Digital Consumer Intelligence for Digital Advertising: fashion luxury e-commerce

In a previous post, we anticipated the release of a new feature that will be strategic for many decision-making aspects about customers and brand lovers.

Now, we’re ready to show you how to perform a Digital Advertising campaign for a luxury-fashion digital store figuring a Buyer Persona comparison for some Brands he sells (Armani, Gucci, and Balenciaga) with just a few clicks.

A Digital Consumer Intelligence platform allows you to do something that you couldn’t do in the past. One of that is the clusterization and profiling of specific audiences simply using a KPI6 account and a couple of Keywords related to the Brand or the Topic you want to analyze.

In the next paragraphs, after a little introduction to our approach, you will see how to use information gathered from those audiences to run a Facebook Campaign.

Are you already dealing with Digital Advertising? Are you aware of the end of the cookies’ era? Get a solution!

Experts from large companies, such as Deloitte, Group M and ENI have already found an innovative way to use this technological cookieless model for their creative strategies.

The biggest difference between Social Media Listening (SML) and Digital Consumer Intelligence (DCI) approach is that the first one studies the phenomena dwelling only on the Listening, while the Digital Consumer Intelligence considers listening just as the first of multiples steps.

DCI has the ultimate goal of understanding the “who” behind the phenomena to activate him with an appropriate advertising campaign, as we will see through this article.

Given the huge amount of online conversations generated by each of the four brands considered in our analysis, we decided to dedicate one query per Maison (i.e. run single research for each Brand and then compare the results).

Any query includes:

When the query is ready, it’s time to test it to understanding if it has been properly built and start our analysis.

Until now, the campaign launch process was like a  “witch hunt” stuff, but in KPI6 we’ve worked hard to make it more simple, pleasing and controlled by Marketers.

When we handled the problem, we asked ourselves: in which way could a Company address a Social Advertising campaign strategy without running old and expensive procedures like preliminary A/B testing for targeting the right audience and have at least the same result with respect to the previous process?

After months of our Devs Dep efforts, last week we released Polygons, answering our question.

In the following paragraphs we show you all the steps needed to run in a few clicks a Facebook Campaign in a new and more performant way:

As we saw above, the first step is to set up a simple query that includes the main keywords associated with your Brand.

Then, in a couple of seconds, we’re able  to display a preview of our research:

In this case, we decided to analyse the data of lasts month.

If you look at the chart and if it seems there is a lack of data,it means you have to add more keywords to your Query. In this way, you will be sure that your query was built in a proper way.

In our analysis 40.000 conversations represent an optimal data source.

For each number of Brand (or Topic) you want to analyze there must be a Query. In this example the queries are three: one for Gucci, one for Balenciaga and another one for Armani.

Since your preview shows that conversations are enough, you can launch associated Query and get a coffee-break while we download all the Data you needed to work with.

When this process is completed you can go fast to the next step.

In this stage, we’ve to extract from the bulk of conversation all the clusters’ audiences we want to observe and analyze in deep.

For instance, we decided to handle only the Brand Lovers for each Fashion house because the final goal of our work was to set up a powerful Ads Campaign on Facebook.

Bringing the people who care most about those Brands gave us the possibility to know a priori which could be the best psychographic pattern to work with during both the creativity phase (including tone of voice) and Facebook set-up phase.

To do this, we’ve simply filtered the conversations based on positive sentiments with even positive emotion, such as joy and admiration. Then, we’ve extracted three clusters of people who were involved in those data based on the entity they were talking about.

At the end, we can find on KPI6’s Audience section three groups of people:

Now, in relation to these groups, our platform shows you demographic insights extracted from the elements like profile pic and bio and interests generated by Artificial Intelligence from users’ feed.

Moreover, they also include more than 2.000 Data Points such as Education, Income, Purchases (both past and intentional in a range of six months), Household status and so on, to better understand them even in a Market Research prospective.

This amount of data allows us to apply both demographic and psychographic segmentation to the current audiences. That provides us a deep insight that complements what we have learned about our audiences till this moment.

At this point, you’re quite ready to use that information in which way you want, especially for a super performing Digital Advertising campaign.

The last step, that is explained in the next paragraph, shows you how to compare these audiences through our new feature “Polygons” in order to find similarities among them and fine-tuning your ads strategy.

The last step to understanding how much the audiences we created differs, our new technology, Polygons, must be applied.

This feature allows representing customers by transforming their interests, extracted, as we saw, during the audience analysis into a single 2-dimensional point. Then, a polygon is calculated for each audience.

Overlapping the generated polygons, we provide an index for their similarity that conveys how much the audiences are similar.

The result suggests to us that Gucci and Balenciaga’s customers have similar interests, while Armani’s customers differ from the other two. The reason for this difference can be interpreted in different ways.

For instance, Armani tends to attract a different kind of Buyer Persona, more aged and with a very different lifestyle habits with respect to his competitors.

Regardless of the platform used, the insights extracted from Polygons can flush the characteristics of an audience.

Whatever it is, an existing audience, (a previously trained Facebook Pixel or an audience on a DMP) or a new one (for a new product launch or a repositioning), we will now explain how to support KPI6 Polygons with your decision-making process.

Starting from our interest comparison between several audiences on a 2-dimensional plane, the percentage of affinity between multiple clusters of users is something we must use in this case.

Considering that our brands are responding to entities and therefore to interests that can be recalled by the editor of the Facebook Ads campaign audience, we can apply these conditions: : 

Despite finding a perfect operational application in the improvement of Facebook campaigns, Polygons is much more than that. In fact, it has a huge number of possible use cases in every industry and situation. 

For instance, one of our customers, leaders in the Telecommunication market, has been applying Polygons to understand how much his customers differ throughout the national territory in order to modify his go-to-market strategy. 

The applications are limitless. Another example comes from our data journalism team, which has been using Polygons to study the relationship between political parties’ audiences and the use of aggressive tone during Political Speech.

And that’s not all! 

In the next months, we will release a lot of new disrupting features that will radically change the Market Research industry forever. So, keep in touch with us.

Have you enjoyed this article and do you want to know more? Watch our Digital Consumer Adventures webinars.

Experts from large companies, such as Deloitte, Group M and ENI have already found an innovative way to use this technological cookieless model for their creative strategies.