every possible combination of items at each portfolio size. For a combo of 3 items from 15, that's 455 combinations. For each combination, it calculates the unduplicated reach: the percentage of respondents who accept at least one of the items in that combination. It also calculates frequency, the average number of items accepted per reached respondent. The optimal combination at each portfolio size is the one that maximizes reach.
The first item selected is always the single most popular item. The second item isn't the second most popular overall. It's the one that adds the most incremental reach by uniquely appealing to respondents who didn't accept the first item. Each subsequent item is chosen for its incremental contribution, not its standalone popularity.
Reading the Output
TURF output is typically presented as a table or chart showing cumulative reach at each portfolio size.
For a beverage flavor study: 1 flavor (vanilla) reaches 42%. Adding a second (chocolate) brings reach to 63%. The third (strawberry) pushes it to 74%. The fourth (mango) reaches 80%. The fifth (mint) hits 83%. Notice that each additional item adds less incremental reach than the previous one. The diminishing returns curve tells you where the portfolio hits a practical ceiling. Going from 4 to 5 items gained only 3 percentage points, signaling that mango might be the natural stopping point if shelf space is limited.
Frequency Analysis
Reach tells you how many people you've covered. Frequency tells you how deeply. If a 4-item portfolio reaches 80% of respondents and the average reached respondent accepts 2.3 of the 4 items, your frequency is 2.3. High frequency means the portfolio creates choice within your line (good for repeat purchasing). Low frequency with high reach means you've covered many people but each person only has one option they like (efficient reach but less depth).
A Worked Example
A snack company tested 18 flavor concepts with 600 consumers. The single most popular flavor (barbecue) reached 48% of respondents on its own. A TURF analysis found that the optimal 5-flavor portfolio was barbecue, salt and vinegar, jalapeno, ranch, and honey mustard, reaching 87% of respondents. The naive approach of selecting the top 5 individual flavors (barbecue, cheddar, salt and vinegar, ranch, sour cream) reached only 79% because barbecue and cheddar had 72% audience overlap. Swapping cheddar for jalapeno and sour cream for honey mustard added 8 points of reach by covering underserved taste segments.
Sensitivity and Threshold Testing
Run TURF at multiple acceptance thresholds to test robustness. If your results were based on "definitely would buy" (top-box), re-run with "probably or definitely would buy" (top-2 box). If the optimal portfolio changes substantially, your results are sensitive to how strictly you define acceptance, which warrants further investigation.
When to Use TURF Data Analysis
- Product line optimization selecting which SKUs, flavors, or variants to offer within shelf-space or manufacturing constraints
- Menu design choosing the combination of menu items that appeals to the widest range of diners
- Message portfolio selection picking which 3-4 value propositions to feature in a campaign from 10+ tested messages
- Feature bundling determining which feature combinations cover the most user needs across customer segments
- Retail assortment planning optimizing category planograms to maximize the percentage of shoppers who find something they want
Common Mistakes
- Confusing individual item popularity with portfolio optimization and defaulting to the top-N most popular items without running the combinatorial analysis that accounts for overlap
- Setting the acceptance threshold too low (including all positive responses) which inflates reach numbers and makes every portfolio look good, masking meaningful differences between combinations
- Ignoring practical constraints like cost, production complexity, or brand fit when presenting TURF-optimal portfolios; always filter results through business feasibility before making recommendations
How Quali-Fi Supports TURF Data Analysis
Quali-Fi's Research plan includes TURF analysis as a built-in analytical tool. Design your product acceptance or concept screening survey, and the platform automatically runs TURF optimization across all possible combinations, displaying reach curves and optimal portfolios at each size. You can filter by segment and adjust acceptance thresholds directly in the dashboard.
Frequently Asked Questions
How many items can TURF handle?
TURF works well with 8-30 items. Below 8, manual comparison is feasible. Above 30, the number of possible combinations grows astronomically, though modern software handles this through efficient search algorithms rather than exhaustive enumeration. The practical limit is more about survey design (respondents can reliably evaluate 15-25 items) than computational capacity.
Can TURF account for items I must include?
Yes. Most TURF implementations let you "lock in" required items and optimize only the remaining slots. If barbecue is already in production and can't be removed, lock it in and let TURF optimize the additional 4 slots around it.
What's the difference between TURF and conjoint for portfolio decisions?
TURF optimizes which items to offer based on acceptance data. Conjoint optimizes how to configure each item based on attribute trade-offs. Use TURF when you're choosing between defined options (which flavors to stock). Use conjoint when you're designing the options themselves (what features should each product variant include).
Related Topics
- TURF Analysis
- MaxDiff Data Analysis
- Conjoint Analysis Data Interpretation
- Cross-Tabulation Analysis
- Sample Size Formula
- Data Collection Methods
Optimize your product portfolio with built-in TURF -- try Quali-Fi free for 14 days.