Consideration-Set Model
Also known as: Consideration Set Composition Model, Consider-Then-Choose Model, Two-Stage Choice Model, Evoked Set Model
Consideration-set models formalize the empirical fact that consumers do not evaluate every available brand but choose from a small subset they actively consider. Choice is decomposed into two stages: first a brand is screened into the consideration (or evoked) set, then it competes for selection only against the other considered brands. John Roberts and James Lattin's 1991 model gave this idea a rigorous, estimable form by treating consideration as the outcome of a benefit-cost calculus — a brand is added to the set when the expected incremental benefit of including it exceeds a cost of consideration. The conditional second stage is typically a logit over the considered brands, so the unconditional choice probability is a weighted sum over possible consideration sets. Modeling the first stage matters because ignoring it biases estimated brand effects and substitution patterns: a brand can lose because it is never considered, not because it loses head-to-head. The framework underlies modern thinking about awareness, screening, and the upper funnel in brand competition.
Key highlights
- Separates non-consideration from unfavorable evaluation, so low share can be correctly attributed to awareness/screening versus product/price.
- Grounds set formation in an explicit benefit-cost calculus rather than an arbitrary inclusion threshold, making consideration predictable from attributes.
- Corrects the bias in single-stage choice models that arises from ignoring which brands are actually evaluated.
- Connects marketing levers (advertising, distribution, awareness) to the consideration stage and product/price levers to the choice stage.
Intuition
This section is available to Pro members. Upgrade to Pro
How it works
This section is available to Pro members. Upgrade to Pro
When to use it
Use a consideration-set model when there is good reason to believe consumers choose from a limited subset of available brands and you want to separate why a brand is not chosen — because it is never considered versus because it loses among considered brands. It is most valuable in categories with many alternatives, meaningful search or recall costs, and managerial questions about awareness, distribution, and the upper purchase funnel. It works best with individual-level data: survey measures of which brands are considered, recall or screening tasks, or panel data rich enough to infer consideration, combined with brand attributes. It is less necessary when choice sets are small enough that everyone effectively considers everything, when you only have aggregate share data with no leverage on the screening stage, or when a single-stage logit or nested logit already captures the substitution patterns of interest at acceptable bias.
Strengths & limitations
- Separates non-consideration from unfavorable evaluation, so low share can be correctly attributed to awareness/screening versus product/price.
- Grounds set formation in an explicit benefit-cost calculus rather than an arbitrary inclusion threshold, making consideration predictable from attributes.
- Corrects the bias in single-stage choice models that arises from ignoring which brands are actually evaluated.
- Connects marketing levers (advertising, distribution, awareness) to the consideration stage and product/price levers to the choice stage.
- Consideration sets are usually latent, so identification leans on strong assumptions or on direct but error-prone survey measures of consideration.
- Summing over all possible consideration sets is combinatorially expensive as the number of brands grows, forcing simplifying assumptions.
- The model assumes a particular benefit-cost mechanism for screening that may not match heuristic or context-driven consideration in practice.
- Individual-level data rich enough to estimate both stages can be costly to collect, limiting application on purely aggregate data.
Common pitfalls
This section is available to Pro members. Upgrade to Pro
Applications
This section is available to Pro members. Upgrade to Pro
Frequently asked
Why model consideration separately instead of just running a logit over all brands?
Because a one-stage logit assumes every consumer evaluates every brand, it misattributes the reasons for non-purchase. A brand that is simply never on the shortlist looks, to a single-stage model, like a brand that consumers weigh and reject, biasing its estimated attractiveness and the implied substitution patterns. Consideration-set models split 'considered?' from 'chosen given considered?', so a manager learns whether the fix is awareness, salience, and distribution (the consideration stage) or product, price, and positioning (the choice stage). These are different strategies, and conflating them with one logit can point investment in the wrong direction.
How does the Roberts-Lattin benefit-cost rule decide what to consider?
A brand is added to the consideration set when the expected incremental benefit of having it as an option exceeds the cost of considering it. The benefit is how much the expected best-available utility rises once the brand is included, so a brand that merely duplicates options already on the list adds little and is screened out, capturing diminishing returns to set size. The cost reflects the cognitive and search effort of evaluating another option. Because utilities are random, this produces a probability that each brand is considered, which can be tied to awareness, prior usage, and marketing exposure rather than an arbitrary cutoff.
What data do I need to estimate a consideration-set model?
Ideally individual-level data with both stages observable or inferable: which brands each consumer considered (from survey recall, screening tasks, or rich panel histories) plus the brand actually chosen and the attributes, price, and promotion of the alternatives. With direct consideration measures the two stages can be estimated more cleanly; without them, the consideration set is latent and the model marginalizes over possible sets, which is identified from choice patterns but rests on stronger assumptions. Purely aggregate share data give little traction on the screening stage, so consideration-set modeling is mainly a disaggregate technique.
Sources
- 1.Roberts, J. H., & Lattin, J. M. (1991). Development and Testing of a Model of Consideration Set Composition. Journal of Marketing Research, 28(4), 429-440.
- 2.Guadagni, P. M., & Little, J. D. C. (1983). A Logit Model of Brand Choice Calibrated on Scanner Data. Marketing Science, 2(3), 203-238.
You have read it. What now?
Cite this page
ScholarGate. (2026, June 23). Consideration-Set Model. ScholarGate. https://scholargate.app/marketing/consideration-set-model