Tracking how people perceive a brand can be very expensive and run into the hundreds of thousands of USD per year and market! Data collection typically accounts for at least half of the total cost of a brand tracker, so reducing data collection costs is the key to making brand tracking more affordable.
To understand how we can reduce data collection costs, let’s first look at what makes collecting brand tracking data so expensive.
The length and complexity of a survey. There are lots of potential metrics to track, from the classic funnel metrics (awareness, consideration, preference) to “unaided awareness”, usage metrics, ad-recall, NPS scores, perception metrics (e.g. whether people “love” a brand). Researchers and marketers often find it difficult to choose which KPI to track and thus go for “all of the above”, which - if tracked for a dozen or more competitor brands - leads to long and often boring surveys that few people want to complete. Getting enough people to accurately complete poorly designed surveys can be more than twice as expensive as getting people to accurately complete short and easy surveys.
Audience segmentation. Getting reliable data on how “men between the age of 18 and 39 who regularly drive a car and live in a city” (a group that is less than 5% of the general population in many countries) view electric car brands often requires complex quota management in classic survey research to ensure that the final sample doesn’t just include 37 people who fit that criteria, thus making it difficult to generate meaningful insights from it. The granularity of audience segmentation can increase the cost of data collection by a factor of 3 or more.
The frequency of data delivery. How often do we need to get updates on our brand? This is a question that many brand managers struggle with, and rightly so: each “wave” comes with a fixed cost on data collection, so having quarterly updates is four times as expensive as having annual updates, and having monthly updates is three times as expensive as quarterly ones.
Combined, these three drivers can have an enormous impact on sampling costs. Here’s an illustrative overview of data collection costs for different setups of a brand tracking study:
| Complexity level | Tracking setup | Effective cost per interview | Required interviews per year | Total data costs per year |
| Very low | Few brands and simple KPIs, general population, bi-annual updates, large market | $1.50 | 2.000 | $3.000 |
| Low | Many brands and complex KPIs, general population, quarterly updates, large market | $3.00 | 4.000 | $12.000 |
| Medium | Many brands and complex KPIs, niche audience/detailed segmentation, quarterly waves, large market | $6.00 | 8.000 | $48.000 |
| High | Many brands and complex KPIs, niche audience/detailed segmentation, monthly waves, large market | $6.00 | 32.000 | $192.000 |
| Very high | Many brands and complex KPIs, niche audience/detailed segmentation, monthly waves, small market | $10.00 | 32.000 | $320.000 |
The costs of data collection poses a dilemma for many brands: either they buy a cheap version that provides only high-level insights with often limited value, or they invest quite substantial sums to get reliable data at a sufficient frequency. Often, only the world’s largest brands are able to afford the latter for all their markets.
How to cut data collection costs by over 90%
Focus on what really matters and make surveys user friendly Some KPIs like unaided awareness require open questions that often lead to inconsistent data and require lots of manual data cleaning and large sample sizes to deliver reliable results on changes. Other common KPIs rarely change significantly (eg usage data), are campaign specific (eg ad-recall) or are difficult to answer adequately (eg most image statements) and thus don’t belong to a general tracker. At Latana, we focus only on the metrics that really matter: brand awareness, understanding, positive and negative perception, purchase drivers, consideration and preference. We also split up longer surveys into smaller modules that can be individually answered in 1-2 minutes to maximize engagement. This alone enables us to reduce data collection costs by over 50% while increasing response quality significantly.
Use bayesian statistics to get reliable estimates on niche audiences. Segmenting on simple demographics like age and gender, or on key segments like “car drivers”, or “living in a city”, is often easy with a sample size of 1.000 interviews. But when it comes to drilling down on combinations of those segments (e.g. “men between the age of 18 and 39 who regularly drive a car and live in a city”), things get tricky and results either come with huge margins of errors or the gross sample size needs to be increased significantly. At Latana we employ bayesian statistics to be able to segment on niche audiences while maintaining high precision and a low margin of error. This enables us to provide a detailed view of the results at less than half the sample size and costs that would be required with non-bayesian analytical approaches.
Syndicate data collection. Most of the KPIs are the same between brands, so instead of running exactly the same setup for multiple clients, we collect data on the shared KPIs in a “syndicated” setup (i.e. generate standard data sets that can be accessed by different brands at shared costs). This enables us to collect a high volume of data and deliver more frequent updates without increasing the cost for our clients.
By following the three steps above, Latana is able to offer a brand tracking solution that delivers the value of a large and sophisticated tracker at a fraction of the price. Our standard setup includes:
Monthly updates on key brand metrics that are accessible through a user friendly dashboard
The insights are based on the continuous collection of 50.000 interviews throughout a year per industry or product category
Ability to drill down on combinations of small segments while maintaining a margin of error that is up to three times lower than industry standards
If you’re keen to explore a more affordable way to get high-quality brand insights, or if you’re looking for more frequent, timely and reliable brand insights, you can register below for a free trial.
