Market & customer research / NOMAS

Product Testing in Saudi Arabia: Evaluating Concepts, Taste and Packaging

A practical guide to separating concept, sensory and packaging tests—and to distinguishing stated liking and purchase intent from observed buying behaviour.

By NOMASPublished 2 October 20267 MIN READ

Product testing stages from concept and taste to packaging and launch decisions

Start with the decision, then separate the questions

Product testing is not a single “Did you like it?” question. Before launch, a team may need to learn whether a concept is clear and relevant, whether the product delivers a desirable experience, or whether packaging communicates useful information and helps people choose. These questions connect, but they are not interchangeable: an idea may sound appealing before use and feel different after trial; a product may be liked in a blind test while its pack fails to distinguish it at the point of choice.

Write down the decision the research should inform and what could change as a result. Is the team reviewing the promise, audience, formulation or packaging direction? Is it choosing between alternatives, identifying what to improve, or deciding whether another test is needed? This determines what participants see, when they see it and which measures can answer the question.

Distinguish concept, sensory experience and packaging

A **concept test** presents a description or proposition and explores comprehension, relevance, distinctiveness and unanswered questions. The description should be balanced and should not promise what the product cannot deliver. Interviews can help explain participants’ language and expectations; structured measures may compare defined responses when the decision calls for it. Liking an idea does not establish that someone will buy it.

A **sensory test** focuses on what a person actually experiences, such as taste, aroma or texture where relevant. Standardise preparation and serving conditions as far as practical, and allow appropriate time between samples when needed. A blind test hides brand identity or associated information to reduce the influence of expectations and brand cues. It does not, by itself, answer how the product performs under ordinary purchase conditions. A branded test reveals the experience when participants know the identity, but prior perceptions may affect responses. A two-stage comparison can be useful when that difference is central to the decision.

A **packaging test** explores reading, comprehension, impression, distinctiveness or findability, using visual or physical stimuli as the question requires. Do not equate choosing a design on a screen with noticing a pack during a real choice; the display method, option order and viewing distance are part of the test conditions. If the question is whether a message is understood, first ask participants what they understood before explaining the intended meaning.

Design comparisons to reduce order and context effects

When participants see several alternatives, sequence, fatigue and direct comparison may shape their responses. Vary presentation order across participants when the design allows, and record sample order and serving method. Use balanced codes in a blind test, and do not reveal identity until the intended evaluation has been captured. Avoid introductions that imply one option is better, and do not describe product attributes before participants give their own responses.

Use neutral, specific prompts: “What caught your attention?”, “What felt clear or unclear?”, or “How would you describe the experience?” Avoid “You liked the taste, didn’t you?” or “Does this pack look more premium?” because they direct attention toward a possible answer. Document what was shown and said, and distinguish participant observations from the team’s interpretation.

Keep liking, intent and purchase distinct

Liking is an evaluation of an experience or concept under defined conditions. Purchase intent is a response about perceived likelihood or willingness in the context of a question; it depends on what the participant imagines about price, availability and alternatives. Actual purchase is observed behaviour under its own conditions. A tasting or hypothetical question does not prove that a purchase will occur later or at a larger scale.

If the decision depends on buying behaviour, consider a suitable, authorised behavioural test, such as a limited offer experiment, and define what will be observed and how conditions will be compared. Do not change purchase terms or customer experience without appropriate approval. Even a behavioural signal from a controlled test remains specific to its audience, offer and context; it does not guarantee a commercial outcome.

A decision tool: which question calls for which test?

Team decision	Possible evidence	What to control	What it does not establish on its own
Do people understand the concept?	Concept exploration and comprehension or relevance questions	Balanced wording, appropriate audience and realistic description	Demand or actual purchase
Which sensory experience needs refinement?	Blind sensory test or sample comparison suited to the question	Preparation, codes, sample order, serving conditions and pauses	The effect of brand or packaging on market choice
Does brand identity change evaluation?	Blind and branded comparison in a defined design	Order, timing of identity reveal and what participants already know	A single cause of preference or future sales
Does packaging communicate the intended information?	Comprehension and readability evaluation in a suitable display	Display size, visible information, context and neutral prompts	In-market shelf performance if the context was not tested
Does interest translate into choice?	Suitable behavioural test where feasible, or authorised available behaviour evidence	Behaviour definition, fair comparison, permissions and constraints	A result generalisable beyond the test conditions

Use the table to identify the evidence gap before choosing a method. If a decision needs more than one evidence type, state what each stage will answer and how the findings will be connected without collapsing unlike measures into one unexplained verdict.

Educational example: an imaginary product improvement

For educational purposes only, suppose a team is reviewing an imaginary food product and wants to learn whether to refine the experience or redesign its pack. It begins with neutral exploratory conversations about expectations of the concept, then tests coded samples in a balanced order to capture sensory experience without revealing the brand. It later presents packaging directions separately and asks participants what they understood and noticed before asking about preference.

If comments point to a sensory attribute or unclear information, those are improvement hypotheses worth examining—not evidence that a change will raise sales. The team can define a testable adjustment, compare the revised version with the original under documented conditions and consider whether observed behaviour is necessary for its decision. This example assumes no preferred result and reports no sales, percentages or field finding.

Reading the case and recognising limits

The illustrative food-product case describes a design attributed to research partner ETK; its figures and orderings are explicitly synthetic and illustrative, not client results. Read it as an example of separating blind tasting from packaging evaluation, not as evidence of product performance or actual sales.

Product tests remain vulnerable to recruitment, memory, language, sample order, venue conditions, recruitment method and how well participants fit the question. An available sample may not represent all buyers or users, and a response in a research session may differ from behaviour in a purchase situation. Describe who participated, what they saw, how responses were collected and what was outside the study. A relationship between two ratings does not prove that one caused the other.

A practical summary

Separate concept evaluation from tasting and packaging, then decide whether blind, branded or sequential testing serves the decision. Balance presentation order, use neutral language and distinguish liking from purchase intent and observed behaviour. Treat findings as evidence within their conditions, and define the next test rather than turning stated preference into a sales forecast.

Frequently asked questions

Should I begin with a blind or branded test?

It depends on the question. Blind testing helps isolate the sensory experience from brand identity; branded testing examines the experience with that identity present. If the difference matters, define the comparison, order and timing of brand reveal in advance.

Does purchase intent mean the product will sell?

No. Intent is a response in a research context, not a transaction. If behaviour is central to the decision, use suitable authorised behavioural evidence and describe its conditions and limits.

Can one study test everything?

A study can include several stages, but each should have its own question, method and limits. If concept, taste and packaging cannot be measured in the same way, do not combine them into one unexplained score.