Brand Measurement: The Proper Process To Explode Company Growth

In a day where everything is measurable, most marketers still don't know how to measure brand the right way.
05/14/2019

Measuring brand is hard. Accurately measuring brand is hard enough that most marketers either don’t attempt to do it all, or if they do, don’t realize they’re doing it wrong.

The reason it’s hard to measure brand is because there isn’t a standardized way to do so accurately. Sometimes we get caught up measuring KPIs that are too focused on ROI and don’t correctly consider the marketing team’s impact. Other times, our brands fail to strike a balance between capturing attention and creativity and end up fraught with biases. Sometimes, historical precedents are discarded in lieu of proxies. And still other times, “brand measurement” can be a foreign concept to myopic digital marketers too focused on bottom-of-funnel metrics.

How-To-Measure-Brand

At the end of the day, marketers who don’t fully understand brand measurement turn to the platforms with which they’re most comfortable or familiar to gain what they think are true and accurate insights into their brand. We call these “proxies” and here’s why it’s a mistake to trust them.

  • Search & website tools: Google Analytics, AdWords, Alexa, et al. can show us search demand for specific topics, services, or sites but the broadness of their application makes them spurious in measuring brand growth.

  • Social media-based services: Buzzsumo, Hootsuite, SproutSocial, et al. can show us the topics people are talking about on social media but the arbitrariness of social media behavior makes them unreliable in determining brand sentiment.

  • DIY consumer insight platforms: Pollfish, SurveyMonkey, Google Surveys, et al. do provide a degree of accuracy in “voice of customer” metrics and can be cost-effective but require a level of marketing expertise to remove biases and achieve statistical significance.

  • Market research and customer experience (CX) companies: Nielsen, Qualtrics, Kantar, YouGov, et al. have long been the standard of market research, customer experience management, and consumer insights but can be cost-prohibitive for all but the largest brands.

Brand-Measurement-Proxies

Proxies can be imprecise, contradictory, expensive, unrefined, and often time-consuming, which is why relying on them to measure brand is problematic. They gather tangentially related data and, too often, marketers seize on it, slice-and-dice it, build insights from it, and present it as fact. That’s not a great way to find the answers you’re seeking or gain true insight into your brand performance.

Brand Metrics to Measure

The first step in measuring your brand correctly is identifying the right metrics that’ll provide the insights you need to show growth and ROI. In a related article, Brand Metrics: Learning to Quantify the Value of Your Brand, we identified six primary metrics marketers can measure to ensure they gain buy-in from leadership and ultimately hit their business goals.

As marketers, we’re conditioned to measure performance-based KPIs like sessions, clicks, impressions, conversion rate, revenue, etc. But the metrics that truly make a difference when measuring brand focus on emotional connections rather than transactional ones.

  • Awareness/Recognition: This is the most important indicator to monitor when growing your brand. People can’t buy from you if they don’t know you exist.

  • Frequency: If consumers hear about you often, it means you're either advertising a lot or you have a ton of word-of-mouth buzz around your brand. Just remember, frequency can undermine your brand if the customer’s experience is poor.

  • Familiarity: This measures consumer perceptions before engaging with your brand. It’s about the gut feelings and intuition consumers have about your product or service.

  • Favorability: This measures a consumer’s direct experience with a brand in terms of quality, superiority, consideration, and personality. This gets to whether your customers like you and if they’d recommend you.

  • Preference: Every brand has competitors. You need to understand if yours want to do business with you or someone else. This metric can be impacted by awareness, familiarity, and favorability.

  • Demand/Purchase Intent: This is about sales performance and market share. Without demand, you have no customers, which means no brand.

Brand-Measurement-Metrics

Building Your Brand Survey

None of the metrics or tools matter unless you can figure out a way to find your consumers and talk to them about you. That’s what measuring brand is about. Whether you end up hiring a market research firm or choose to do it yourself, calculating robust brand metrics requires a few things in addition to asking real people: 1) a surveying tool; 2) a compensation system; 3) a survey/study methodology; 4) ability to analyze results; and 5) a dashboard to communicate results.

Here’s a step-by-step process that defines how to build an effective survey. Following this process can put you on the path toward understanding your brand and its strengths and weaknesses.

Brand-Measurement-Process

Step 1: Plan

Survey Platform Selection: Survey platforms vary in their accuracy, pricing plans, and analysis capabilities. Google Surveys, SurveyMonkey Audience, and Pollfishrepresent good options for search marketers. Pro tip: These platforms can be very helpful but they require some marketing and statistical expertise to ensure valid results.

While each of these do-it-yourself survey platforms have some major differences, most of them follow the same basic steps.

First, the targeting. The goal here is to find an audience that’s representative of the entire potential market for a brand. The more mainstream the product, the broader your targeting needs to be. The more niche a brand is (i.e., B2B brands only appeal to a small segment of the entire population), the more sophisticated your screening and targeting tactics have to be.

Pollfish-Brand-Survey

Next, it’s time to build out what the survey looks like to respondents. Again, every platform is different when it comes to question and response formats, question grouping, ordering logic, and other configuration options that can help increase accuracy and reduce the bias marketers often bring to their studies.

Pollfish-Brand-Study

Most brand studies should contain a healthy mix of both core brand questions needed to support the beacon metrics above, and custom questions that will guide the most important project within an organization during the upcoming fiscal period (i.e., next quarter).

In the end, most brand measurement studies should consist of between 15-20 questions ranging between the core brand metric questions and custom questions dedicated to understanding the day’s current business problems.

