Conjoint Analysis and Pricing Research for Caribbean Markets

Pricing decisions in Caribbean markets carry extra weight. Get it wrong in a 1.4-million-person market like Trinidad and you can't hide behind volume. The margin hit is immediate and the competitive response comes fast in markets where every brand manager knows every other brand manager.

We run the full range of pricing and product optimization techniques: choice-based conjoint (CBC), Van Westendorp, Gabor-Granger, MaxDiff, TURF and Kano analysis. The question isn't which technique exists. It's which one fits the business decision, the budget and the sample size your fieldwork can actually deliver in-market.

Choice-Based Conjoint (CBC)

CBC is the right tool when the purchase decision involves trade-offs between multiple attributes. Telecom bundle design is the classic Caribbean application: a customer choosing between plans that differ on data cap, speed, price, contract length and included streaming services. The conjoint model estimates the utility of each attribute level and predicts market share for any hypothetical product configuration.

Sample Requirements

CBC needs a minimum of n=300 per market for aggregate-level analysis. If you want segment-level utilities (and you usually do), plan for n=500+. For a three-market Caribbean study with segment analysis, that's 1,500 interviews before you account for quotas and incidence rates. This is a real constraint. In smaller Eastern Caribbean markets, hitting n=300 on a specific target population may require mixed-mode fieldwork or extended field periods.

Design Considerations

The attribute list is where most conjoint studies succeed or fail. Six attributes with 3-4 levels each is a practical ceiling for Caribbean respondents completing the survey online. Go beyond that and task complexity produces random responding, especially on mobile devices. If you have eight attributes that all feel critical, consider a two-stage approach: MaxDiff to identify the four most important, then conjoint on those four.

Van Westendorp Price Sensitivity Meter

Van Westendorp is the workhorse for straightforward pricing questions. Four questions ("at what price would this be too expensive," "too cheap," "expensive but still worth considering," "a bargain") produce a price sensitivity curve with an optimal price point, a range of acceptable prices and indifference points.

It works with smaller samples than conjoint (n=200 is workable) and adds only 2 minutes to a survey. For a single product with a clear value proposition (a new beverage SKU, a basic service tier, a consumer electronics accessory), Van Westendorp gives you a defensible price range fast.

The limitation: it doesn't capture feature trade-offs. If the pricing decision depends on which features are included at which tier, you need conjoint or Gabor-Granger. Van Westendorp assumes the product is fixed and only the price varies.

Gabor-Granger

Gabor-Granger estimates a demand curve by testing purchase intent at discrete price points. More useful than Van Westendorp when you need revenue optimization (price x volume) rather than just price acceptability.

The technique presents a series of prices in sequence (usually ascending) and measures purchase likelihood at each. From this you can model the revenue curve and identify the price that maximizes revenue, not just the price consumers find acceptable.

Sample requirements scale with the number of price points tested. Four price points work fine at n=200. Eight or more price points need n=400+ to keep the estimates stable at each level. In Caribbean research we typically test five or six price points, which balances analytical precision against survey length.

MaxDiff (Best-Worst Scaling)

MaxDiff forces respondents to discriminate between items instead of rating everything "very important" (which is what happens with 5-point Likert scales in Caribbean studies). Each task shows a set of items and asks which is most important and which is least important. The output is an interval-scale ranking that tells you how much more important item A is than item B, not just that both scored 4.2 out of 5.

We use MaxDiff for feature prioritization, message testing, brand attribute ranking and any situation where the client needs a clear hierarchy. It's especially valuable in Caribbean markets where cultural politeness and acquiescence bias produce flat distributions on rating scales. MaxDiff cuts through that.

Sample requirements are modest: n=200 per market for a 15-item MaxDiff with 4 items per task. Survey burden is low (respondents find the tasks intuitive) and the analytical output is clean.

TURF Analysis

TURF (Total Unduplicated Reach and Frequency) answers a specific question: given a constraint on how many options you can offer, which combination maximizes the number of people who find at least one option acceptable?

The Caribbean FMCG application is direct. A brand launching a line extension with budget to support four SKUs out of eight candidates needs to know which four maximize unduplicated reach. TURF identifies that combination. The data comes from a standard concept evaluation survey. The analysis is the differentiator.

TURF is one of the highest-ROI techniques available for portfolio decisions and we think it's underused in the Caribbean. The data collection cost is minimal (it piggybacks on concept testing you're probably doing anyway) and the output directly informs a decision with real money behind it.

Kano Model

Kano classifies product features into categories: must-haves (expected, their absence causes dissatisfaction), performance features (more is better, linear relationship with satisfaction) and delighters (unexpected features that create disproportionate satisfaction).

The technique uses paired functional/dysfunctional questions for each feature. It's useful early in product development when you need to know which features are table stakes versus differentiators. For Caribbean financial services and telecom, where products tend to converge on the same feature set, Kano identifies where a brand can create separation.

Choosing the Right Technique