Market research tool
TURF analysis calculator
How results are checked
Named statistical review pending. Selected automated reference cases are published on the validation page.Best 3-flavor line-up
80.3%reach
Sea salt + Chili lime + Salt & vinegar
482 of 600 respondents · frequency 1.26 · 56 combinations tested
Best reach by line-up size
TURF analysis (Total Unduplicated Reach and Frequency) finds the set of products, flavors or features that appeals to the most people at least once. Instead of picking the individually most popular items — which often appeal to the same respondents — it counts each respondent once and ranks combinations by unduplicated reach, using frequency to break ties.
Run TURF on your data
Open a respondent-level file with one column per item (for example purchase interest on a 1–5 scale), tick the items, choose what counts as "reached" and the portfolio size. The analysis runs in this browser tab — free, no account, and the file is never uploaded. "Use sample data" loads a synthetic snack-brand study with eight candidate flavors.
CSV, Excel, SPSS, Stata, SAS, R, Parquet, SQLite or JSON. The file is read in this browser tab and never uploaded.
Why popularity alone picks the wrong line-up
Two flavors that each appeal to 40% of buyers add nothing to each other if they appeal to the same 40%. TURF looks at overlap: a respondent counts toward reach once, no matter how many items in the portfolio appeal to them. A niche item that reaches people the rest of the line misses can therefore beat a broadly liked item that duplicates existing appeal.
The calculator evaluates every combination when there are at most 50,000 of them, and otherwise falls back to greedy forward selection, which it labels as an approximation. Respondents who skipped every selected item are left out of the base. For the underlying crosstabs of item appeal by segment, use the survey crosstab tool.
The method
Formula
Reach(S) = Σ wᵢ · 1[i reached by ≥ 1 item in S] / Σ wᵢ
where
- S
- a candidate portfolio of k items
- wᵢ
- respondent weight (1 when unweighted)
- Frequency(S)
- Σ wᵢ · (number of items in S reaching i) / Σ wᵢ — the mean number of portfolio items that reach each respondent
- Incremental reach
- Reach(S ∪ {item}) − Reach(S): the new respondents an added item brings in
Worked example
In the synthetic sample (600 respondents), purchase interest for eight flavors is rated 1–5 and a respondent is "reached" by a flavor at top-2-box (4 or 5). The most popular single flavor is sea salt at 48.7%, followed by cheddar (40.0%) and chili lime (39.3%).
Evaluating all 56 three-flavor combinations, the best line-up is sea salt + chili lime + salt vinegar, reaching 80.3% (482 of 600) with a frequency of 1.26. The runner-up, sea salt + chili lime + cheddar, reaches 78.8%. Simply taking the three most popular flavors reaches 78.8%.
The reach ladder shows diminishing returns: sea salt alone reaches 48.7%; adding chili lime brings +20.5 points, then salt vinegar +11.2, and the fourth item only +6.0.
Using it well
When to use it
Use it when
- Use: Choosing which product variants, flavors, or SKUs to launch or keep in a line.
- Use: Picking a set of features, messages, or media vehicles that together reach the widest audience.
- Use: You have respondent-level ratings or yes/no appeal for every candidate item.
Not the right tool when
- Avoid: You only have aggregate percentages per item — TURF needs to know which respondents overlap.
- Avoid: Volume or revenue matters more than breadth of appeal; reach treats a light and a heavy buyer the same.
- Avoid: Items cannibalise each other or depend on price — use a choice-based method instead.
How to interpret it
Rule of thumb
Pick the smallest portfolio where the reach ladder flattens out: once an extra item adds only a point or two of reach, it mostly re-reaches people the line already covers. Among combinations with near-equal reach, prefer the higher frequency (more respondents with several appealing options).
Questions
Frequently asked questions
What does TURF stand for?
Total Unduplicated Reach and Frequency. Reach is the share of respondents reached by at least one item in a combination, counting each person once; frequency is the average number of items in the combination that reach each respondent.
Should I use top-box or top-2-box as the reach rule?
Top-2-box (for example 4 or 5 on a 5-point purchase-interest scale) is the common default. Top-box is stricter and suits cases where only definite interest counts. If your data already has a 0/1 'would buy' flag per item, choose the binary rule.
Why is the result sometimes labelled greedy?
With many items and a large portfolio the number of combinations explodes. Above 50,000 combinations, the calculator adds one item at a time, each time choosing the item with the largest reach gain. Greedy selection is fast and usually close to the best, but it is not guaranteed to find the optimal combination.
How are missing answers handled?
A missing answer on an item counts as not reached by that item. Respondents who skipped every selected item are excluded from the base, and the base size is reported with the results.
Can I force an item into every combination?
Yes. Mark an item 'Must include' to keep it in every portfolio (for example an existing flagship product), or 'Exclude' to leave it out of the search. Required items also enter the reach ladder first.
Sources and next steps
References & further reading
- [1]Miaoulis, G., Free, V., & Parsons, H. (1990). TURF: A new planning approach for product line extensions. Marketing Research, 2(1).
Try it on your own data — free, no signup
Upload a CSV, Excel, SPSS, Stata, SAS, R, Parquet or JSON file. Everything runs in your browser; your file never leaves your device.
Open the workspace →