Measure how people remember your brand, follow the path from awareness to consideration and usage, and compare the image you own against the brands customers see as alternatives.
Skari connects awareness, conversion, image, and competition in one brand view.
FROM SETUP TO FIELDWORK
No separate preview step — mark your brand and the whole question set builds around it.
Brand Tracking
BrandCustom tasks · min. 100 respondentsBrands (mark yours)
Image attributes
Which brands come to mind first? (up to 3) *
Left: the brand and attribute list every tracking study starts from. Right: exactly what respondents see — built from the brands you named.
FROM RESPONSES TO RESULTS
Every figure below comes from one real example study — 30 respondents across 3 brands, run through the same engine your own data runs on.
Awareness alone is one number. Skari pairs it with Brand Health, Brand NPS, and the strongest perceived attribute in a single summary.
Top-of-mind, unaided, aided recall, and 12-month usage — ranked side by side against every competitor.
The brand funnel tracks conversion stage by stage; the Equity Index rolls awareness, usage, and image into one comparable number per brand.
Funnel · Our Brand
90% awareness-to-consideration conversion.
Equity Index
40% aided + 30% usage + 30% image.
A perceptual map and pairwise brand distances show which brands occupy the most similar positions.
Six personality dimensions and a composite Brand Health score summarize both awareness and customer perception.
Brand Personality
Perception Health
Perceived market leader and co-recall pairs state the facts; the conclusions are generated directly from the analysis results above them.
Perceived Market Leader
Co-recalled pairs
A COMPLETE REPORT
The six cards above are fragments of this. Scroll the real output end to end, right here on the page.
IN THE REPORT
Measure recall, follow the funnel, compare competitors, and understand the image customers associate with your brand.
Three awareness tiers
Top-of-mind, unaided, and aided recall for every brand.
Brand funnel
Awareness → trial/usage → consideration conversion.
Brand Equity Index
A single 0–100 composite score per brand.
Perceptual positioning map
Correspondence analysis of brands and image attributes.
Brand distance diagnosis
Exact Euclidean distance between every brand pair's image profile.
Brand personality
Six bipolar dimensions, from Traditional/Modern to Risky/Trustworthy.
Perception health & Brand NPS
Trust, differentiation, and values-fit rolled into one health score.
Competitive landscape
Perceived market leader and co-recall pairs reveal who you really compete with.
Turn awareness, funnel, and perception data into a clearer brand strategy.
30 responses · 3 brands · 20 aware of Our Brand.
Executive Summary
30 respondents · 20 aware of Our Brand
30 respondents were surveyed on 3 brands; 20 who recognized Our Brand (aided awareness 66.7%, rank 3 of 3) were then routed via skip logic into a deeper personality, trust and NPS block. On the perceptual map, Our Brand sits closest to Competitor A and furthest from Competitor B — its real competitive set for image. Its strongest perceived attribute is "Differentiation" (80/100), while "Values Fit" trails at 71.3/100. The word most associated with Our Brand is "Professional" (50% of aware respondents). Brand NPS stands at +25 (n=20).
| Metric | Value |
|---|---|
| Respondents | 30 |
| Design | Skip-logic gated (all → awareness, aware-only → perception) |
| Period | 2026.06.14 – 2026.08.09 |
| Brands compared | Our Brand · Competitor A · Competitor B |
| Image attributes | Innovative · Trustworthy · Premium · Affordable |
| Your brand | Our Brand |
| Aware of your brand | 20 (66.7%) |
The share of respondents who named each brand first when asked, with nothing shown to them, which brands came to mind.
Interpretation
20% of respondents named Our Brand first, which is rank 3 of 3.
0% of respondents named no brand at all. When that share is large, read awareness of the category itself before the ranking between brands. Tie the message to the situations in which you want the brand to come to mind, repeat it, and check how this figure moves in the next survey.
The share who brought the brand up themselves alongside the share who recognised it from a list. Brands that were not on the list have no second figure.
Interpretation
The widest gap between the two figures is Competitor B — recognised by 93.3% from the list but brought up by only 66.7% unprompted (26.6%p apart). A brand people recognise but do not recall is not a reach problem; it is a problem of which situation brings it to mind.
The results for every brand compared, gathered into one table. Items that were not surveyed are marked “no data”.
| Brand | TOM | Unaided | Aided | Estimate range | Usage |
|---|---|---|---|---|---|
| Competitor A | 43.3% | 80% | 96.7% | 83.3–99.4% | 60% |
| Competitor B | 36.7% | 66.7% | 93.3% | 78.7–98.2% | 63.3% |
| Our Brand (you) | 20% | 50% | 66.7% | 48.8–80.8% | 26.7% |
• Competitor A / Competitor B, Competitor B / Our Brand — these estimate ranges overlap, so this sample cannot settle their order.• 0% named no brand at all (range 0–11.4). When that share is large, read category awareness before the ranking between brands.
Interpretation
Our Brand's aided awareness is 66.7% (rank 3 of 3), vs. Competitor A's 96.7%.
Our Brand's response share at each stage. Compare the stages that were surveyed to see whether awareness or usage is the part to look at first. Stages that were not surveyed are marked “no data”.
48.8–80.8%
14.2–44.4% · 40% of previous asked stage
42.3–75.4% · 224.7% of previous asked stage
16.7–47.9% · 50% of previous asked stage
90% awareness-to-consideration conversion.
The arrow figure is the share relative to the previous stage that was surveyed. A * marks that an unsurveyed stage sits between the two. Read the stage figures together with the questions that were asked to decide what to check next.
A composite index across all brands (40% aided awareness + 30% usage + 30% image).
Where respondents who are aware of Our Brand encountered it.
