Brand tracking metrics a board can read: awareness, consideration, mental availability and share of search
Nine brand tracking metrics in two groups, from awareness to share of search, with survey wording, worked numbers and a one-slide board template.
The brand tracking metrics a board can read are nine measures in two groups. Six form the core tracker: prompted awareness, unprompted awareness, consideration, preference, mental availability and share of search. Three are outcomes: brand perception, loyalty and usage. Run survey measures quarterly and share of search monthly. Share of search is the one leading signal a chief financial officer (CFO) can check against free public data.
Which brand tracking metrics hold up in a board meeting
Report nine measures in two groups. Six form the core tracker: prompted awareness, unprompted awareness, consideration, preference, mental availability and share of search. Three are outcomes: brand perception, loyalty and usage.
Our practice is to run the survey measures quarterly and share of search monthly. Tie budget to share of search. In our reading, it is the one leading signal a chief financial officer (CFO) can check against free public data.
| Measure | Question or formula | Data source | Board question it answers | Leading or lagging (our reading) |
|---|---|---|---|---|
| Prompted awareness | Recognition from a list of category brands | Tracking survey | Do buyers know us when they see us | Leading |
| Unprompted awareness | Brands named with no list shown | Tracking survey | Do we come to mind first | Leading |
| Consideration | Brands a buyer would consider next time | Tracking survey | Are we on the shortlist | Leading |
| Preference | Brand a buyer prefers over rivals | Tracking survey | Do we win the shortlist | Middle |
| Mental availability | Share of category entry points linked to us | Tracking survey | In how many buying situations do we come to mind | Leading |
| Share of search | Our brand searches divided by all category brand searches | Google Trends | Is demand for us moving before sales do | Leading |
| Brand perception | Agree or disagree statements about the brand | Tracking survey | Do buyers believe what we claim | Middle |
| Loyalty | 0 to 10 likelihood to recommend | Customer survey | Will current buyers stay and refer | Lagging |
| Usage or penetration | Category buyers who buy us, divided by all category buyers | Survey or sales data | How many buyers do we actually have | Lagging |
Perception. Qualtrics suggests statements on a Likert scale (Strongly Agree to Strongly Disagree), from product claims to values. Report the three or four attributes the strategy depends on.
Qualtrics, Brand Metrics: What to Measure in Your Brand Tracking Study, 6 November 2020
Loyalty. Net Promoter Score (NPS) asks: "On a scale of 0-10, how likely are you to recommend [brand] to your family and friends?" Qualtrics says it is best used in combination with more nuanced measures, so pair it with repeat purchase rate.
Qualtrics, Brand Metrics: What to Measure in Your Brand Tracking Study, 6 November 2020
Usage. The survey wording Qualtrics gives is "Please select which of the following brands you buy or use regularly." Where sales data exists, use it in place of the survey answer.
Qualtrics, Brand Metrics: What to Measure in Your Brand Tracking Study, 6 November 2020
Label the groups on the slide. Leading signals tell the board where sales are heading. Lagging outcomes confirm what already happened. A board that sees only lagging numbers can only react to them.
Pairing survey measures with behavioral ones matters. Les Binet said he found "the relationship between tracking metrics and actual purchase behaviour is often surprisingly weak". He also proposed reading the base level of Share of Search without advertising as a measure of brand strength.
The stakes are commercial. Gartner reported that companies with a strong brand strategy are 2x more likely to exceed their growth goals. Julie Reeves of Gartner said "CMOs need to move beyond tracking brand metrics in isolation". The board deck template shows where each group sits on the slide.
Gartner, Gartner Marketing Survey Finds 84% of Companies Are Stuck in a “Brand Doom Loop”, 10 June 2026
How to measure brand awareness, consideration and preference
Each of the four funnel metrics is a survey question. The wording and the order of the questions decide what the number means, so fix both before the first wave and keep them unchanged after that. The templates below are ours. Adapt the category phrase, then lock it.
Sample wording, in the order to ask it
- Unprompted awareness. When you think of [category], which brands come to mind? Use an open text box with no list. Count your brand if it appears anywhere in the answer, and record separately whether it came first.
- Prompted awareness. Which of these brands have you heard of, even if only by name? Show your brand and competitors in rotated order. A Qualtrics guide also suggests a follow-up on source: "In the past year, where have you seen or heard of our brand?"
- Consideration. Next time you buy [category], which of these brands would you consider? Show only the brands the respondent recognized.
