Choose a portfolio for additional coverage, not only popular individual items

Define what reached means in this study

Ask the analyst to translate the reach rule into plain language. Does a person count when at least one item exceeds a threshold, when their first choice is present, or through a probability-based method? Preserve the exact rule in the report. A percentage labeled reach can be interpreted as actual media exposure even when it describes only modeled preference coverage.

Sawtooth Software describes TURF as selecting portfolios to maximize respondents reached and offers different reach methods. It notes that several portfolios can have similar results and permits inclusion constraints. These are model-specific design choices. They should be documented before the output is treated as an operational assortment recommendation.

Review the candidate universe and constraints

List the items eligible for selection, those required in every portfolio and those unavailable. Explain each restriction. A mandatory service option, supplier commitment or incompatible pair can materially change the feasible result. Do not hide such constraints by removing items from the file without recording why they disappeared.

Choose the portfolio size from the real decision. Shelf space, production capacity and communication attention impose different constraints, and each may require more than a simple item count. Ask the operations owner whether the proposed combination can actually be delivered. Keep customer appeal and operational feasibility as separate evidence fields rather than assuming the optimization knows every business rule.

Compare combinations, not only item rankings

Request the recommended combination and plausible alternatives with comparable coverage. Inspect which audiences each adds or leaves uncovered, subject to the evidence available. A slightly lower-scoring item may contribute a different group of people, while another popular item mostly duplicates the existing set. The important question is its role within the proposed portfolio.

Ask how sensitive the recommendation is to sample variation, reach definition and uncertain inputs. A tiny difference between near-equivalent portfolios should not become a rigid prescription. Keep the analyst’s uncertainty assessment beside practical considerations such as replenishment, shelf life or production complexity. Do not invent a monetary value for a small coverage difference unless a separate defensible model supports it.

Translate the model into a testable operating choice

Select the portfolio with an explicit explanation of the tradeoffs and any exceptions. Record which items will be offered, where and for how long before review. If a recommended item cannot be stocked, rerun or reassess the feasible scenario rather than quietly substituting a convenient product and retaining the original coverage claim.

Use a suitable pilot to observe real behavior when the stakes justify it. Track availability and implementation fidelity alongside customer response. A weak result may reflect poor execution, a limited preference measure or both. Keep observed purchases separate from modeled reach and use the comparison to improve the next decision, rather than declaring the optimization either infallible or useless.

Make the reasoning concrete.

A fictional workshop provider can promote three class themes. The three most popular themes individually attract largely the same audience, while a less popular practical theme interests a different group.

The provider selects a feasible combination after reviewing overlap and uncertainty. It describes the result as preference coverage in the study and separately measures actual class bookings.

Take the questions into your next brief.

Use these prompts in your own notes. Nothing is submitted from this page.

1. Reach rule

   What must happen for a respondent to count as reached?

2. Universe

   Which items were eligible, excluded or forced in?

3. Constraints

   What limits the feasible portfolio beyond item count?

4. Alternatives

   Which near-equivalent combinations have different operational costs?

5. Validation

   How will real availability and customer response be observed?

Watch for the wrong conclusion.

• Confusing modeled preference coverage with ad impressions.
• Selecting top individual items and calling the result an optimized portfolio.
• Hiding forced inclusions or unavailable options.
• Substituting an item after analysis while keeping the original reach claim.

Sources & scope

Workflows, worksheets and hypothetical examples are original editorial analysis. Platform documentation supports only identified technical principles. No measured business result or named human reviewer is claimed.

• Sawtooth Software: TURF optimization →Primary research-software provider methodology guidance · Supports: Portfolio reach optimization, alternative reach definitions, inclusion settings and near-equivalent portfolios. The operational decision and pilot framework is original editorial analysis.