Step 2: Calculate

Panel & Study Creation: Parameters are defined for respondent panel composition, desired levels of statistical confidence/margin of error, and brand measurement questions. Pro tip: Speak to someone familiar with statistics to ensure your questions don’t include biases, assumptions, or inconsistencies.

Step 3: Launch

Research Commences: During the survey collection period, respondents are recruited via apps (e.g., Google Opinion Rewards), survey-gated content, or programmatic advertising. Pro tip: Let your survey finish before you look at any of the data. Try to avoid biasing your own perceptions by looking at results and building hypotheses too soon.

Studies range widely in the amount of time they take to complete. Results can come in overnight if you launch a small study to general population in a heavily populated area. But, studies can take much longer to complete—sometimes a few weeks or even a month or two in extreme cases—if they contain advanced targeting, screening, or quota logic.

Step 4: Analyze

Brand Metrics & Analysis: Once the study is completed, a thorough analysis of the data occurs via human and (increasingly) machine efforts. Pro tip: Filter your first round of data for any answers/trends/responses that are nonsensical or counterintuitive. This may indicate survey flaws that could require a new survey.

This is the part that can make or break your study. If you simply report back the answers to every question, your study could end up being dismissed as something novel, but irrelevant or worse, unactionable.

If you’ve followed the previous steps correctly, the analysis phase is where the magic happens. There are almost always only a small handful of insights in any product category or business vertical that truly matter. The goal of this analysis is to find those 2-5 truths upon which the entire brand must be built in order to become the market leader. Plain and simple.

Brand-Measurement-Insights

And when those truths become clear, curate only as much as what your stakeholders need to see in order to act. Forget the data; focus on the insights.

Step 5: Impact

Results & Recommendations: Brand metrics should be refreshed quarterly, biannually, or annually—depending on the brand’s growth rate—and should accompany the goal-setting processes. Pro tip: Human emotions and perceptions take time to change. Try to avoid projecting aggressive changes to brand metrics too quickly.

Done correctly, this process becomes a tool to guide the company through uncertain market waters and gives its leaders more data-driven ways to build a purpose-driven brand.

Refreshing brand metrics consistently also provides marketers with a way to take on a more consultative role to the rest of the business, which is increasingly necessary whether you work at a venture-backed startup or a Fortune 500 struggling to transform digitally.

Finally, it puts us in a position to change the business on a foundational level if need be, which is sometimes necessary in order to live up to the brand promise needed to become the number one brand in a category.

Calculating Sample Size & Margin of Error

As soon as you read “calculating” in the subhead, there’s a chance your eyes glazed over. There’s a chance we’ll lose some readers here but there’s no reason to be afraid of the math. We’ll walk you through it and give you a free tool that’ll make things easy.

Sample size dictates the amount of information you must gather to be confident (i.e., achieve statistical validity and reliability) in your data. That said, it’s extremely important to get your sample size right before building out your survey.

Sample size is essentially an estimate and, as such, “The more variable the population, the greater the uncertainty in our estimate,” according to UK-based Select Statistical Services. “Similarly, the larger the sample size, the more information we have and so our uncertainty reduces.”

That means you need a sample size that’s large enough to be representative of the population you’re studying. But you also need to consider margin of error, or, the level of precision you require. Select Statistical Services defines it as, “the range in which the value that you are trying to measure is estimated to be and is often expressed in percentage points (e.g., ±2%).”

Put simply, a narrow margin of error means more reliable data but it requires a larger sample size to achieve.

There are several platforms out there that will calculate sample size, confidence level, and margin of error for you but the SurveyMonkey calculator is simple and effective. Here’s an example of how it works.

Brand-Measurement-Margin-Of-Error

Say we’re trying to poll U.S. women who purchase jewelry over the internet every year. And, we want to get the most reliable data we can, which means a confidence level of at least 95 percent. Here’s how we’d break that down in SurveyMonkey’s calculator.

We know there are approximately 158.3 million women in the U.S. Our additional research shows 88 percent of them use the internet. Multiply those values together and you’ll get 139.3 female internet users. Multiply that value by the number of women who have purchased jewelry over the internet in the last year (23.5 percent) and you get 32.7 million. That’s your population size.

Then, it’s just a matter of choosing your confidence level and margin of error. SurveyMonkey allows you to choose a confidence level that ranges between 80-99 and any margin of error you’d like—just remember, a narrow margin of error is better, but requires a larger sample size. Once you’ve done that, SurveyMonkey spits out the size you need to sample. It’s that simple.

The Bottom Line

To measure your brand effectively, you need to know what metrics to measure, the proxies to avoid, how to build out the right surveying mechanism, and where to find real people to provide the insights you’re seeking.

As we said at the beginning, measuring brand is hard. It’s a process—one that’s often done incorrectly—and it’s only as strong as its weakest component. Gathering accurate insights from a true walkin’ talkin’ audience is difficult, but investing time and resources in the right process will help you build a robust brand and, eventually, achieve your business goals.

What’s more, understanding how to measure brand effectively can drive success in other areas of the business. A strong brand gives your organization an identity. Accurate brand measurement can set the tone for sometimes difficult discussions about decline, development, and growth. And, if measured correctly, brand metrics can become the KPIs you benchmark against to set organizational priorities and, ultimately, build your marketing strategy.

Author: Josh Braaten

CEO - Brandish Insights

Get Free Brand Measurement Resources

Subscribe to our newsletter to be the first to get these freebies.