Qualtrics, Brand Metrics: What to Measure in Your Brand Tracking Study, 6 November 2020
Ask unprompted awareness before any brand list appears. A respondent who has just seen your logo cannot give a clean unprompted answer.
Worked example: reading the funnel as conversion
This example uses round inputs and is not a benchmark. Suppose a wave has 1,200 respondents and 800 recognize your brand when prompted. That is a prompted awareness of 67%. If 600 of those 800 would consider you, then 75% of aware buyers consider you. If 400 of those 600 prefer you, then 67% of considerers prefer you.
The board learns more from the conversion between stages than from the headline awareness figure. We think a drop between consideration and preference points to product, price or proposition. A drop at awareness points to reach.
Comparison table
| Metric | How it is collected | How often (our reading) | What moves it (our reading) | How it misleads (our reading) |
|---|---|---|---|---|
| Unprompted awareness | Open text, before any list | Continuous fieldwork, reported quarterly | Sustained reach and distinctive assets over many months | Coding of misspelled or partial brand names can shift the score between waves |
| Prompted awareness | Rotated brand list | Continuous fieldwork, reported quarterly | Any exposure, including weak or negative | Close to its ceiling for established brands, so it barely moves |
| Consideration | List limited to recognized brands | Continuous fieldwork, reported quarterly | Relevance, availability, price position | Rises with awareness, so it can look like progress when only reach changed |
| Preference | Single choice from a list | Continuous fieldwork, reported quarterly | Product experience, proposition, competitor moves | Small subsamples make wave-to-wave swings look meaningful when they are noise |
Record these four lines on one page of your marketing scorecard, with the base size shown under every figure. For definitions of terms, see the brand tracking glossary entry.
What mental availability is and how category entry points measure it
Awareness tells the board whether buyers know the brand. Mental availability asks a narrower question: does the brand come to mind at the moments when buyers start to think about buying in the category. The Ehrenberg-Bass Institute builds that measure on category entry points (CEPs).
CEPs are essentially moments when someone mentally enters a product or service category, such as a hot day at the beach or a conversation about saving money on energy. These moments act as cues that trigger your memory to recall brands. A brand linked to more of the relevant cues has more chances to be recalled before a purchase.
marketingscience.info, Category Entry Points Dissected: How They Really Contribute to Growth |, 28 November 2025
The two-stage method
The Ehrenberg-Bass approach runs in two stages.
- Elicit. The first stage is an Ehrenberg-Bass designed elicitation survey that asks category buyers a series of tailored questions. The aim is to collect the situations, needs and occasions buyers actually report, in their own words.
- Quantify and score. The second stage is quantifying the CEPs to understand the usefulness/relevance of each, then measuring your brand’s performance as well as the performance of key competitors. This also highlights CEPs for the category that are more competitive (i.e. where you might need a specific strategy).
marketingscience.info, Identifying and Prioritising Category Entry Points |, 14 September 2026
A worked example
These are round example inputs, not benchmarks. A tracker surveys 1200 category buyers and tests two CEPs.
| CEP | Buyers linking any brand | Buyers linking our brand | What it suggests (our reading) |
|---|---|---|---|
| CEP A | 800 | 400 | Large cue where our brand holds half the links |
| CEP B | 600 | 400 | Smaller cue where our brand is strong |
At CEP A, our brand holds 400 of 800 links, or 50%. At CEP B, it holds 400 of 600, or 67%. In our reading, CEP A is the growth question for the board: it is the larger cue, and competitors hold the other half of the links. CEP B is a position to defend in creative work.
The mistake to avoid
The common error is treating the CEP count as the goal. More CEPs don’t automatically mean more growth. The Ehrenberg-Bass finding is that linking a brand to as many of these relevant cues as possible increases mental availability. The important word is relevant. A brand that adds ten fringe occasions to its messaging may score higher on a list and gain nothing in the moments that drive volume.
marketingscience.info, Category Entry Points Dissected: How They Really Contribute to Growth |, 28 November 2025
For the board, we would report a short set of high-relevance CEPs from stage two, the brand's share of links at each, and the gap to the main competitor. Track the same set each wave inside your brand tracking program so the movement is comparable.
What share of search is and how it is calculated
The ratio was covered earlier: one brand's searches over all brand searches in the category. Two build details matter. It counts organic Google search queries, "not share of paid search advertising", and Google Trends can offer weekly frequency.
