Rich, Deep and Powerful Insights Have Never Been Easier

Don't forget to share this post!


When we think about social media intelligence platforms, we often associate richness and strength with complexity. Indeed, historically, you had to choose between robustness of the platform allowing you deeper insights and simplicity of use. In this article, we will show you how, at Linkfluence, we reconcile powerful data structuring which is key to surfacing deep and valuable consumer insights with simplicity. You no longer have to choose!

What is Advanced Data Structuring and Why does it Matter?

Definition of Data Structuring

Data Structuring is the ability to organize a large volume of data in a way that makes it easier to analyze and derive high-value and actionable insights. 

Once raw data is structured in the right way, analysts can manipulate it to get a much deeper analysis than basic social listening platforms can bring.

It is therefore an essential step to perform deep consumer intelligence analysis, surfacing the type of insights that historically required traditional market research, but leveraging social data’s ability to be always-on, massive and relying on spontaneous expression.

How does it apply concretely to brands

To better understand how critical advanced data structuring is, I will share some examples of how it applies to brands and how it offers more than standard listening platforms alone.

Revealing moments of consumption

A wine and spirit company wanted to understand how and when consumers use its products, to create more engaging activation campaigns. With the help of Linkfluence's experts, they categorized billions of pieces of content into several categories in order to reveal consumption moments and better address consumers based on their behaviors. Doing this requires the platform to understand the many different ways social media users can express what we categorize as “Small Celebration” or “Big Night Out”. And as you might have guessed, this goes way beyond #smallcelebration or #bignightout… which is where the ability to combine AI techniques such as advanced semantic models or computer vision with industry expertise is critical. 

Data Structuring - Moments of consumption


Understanding topics of interest

Another customer evolving in the fashion and luxury world wanted to better understand topics discussed by its target audience to adapt its content strategy and better engage them. Data structuring allows customers to reveal topics and sub-topics of interests. For example, the topic of society helps customers better understand how this target audience talks about societal topics like LGBTQ+, feminism, Black Lives Matter, or inclusivity. With this knowledge, this brand is able to create much more engaging content. Here again, the essential step is to understand the many different ways abstract concepts such as feminism or inclusivity can be expressed by social media users.

Data Structuring - Understanding topics of interest


Providing insights during an acquisition

Other wine and spirits customers used Radarly and its huge data structuring capabilities to help the team in charge of Mergers and Acquisitons to make the right decision. They gave them a lot of valuable insights regarding who was the best candidate for the acquisition depending on their association with moments of consumption. Thanks to the data structuring put in place in this project, they’ve been able to  perform a multi-dimensional analysis and visualize it through the following pivot table enabling them to instantly identify the white space.

Data Structuring - Associations brands and moments of consumption


Find the best brand ambassador to work with

The last great example that I’d like to share with you is the one of a brand selling sports performance equipment which wanted to evaluate the best athletes to work with, depending on what topic they are most relevant in. Data structuring allowed them to categorize billions of content by several ambassadors (here, athletes) and several key topics like health, pop culture, entertainment, social justice, and fashion. They then decided which athlete was the best fit to be the spokesperson for each specific topic in order to better engage the fans.

 Data structuring - ambassadors


To summarize what data structuring brings, let me give you a comparison of the different levels of consumer insights depth:

  • Let’s start with social monitoring which collects relevant and comprehensive data coverage. Here is an example of insights you could get with it:

Cristiano Ronaldo was the Nike athlete most mentioned last month

  • Let’s go a bit further with social listening. In addition to the same level of data collection, social listening will bring generic AI-enrichment like sentiment, emotions, topics, keywords, location or even author’s demographics. Taking the same example here what you will know:

Cristiano Ronaldo was the Nike athlete most mentioned in a positive way last month. Cristiano Ronaldo 4.5x more positive than negative mentions last month.

  • Finally, let's see what an AI-enabled consumer insights platform with advanced data structuring capabilities brings. It allows:
    • To add custom AI-enrichments and topic models like industry-relevant topics, market segments, product categories, brand perception drivers, moments of consumption, etc. 
    • A multi-dimensional analysis: being able to pivot and analyze data by combining multiple queries, filters, custom enrichments and custom topics. 

