How to track brand metrics by target audiences using Google Data Studio [step-by-step guide]
Understand how certain types of people do perceive your brand by creating an interactive Data Studio dashboard
See what we did above? We managed to understand if women aged 26–35 consider using the brand Oreo. So specific!
If you want to know how your brand performs with different audiences, you can use Google Data Studio to combine different demographics or attributes to get a better understanding of who knows your brand and what certain groups of people do think about it.
This comes in very handy for the brand managers out there who need to understand how their brand is positioned. Once you explore this dataset, you will be able to make stronger decisions and improve your brand positioning.
In a previous article, we did go over tracking brand performance using Google Data Studio, without going much in detail about segmentation by audiences.
In this one, we will create a dashboard to visualize how your brand performs with specific audiences.
Objective: measure brand performance across different target audiences
The objective is to visualize how different brand metrics perform with different target audiences.
The measured brand metrics in this fictional brand tracking exercise are the following:
- Unaided brand awareness (UBA): when asked about your business category (e.g. cookies), what do people respond?
- Aided brand awareness (ABA): do people recognize your company’s logo and name amongst others?
- Brand consideration (BC): who considers buying your brand?
- Brand associations: what attributes do people associate with your target?
Data collection: combine surveys with a data science model
For this set-up, we will assume that Cookie Boonie (fictional cookie brand) conducted mobile surveys to collect data.
Given how Cookie Boonie is keen to understand how its brand is performing with a multitude of audiences (on top of its target audience), it collected more than 1.000 responses and leveraged the data science model MRP (Multi-level Regression and Post-stratification) to predict how specific types of audiences perceive brands. Here’s the Columbia University’s research paper on the topic.
Let’s dive in the different steps to build the dashboard above:
Step #1: create unique rows with one single number/result
In Google Sheets, we will create two tabs:
- Brand KPIs: with the UBA, ABA, and BC.
- Brand Associations: given how each brand can be associated with five different terms (quality, variety, healthy, unique, delicious), the sheet structure is different, hence we separate it from the other 3.
For each demographic, attribute, market or brand component (KPI, brand name), we will create a single column.
In this example, we have the following:
- Country
- Demographics: gender, age, education, region, and income
- Attributes: conscious consumer, frequent traveler and target group
By structuring your data this way, each row will be unique. Therefore, once you select all demographics, attributes and brand component, you will have one unique variable.
You may ask: “Why did we write “All?” in most of the cells”?. The idea behind this is to fill the cells with a default text so that you can easily filter later on by the audience of your choice.
For instance, the following row: Country = El Salvador, Gender = All, Age = All, Region = All, Conscious Consumer = All, Frequent Traveler = All, Income = All, Target Group = All, will show you how a brand performed with people in El Salvador as a whole (gen pop).
If you switch the Gender = All to Gender = Female and keep the other attributes equal, then the result you will get is for all women in El Salvador.
If you decide to keep Gender = All and on top of that put Age = 26–35, you get El Salvador women’s aged 26–35 perception of your brand.
Each row represents a different kind of situation for analysis of brand performance, with some being for general situations, thus having ‘All’ in all cells and others representing for specific situations (a specific gender, age range, etc.)
Once your sheet is ready, you can easily connect Google Data Studio to Google Sheets with the Create New Data Source.
Step #2: set all the Filter controls to Single select
This way users will only have one choice and won’t be faced with messed data (recall how each row is unique, therefore we cannot have more than one selection):
To do that, we should check the “Single select” option in the style configuration of our filter boxes.
Style > Single select
Step #3: set a “standard” page and analyze audiences from there
For instance, if you want to know how high-income people do perceive your brand, all you will have to do is select High in the Income box.
Make sure all the other filters are set to a default wording:
Style > Default > Type your default wording (under brand Cookie Boonie, under country El Salvador, under the demographics All and under target audience None):
By doing so, anyone you share with the GDS with will first be presented with the overall view, and then will have to select demographics or attributes to understand how the brand is perceived by different audiences.
Step #4: create multiple perspectives to visualize your data
As you recall, we do analyze both the Brand KPIs (UBA, ABA, BC) and Brand Associations. For each of these, we will have two perspectives:
- Per market: to separate by countries
- Per competitors: to compare by competitors
This is how the data is set for the “per market” view for the Associations:
Dimension: Association
Metric: Result
And for the “per competitors” view for the Associations:
Dimension: Association
Breakdown dimension: Brand
Metric: Result
Step #5: remove misguiding numbers from the selections
Not to confuse the user with useless numbers, select your Filter control and:
Data > untick Show Values
Step #6: reset to default mode to explore a new audience
There is a small issue with this dashboard we’ve built — if you want to explore a new audience, you have to go back to the default mode, otherwise you will be stuck with your current audience.
For instance, if you select Age 26–35 and Gender = Female, you will not be able to select Education = High.
The reason behind this is because we did not collect enough data to combine a certain set of demographics (we cannot know what a highly educated woman with low income living in a rural area and is a conscious consumer thinks about our brand).
We are limited with this binary option in certain instances (conscious consumer or not conscious consumer). Therefore, we always have to return to default mode in the GDS:
GDS: fancy playing with the tool?
Bravo! You can now build your own dashboard for your brand.
This also helps you understand how you are positioned against competitors, and pushes you to set goals to either keep your number position or aim for it.
Here’s the link to this mock-up GDS.
Teach me more GDS tricks!