A worked example
Take four brands in one category with these monthly search volumes. They are example inputs, not a benchmark. Add them: 1200 plus 800 plus 600 plus 400 gives a category total of 3000.
| Brand | Example searches | Share of search |
|---|---|---|
| A | 1200 | 40.0% |
| C | 600 | 20.0% |
| D | 400 | 13.3% |
Brand A's 1200 searches out of a category total of 3000 give a 40.0% share. Brand D's 400 searches out of the same 3000 give 13.3%. The useful number for a board is the gap between each share and that brand's market share.
What Binet found
The gap is named. Binet calls it Extra Share of Search (ESOS) and describes it as a particularly solid indicator of market share movements. We think a brand whose search share sits above its market share tends to gain. One that sits below tends to lose.
He also linked the metric to spend. Sustained advertising grows share of search, with 60% of searches coming from the long-term effects and 40% from short-term effects. When share of voice falls below a brand's equilibrium level, the Share of Search tends to fall over the next two years.
Where it misleads
Binet says it is not a perfect predictor. Conversion is affected by other factors, "particularly price." A brand can win searches and lose sales to a cheaper rival. And not all search activity is positive: a recall, an outage or a scandal lifts search volume while hurting the brand.
In our reading, that gives three rules for a board pack:
- Report ESOS next to share of search, not the raw share alone.
- Annotate any month where price changes or bad news could explain a spike.
- Read the trend over quarters, given the lags above.
For definitions, see share of search and brand tracking.
How Google Trends normalizes its index and why that matters for share of search
Share of search rests on an index, not a count. The board pack should explain how that index is built before anyone reads a trend line from it.
Google describes three steps. First, each data point is divided by the total searches of the geography and time range it represents. Second, the resulting numbers are then scaled on a range of 0 to 100, with 100 marking the peak.
support.google.com, FAQ about Google Trends data - Trends Help
Why one query matters
Every query is scaled to its own peak. Two brands pulled in separate queries each get their own 100, so their numbers cannot be added or compared.
Here is an example, not a benchmark. Brand A's weekly index points sum to 1200 and brand B's to 800 across a year, a combined 2000. That arithmetic holds only if both brands sat in the same query, on the same scale.
The working rules
| Rule | What Google says | How we apply it (our practice) |
|---|---|---|
| Five terms per pull | You can explore up to five topics or terms | Fit the category's main brands into one query. If there are more, repeat one anchor brand in every query and rescale against it |
| Topics over search terms | Topics are generally considered to be more reliable, and they include misspellings and acronyms | Use the brand topic where one exists |
| Like for like | topics should be compared with other topics, and search terms should be compared with other search terms | Never mix a topic for one brand with a search term for another |
| Small brands | search terms with low volume appear as "0." | Report a challenger with frequent zeros as a range, not a share |
| Geography | Different regions that show the same search interest for a term don't always have the same total search volumes | Compare share within a country, not index levels across countries |
Expect revisions
Because the data is sampled and noised, we think a repeat pull of the same query will rarely match the last one exactly. Our practice is to save each pull with its date, settings and raw export. Treat a move of a point or two as noise and act on sustained shifts across several quarters.
The share of search glossary entry carries the full definition.
A worked share-of-search example for one category
Here is the method run end to end for one category. The four brands and every number below are an example we built to show the arithmetic. They are not a benchmark.
Step 1: Pull all four brands in one query
Enter the four brands as topics in a single Google Trends comparison, for the same country and the same 12 months. One query keeps all four on the same index, for the reasons covered in the previous section. Export the weekly values and sum each brand's column over the year.
Step 2: Sum the category and compute shares
In our example, the summed index values come to 1,200 for Brand A, 800 for Brand B, 600 for Brand C and 400 for Brand D.
Each brand's share of search is its sum divided by that total.
Step 3: Set share of search against market share
Take market share from your own sales data or a syndicated panel for the same period.
| Brand | Summed index (example) | Share of search | Unit sales (example) | Market share | Extra share of search (our calculation) |
|---|---|---|---|---|---|
| A | 1,200 | 40.0% | 1,200 | 40.0% | 0.0 points |
| B | 800 | 26.7% | 600 | 20.0% | +6.7 points |
| C | 600 | 20.0% | 800 | 26.7% | −6.7 points |
| D | 400 | 13.3% | 400 | 13.3% | 0.0 points |
Brand B's share of search of 26.7% minus its market share of 20.0% is an extra share of search of 6.7 points. Brand C's share of search of 20.0% minus its market share of 26.7% is a gap of minus 6.7 points.
Step 4: Read the gap as a warning, not a verdict
Brand C sells more than its search interest suggests. That is the pattern the earlier section described as a signal of market share movement. In Binet's work, the lead time could be up to a year for cars, and he described the measure as a possible early warning system for brands on market share.