This way, you’ll learn:

Cristiano Ronaldo was mostly mentioned in relation to the Health topic last month. 

The Nike athlete most mentioned for Entertainment last month was Serena Williams.

The Nike athlete most mentioned for Social Justice by women last month was Megan Rapinoe.

Through this example, you can see that with data structuring you can understand the correlations with complex topics such as Health or Entertainment, and cross tabulate the data across several dimensions.


How is data structuring processed in an AICI platform

Concretely, data structuring comes into play pretty quickly in the process: once we have retrieved the data, enriched it through AI and gathered everything in an index dedicated to the client’s project, data structuring is the last important step before being able to play with the platform.


To create the best data structuring for your project, here are the steps you need to follow:

Step 1: Define exactly what business questions you want to answer

Depending on this, the data structuring methodology can vary. For instance, if you want to monitor your brand reputation, we will focus on the three main pillars of brand reputation which are your brand, your major corporate topics (like sustainability or working environment) and your C-suite.

Step 2: Create the project architecture that will provide the multi-dimensional, industry relevant framework of analysis

This step is fundamental to your project in enabling you to perform thorough and deep analysis. As a point of comparison, this is as important as getting well-defined blueprints when you want to build a house.

Step 3: Structure the data that matches the project architecture by creating lots of queries, adding custom fields, creating a corpus of authors, and more. This is necessary to capture, segment and aggregate the data and finally create transversal analyses of the data set, which is done by sorting, filtering, organizing the data captured in your project.


Linkfluence, a powerful platform allowing advanced data structuring


Our technical capabilities allow you to enrich, organize, and filter the high volume of data captured in your project, in particular:

  • Custom index: Linkfluence Radarly platform builds a dataset tailored to your specific needs into a custom index. This enables you to add or remove data sources based on their relevance to the use case and have a longer historical archive, starting from the beginning of the project.


  • Rule-based custom tagging: ability to automatically tag all of the content in the index in ways that are specifically useful to the project (sentiment, language, countries, keywords, or even your own custom fields). Linkfluence Radarly enables you to do so at scale by creating rules that can apply to new content or historical data. 


  • Topic modeling: AI technique that analyzes millions of social media posts and automatically categorizes them into a 3-level ontology of categories. This makes the data easier to understand, quickly and at scale (as it would be impossible for humans to do manually).


  • Complex queries: Users can build complex search queries on their data set, on multiple levels in a way that simplifies complex projects and enables multi-dimensional analyses. For instance, they have the ability to mix search and focus queries or to leverage our topic modeling capabilities.

As you can imagine, when it comes to large global brands using consumer intelligence platforms like ours to inform lots of decision-making processes, it involves managing hundreds, even thousands, of search queries, focus queries or custom fields. This situation could be stress-inducing but don’t worry, we have two solutions for you!


A brand new powerful User Interface allowing to do this advanced data structuring easily


As McKinsey says, as technologies and business models continue their rapid evolution, companies are experiencing a step change in the workforce skills they need to thrive and grow. Consumer Intelligence is no exception. 

We see more and more large brands trying to address this skill shortage by hiring consumer intelligence experts in charge of exploiting and mastering AI-enabled consumer intelligence platforms, without being dependent on third-party players to master it.

That’s why the new Radarly setting interface makes managing data structuring for small or large projects easier than ever. 

New UI interface - Settings


We’ve started by delivering a new interface to create and manage your queries that follows the best UX practices to make your life easier. 

new queries tab - Radarly


During the next few months, we will keep polishing the entire settings interface to deliver a seamless experience that provides high levels of autonomy so that you can master your own advanced data structuring to the extent desired.


Even though it's never been easier to directly leverage advanced data structuring capabilities, we will continue providing turnkey solutions including managed services where our team structures the data for you. And we can even combine managed services with self-service to offer the best of both worlds to our clients and adapt to their users’ technical maturity levels with social data that can greatly vary across markets and roles. 

It has never been easier to take the most out of Radarly to run your Consumer Intelligence Program! Do not hesitate to contact your Account Manager if you have any questions or if you’d like to request a demo now to see it in action. 

Don't forget to share this post!