Only people who know the brand rated it. The darker the colour, the higher the score on that item.
| Brand | Innovative | Trustworthy | Premium | Affordable |
|---|---|---|---|---|
| Our Brand (you)30 rated | 71.7 | 74.2 | 66.7 | 78.3 |
| Competitor A30 rated | 84.2 | 78.3 | 84.2 | 81.7 |
| Competitor B30 rated | 83.3 | 83.3 | 84.2 | 85 |
Interpretation
The item on which the brands differ most is Premium (17.5 points between the highest and the lowest). Look at where the compared brands are strong on the items with wide differences, and decide which image your brand will emphasise.
The difference from the average of the compared brands. The further to the right, the higher the item was rated relative to that average.
Interpretation
The bars are the difference from the average of the compared brands. Affordable scored 78.3, -5 against that average; Premium scored 66.7, -17.5. Reflect the strengths in the message and the experience, and for the low-rated items find the concrete inconvenience behind them before improving.
Euclidean distance between brands' attribute-rating profiles — 0 = identical image, larger = more differentiated.
| Brand | Our Brand | Competitor A | Competitor B |
|---|---|---|---|
| Our Brand | – | 0.89 | 0.96 |
| Competitor A | 0.89 | – | 0.24 |
| Competitor B | 0.96 | 0.24 | – |
Interpretation
Our Brand is closest to Competitor A (distance 0.89) and furthest from Competitor B (distance 0.96).
The image ratings of 3 brands arranged on two axes. Brands that sit close together were rated in similar ways. The horizontal axis separates the characteristics of Affordable · Trustworthy from those of Innovative · Premium. The two axes carry 100% of the information in total.
Our Brand (you)
Competitor A
Competitor B
Explains 100% of variance
Right is not “good” and left is not “bad”. The direction of each axis is set arbitrarily by the calculation — flipping an axis whole gives the same data, as the second map below shows. Read a position by which attributes pull that side of the axis, in the table that follows.
Axis interpretation — correlation between each attribute's brand-level ratings and the two map dimensions, with how much of that attribute this two-dimensional picture carries. Small-n (one point per brand): read as directional.
| Attribute | Dim 1 | Dim 2 | Carried by this map | How to read |
|---|---|---|---|---|
| Innovative | -0.97 | -0.27 | 100% | Pulls the left of the horizontal axis |
| Premium | -0.95 | -0.33 | 100% | Pulls the left of the horizontal axis |
| Trustworthy | -0.63 | -0.79 | 100% | Pulls the bottom of the vertical axis |
| Affordable | -0.67 | -0.75 | 100% | Pulls the bottom of the vertical axis |
Shows the same rating results with the direction of the horizontal axis turned around. Look at the positions by the image items tied to each axis. Nothing else changed — which side counts as “right” is set arbitrarily by the calculation, so “further right, therefore ahead” is not a reading this map supports.
Respondents were resampled 300 times and the map re-solved each time (each solution aligned to this one, since axis direction is arbitrary). The map is 0.06 wide.
How far each point moved
95th percentile of displacement, in map units.
Distance between brands
| Pair | Distance | Range |
|---|---|---|
| Competitor A ↔ Competitor B | 0.02 | 0.01–0.07 |
| Our Brand ↔ Competitor B | 0.04 | 0.02–0.09 |
| Our Brand ↔ Competitor A | 0.06 | 0.03–0.11 |
The point that moves most is Our Brand (0.03, against a map 0.06 wide). No pair’s range reaches down to the noise level, so the relative positions hold up.
Words respondents most associate with Our Brand, among those aware of it.
Positive 95.2% · Negative 4.8%
Bipolar personality-adjective ratings — mean position between each pole, among respondents aware of your brand.
Trust, differentiation and values-fit scores, plus a composite brand health index — all normalized 0–100.
Brand Health
68.4
/ 100
Brand NPS
+25
-100 ~ +100
Strongest
Differentiation
80/100
Weakest
Values Fit
71.3/100
"How likely are you to recommend Our Brand?" — asked only of respondents already aware of it, via skip logic.
Net Promoter Score
+25
Based on 20 responses
Promoters (9-10)
9
Passives (7-8)
7
Detractors (0-6)
4
Which brand respondents name as the category leader, and which brands are most often recalled together.
The groups to review first, the direction for promotion, and the figures to check next.
Our Brand's aided awareness is 66.7% (rank 3 of 3), vs. Competitor A's 96.7% — the category leader.
Awareness-to-consideration conversion is 90% — once people know Our Brand, a reasonable share go on to consider it. The bottleneck is reach, not appeal.
Social media is the dominant discovery channel (70% of aware respondents) — concentrate media spend there before diversifying.
On the perceptual map, Our Brand sits closest to Competitor A (distance 0.89) and furthest from Competitor B (distance 0.96) — that's the real competitive set for image, not necessarily for awareness.
Among respondents aware of Our Brand, "Differentiation" is the strongest perceived attribute (80/100) and "Values Fit" the weakest (71.3/100) — a concrete target for messaging.
Brand NPS among aware respondents is +25 (n=20) — net positive, a base to build advocacy on.
3 brands × 4 image attributes · 30 responses (20 aware of Our Brand). Perceptual map: correspondence analysis on brand×attribute mean ratings. MDS: classical scaling of the same distance matrix. Distance = Euclidean on attribute-rating profiles. Equity = 0.4·aided + 0.3·usage + 0.3·image. Brand Health = average of attribute scores, semantic differential percentages and normalised NPS, over aware respondents only. Axis correlations computed across brand-level means (n = number of brands) — directional, not inferential. No cross-wave trend or "ideal point" section is shown here — both need data this template/wave doesn't have (multiple real waves via Survey Series, and a dedicated ideal-brand rating question, respectively).