We would put Brand C's gap on the board page as an amber flag with three checks beside it:
- Is the gap persistent? Rerun the query for the latest quarter. A gap that holds across two or more quarters carries more weight than a single reading.
- Is search negative? Look at related queries for recalls, complaints or price rises.
- Is price doing the work? A brand winning on discount can hold sales while losing interest.
Google itself says Trends data should always be considered as one data point among others. We think that means the gap should open a conversation with the CEO and finance, alongside brand tracking survey results and marketing mix modeling, before anyone moves budget on it.
support.google.com, FAQ about Google Trends data - Trends Help
How often brand tracking should run
Cadence is a method choice, and we set it by how often each data source changes, not by how often the board meets. The rule is to read each metric at the rhythm its data supports and to report it at the rhythm decisions get made.
Share of search: read weekly, report monthly
Share of search data from Google Trends is free and generally goes back to 2004. That makes a weekly read cheap and gives a long baseline to compare against.
Weekly does not mean weekly reaction. Les Binet found that share of voice has a big short-term effect that dies away quickly, plus a smaller, longer-term effect that decays away slowly. In our reading, a weekly spike after a campaign burst is mostly the first effect. The trend line the board should see is the second one, so we smooth weekly data into a rolling average before it leaves the marketing team. Binet also said "the data needs to be interpreted with care".
Survey trackers: monthly or quarterly
Awareness, consideration and preference come from surveys, and each wave costs money. Qualtrics describes the value of continuous monitoring as spotting when a brand is on track and when it wobbles. In our reading, for most companies in the $20M to $2B range, that means a monthly fielding pooled into a quarterly board read, or a single quarterly wave.
Qualtrics, Brand Metrics: What to Measure in Your Brand Tracking Study, 6 November 2020
Pooling helps with sample size. As an example, not a benchmark: a monthly wave of 400 respondents gives 1200 over three months. If you then cut that quarter into three regions of 400, each region holds a third of the pooled sample, and a regional move between quarters will be noisier than the national figure.
Category entry point studies: annually
The two-stage category entry point study described earlier is a heavier piece of work. In our reading, buyer memory structures move slowly, so we run it once a year and use the survey tracker to watch a short list of priority entry points in between.
| Metric | Data rhythm | Board read (our reading) |
|---|---|---|
| Share of search | Weekly, free | Monthly trend, quarterly summary |
| Awareness, consideration, preference | Per survey wave | Quarterly |
| Category entry points | Per study | Annually |
Rules for not over-reading noise
- Agree a minimum change before anyone calls a move. Set it with your research agency from the sample size, and write it into the brand tracking brief.
- Treat a single wave in one direction as a watch item. Act on consecutive waves moving the same way.
- Check a survey move against share of search for the same period. When both move together, we think the signal is stronger.
- Keep the questionnaire wording and sample frame fixed. A change in either resets the trend line.
- Read brand trends alongside marketing mix modeling results, which run on their own cadence.
How to report brand health to the board on one slide
Gartner's survey of brand measurement, published in June 2026, put a name to the problem this slide is meant to solve. It found that 84% of companies were stuck in a "brand doom loop" that keeps marketing leaders from proving brand's effect on growth. The loop, in Gartner's description, forms when companies underinvest in brand measurement, lack confidence in the results and then attract even less funding. The survey covered 426 senior marketing leaders between September and October 2025.
Gartner, Gartner Marketing Survey Finds 84% of Companies Are Stuck in a “Brand Doom Loop”, 10 June 2026
The same release found that 43% of C-suite executives want a clear, simple story about brand health and business performance. One slide is that story.
Gartner, Gartner Marketing Survey Finds 84% of Companies Are Stuck in a “Brand Doom Loop”, 10 June 2026
The template
Copy this into the brand page of your board deck. Fill the brackets each quarter. Keep every definition to one line, so a director can read the panel in under a minute.
| Metric | One-line definition | Current value | Four-quarter trend | Competitor benchmark | What changed |
|---|---|---|---|---|---|
| Unprompted awareness | Share of category buyers who name us without a prompt | [x%] | [Q1, Q2, Q3, Q4] | [top two rivals] | [one line] |
| Prompted awareness | Share who recognize us from a list | [x%] | [Q1 to Q4] | [top two rivals] | [one line] |
| Consideration | Share of aware buyers who would include us on a shortlist | [x%] | [Q1 to Q4] | [top two rivals] | [one line] |
| Preference | Share who would choose us first | [x%] | [Q1 to Q4] | [top two rivals] | [one line] |
| Mental availability | Our share of links to the category entry points we track | [x%] | [Q1 to Q4] | [top two rivals] | [one line] |
| Share of search | Our branded searches over all brands in the category | [x%] | [Q1 to Q4] | [top two rivals] | [one line] |
Write one cause per row, such as a campaign, a price move or a competitor launch. Leave it blank when nothing moved.
