How KwantumLabs combines CEP mental availability and brand association density to deliver actionable growth recommendations.
Direction: CEP → Brand. Given a buying situation, which brands come to mind? Brand A is linked to 4 CEPs (high mental availability). Brand C is linked to only 1 (low). Measured via aided assessment with a brand list to avoid popularity bias.
Direction: Brand → Associations. Given a brand name, what comes to mind? Density = distinct associations x inter-connectedness. Dashed lines show associations that reinforce each other ("reliable" connects to "established" connects to "big companies"). Measured via open free association, no pre-defined list.
Both models describe parts of the same memory network. The CEP model measures how many doors lead into the brand (entry points from buying situations). The density model measures how reinforcing the internal structure is once you're in (how strongly associations connect to each other and back to the brand).
A brand can have high mental availability (linked to many CEPs) but thin association density (people know the name but can't say much about it). Or a brand can have rich, dense associations but only be linked to one or two buying situations. Neither alone gives you the full picture. Both together tell you exactly where marketing effort should go.
Business event activates a Category Entry Point
CEP ModelCEP-brand links determine which brands are retrieved
CEP ModelDenser associations make the brand "stick" and feel more substantial
Density ModelBrands with both retrieval and rich associations get evaluated
Both ModelsBrand is selected (or not) based on fit with needs
Both Models| Dimension | CEP Mental Availability | Brand Association Density |
|---|---|---|
| Cross-sectional correlation with market share | 0.74 (soft drinks, 10 associations). Consistent across 40+ categories. | 0.64 to 0.95 depending on category. Highest for smartphones (0.93) and toothpaste (0.95). Lower for fragmented markets like beer (0.64). |
| Proven to predict FUTURE sales growth? | Not directly proven. Cross-sectional data shows bigger brands have more mental availability. Logical inference, not causal proof. | Not directly proven. Same limitation. Correlation with current market share, not demonstrated prediction of future share changes. |
| Breadth of evidence | Very broad. Dozens of categories, B2B and B2C, multiple countries, studies spanning decades. | Three consumer categories (smartphones, beer, toothpaste). No B2B categories tested yet. |
| Actionability for marketing | High. Tells you exactly which CEPs to target and where competitors are stronger. Direct input for media planning and messaging strategy. | Moderate to high. Tells you whether your brand's associations are thin or dense, and what language people use. Useful for creative and messaging, less direct for media planning. |
| Diagnostic power: Samsung vs. Apple example | Would show both have high mental availability across CEPs, but limited insight into why Apple's market share is higher despite Samsung having more total associations. | Samsung had MORE distinct associations but LOWER density (less inter-connected). Apple had fewer but more tightly inter-connected associations. Density explains the gap. |
| Best data collection method | Survey with aided brand list (per Romaniuk). Avoids popularity bias. Scales well to large samples. | Open free association question. Pre-defined attributes fail in most categories. Interviews are ideal. |
B2B studies: Category Entry Points in B2B are typically business events (growth, turnover, contract expiry, compliance changes) that can be identified collaboratively with the client's sales team before fieldwork. Stage 1 serves as a brief warm-up and validation that the pre-built CEP list is complete. If a participant mentions a buying trigger not on the list, that's a valuable discovery.
B2C studies: Consumer categories often have many diverse CEPs across occasions, need states, and life situations. Stage 1 should NOT be included in the measurement interview. Instead, conduct a separate CEP discovery study (~60 interviews) beforehand to generate the CEP list. The measurement interview starts directly at Stage 2.
"Think back to the last time your company needed a recruiting solution. What was happening that triggered the search?"
"I'll read a list of companies. For each, tell me yes or no: would you associate them with this situation?"
"When evaluating a solution for high-volume hiring, what matters most to you?"
"How would you describe [Brand X] to a colleague? What comes to mind when you think of them?"
"Are you familiar with [Brand Y]? Any reason it didn't come up earlier?"
Which buying situations is the client's brand linked to vs. competitors? Where are the gaps and white space?
Action: "Focus marketing on [CEP X] and [CEP Y] where you're under-linked relative to your category penetration."
Mental penetration, mental market share, and network size for the client brand vs. competitors.
Action: "Your mental penetration is 34% vs. competitor's 61%. The growth ceiling is awareness, not perception."
How many distinct associations does the brand have, and how inter-connected are they? Is the brand's mental structure thin or dense?
Action: "Your brand has many associations but low density. Messaging should reinforce connections between existing associations rather than adding new ones."
The actual words and phrases buyers use to describe brands and what matters to them. Not marketing language. Buyer language.
Action: "Buyers say 'takes the hassle out of compliance hiring,' not 'streamlines regulatory workforce acquisition.'"
What portion of the target audience is unaware vs. aware-but-not-linked vs. actively rejecting? Typically unawareness is 4-16x higher than rejection.
Action: "72% of HR leaders don't know your brand exists. Fix reach before fixing messaging."
How each competitor is perceived in the buyer's own words, overlaid with which CEPs they own. Reveals both positioning reality and positioning opportunity.
Action: "Competitor X owns 'fast hiring' but is perceived as 'impersonal.' You can own 'fast + human' if you link to that CEP."
Both models show that brands with more memory connections (whether CEP links or association density) correlate with higher market share. Growth comes from building more connections, not defending a narrow position.
Romaniuk found unique associations don't predict brand performance. Vriens found adding uniqueness to density metrics didn't improve correlations. Differentiation is less important than presence.
Romaniuk advocates aided brand lists for CEP measurement (to avoid popularity bias), but Vriens shows free association beats pre-defined attributes for capturing what people actually think. Both favor methods closer to natural memory processes.
Surveys can run the CEP model at scale, but they cannot capture brand association density or inter-connectedness. They produce checklists, not networks. And pre-defined attribute lists fail outright in many categories.
Interviews can run both models in a single 30-minute conversation. The CEP measurement (with aided brand list) gives you the same quantitative metrics a survey would produce. The free association portions give you the density metrics and qualitative richness that surveys structurally cannot deliver. With 100 interviews, you have enough data to compute all metrics from both models, plus the natural language and emotional texture that turns data into actionable creative and messaging recommendations.
No other methodology delivers both in one study. That is the offering.