Five steps to set it up
- Fix the competitor set: the same two or three rivals on every row, every quarter.
- Lock the survey wording and the search terms before the first wave, so the trend line compares like with like.
- Pull four quarters of history where you have it. Label any row with fewer as new.
- Agree the definitions with finance, using our CEO and CFO alignment guide.
- Put the panel on a fixed date in the board calendar.
The decision you have to make
Pick one row as the brand target tied to budget. The rest stay as context. We think a single target is what moves brand out of the cycle Gartner's Julie Reeves described: "This creates a cycle where brand is undermeasured, underfunded and undervalued." Choose the metric your board will accept as a lead indicator of revenue, then link its target to the brand and performance split in next year's plan.
What this page does not cover
This page does not cover how to choose or brief a research agency, how to set sample sizes and significance thresholds, or how to build a marketing mix model. The worked numbers are examples built to show the arithmetic, not benchmarks for any category. Nothing here is financial or legal advice.
Related on CMO Tribune: Marketing mix modeling, explained for the CMO who has to choose a measurement model; In-house vs agency cost, compared with your own numbers; Marketing team structure by revenue band: functional, pods, hub-and-spoke and business-unit models.
Questions
What are the most important brand tracking metrics?
Report nine measures in two groups. The core tracker covers prompted awareness, unprompted awareness, consideration, preference, mental availability and share of search. The outcomes are brand perception, loyalty and usage. Label leading signals and lagging outcomes on the slide, because a board that sees only lagging numbers can only react to them.
What is share of search and how is it calculated?
Share of search is your brand's searches divided by all brand searches in the category, using organic Google search queries from Google Trends. If four brands have 1,200, 800, 600 and 400 searches, the category total is 3,000 and the first brand's share is 40.0%. Les Binet found it correlates with market share and moves ahead of it. The gap between share of search and market share, which he calls Extra Share of Search (ESOS), is the number to watch.
How do I measure brand awareness?
Ask unprompted awareness first, with an open text box and no brand list: "When you think of [category], which brands come to mind?" Then ask prompted awareness with a rotated list of your brand and competitors. If 800 of 1,200 respondents recognize your brand when prompted, prompted awareness is 67%. Fix the wording and question order before the first wave and keep them unchanged.
What is mental availability in marketing?
Mental availability is whether a brand comes to mind at the moments buyers start to think about buying in the category. The Ehrenberg-Bass Institute measures it through category entry points (CEPs), such as a hot day at the beach. Its method first elicits the situations buyers report, then quantifies each CEP and scores your brand and key competitors. The goal is links to relevant cues, not a higher count of cues.
How often should brand tracking run?
Read share of search weekly, since Google Trends data is free, but smooth it into a rolling average and report it monthly. Field awareness, consideration and preference surveys monthly and pool them into a quarterly board read, or run a single quarterly wave. Run the heavier category entry point study once a year. Act on consecutive waves moving the same way, not on a single wave.
How do I report brand health to the board?
Use one slide with six rows: unprompted awareness, prompted awareness, consideration, preference, mental availability and share of search. Each row carries a one-line definition, the current value, a four-quarter trend, a competitor benchmark and one cause for any change. Gartner found that 43% of C-suite executives want a clear, simple story about brand health and business performance. Pick one row as the brand target tied to budget and keep the rest as context.
Sources
- Brand Metrics: What to Measure in Your Brand Tracking Study, Qualtrics, 6 November 2020
- Binet presents fast, cheap, predictive Share of Search metric to EffWorks, IPA
- Gartner Marketing Survey Finds 84% of Companies Are Stuck in a “Brand Doom Loop”, Gartner, 10 June 2026
- Category Entry Points Dissected: How They Really Contribute to Growth |, marketingscience.info, 28 November 2025
- Identifying and Prioritising Category Entry Points |, marketingscience.info, 14 September 2026
- FAQ about Google Trends data - Trends Help, support.google.com
- Basics of Google Trends - Google News Initiative, newsinitiative.withgoogle.com
Written by the CMO Tribune Desk and edited by Lila Odinaev. Every factual sentence was checked against a passage on the source it cites. See how we work, or report an error on the corrections page.