• en
  • fr
Intelligence économique et veille : pour que l'information se transforme en action

Social listening: avoiding the pitfalls of a hasty analysis

Photo credits : George Pagan – Unsplash

Many are the businesses wishing to speed up their digital transformation by implementing new services for their marketing and communications teams, or by integrating new technologies as data lakes or artificial intelligence.

Social listening has been part of this trend through editors promoting innovative technologies able to provide insights as fast as we heat up Asian noodles in a microwave.

While it is true that platforms are continuously improving in terms of settings, artificial intelligence, data filtering and data visualization, let us remind you the basics of a quality insight beforehand.

Data Quality

Crédit photo : Marcus Spiske – Unsplash

The quality and volume of information we collect through social media depends on the platforms, industries and monitoring topics we deal with. For our clients, we compare on a regular basis the level of source coverage demonstrated by the different platforms, their evolutions and thus provide them with the most appropriate one in terms of industry and business needs/resources.

If you are familiar with business intelligence and data collection, the main topic being on everyone’s mind in the social listening industry is source availability.

Do we have to worry about that? The answer is yes when the report from a study ordered to your agency shows a limited number of sources and foreshadows the strategic orientation of a brand. Because if we stick to the prismatic vision of a data collection implemented without checking the sources, their possible limitations or their relevance, the resulting recommendations are likely to be biased and deprived of added value.

The data-driven paradigm collapses like a house of cards

Beyond social listening, the subjects of data governance, homogeneity and quality are well at the heart of the matter.

How to improve your insights’ reliability?Insights? 

Implementing a data quality program can consequently improve the quality of the data you collect. Data made reliable means a better relevance to the subsequent analysis.

Different approaches are to be found at these different levels:

  • Checking the source coverage on the platform
  • Checking SPAM and content farms
  • Queries’ perimeters too or not enough exclusive regarding the question asked
  • Data filtering regarding the required analysis
  • Manual cleaning of inconsistencies

All these background tasks must be carried out on a regular basis, and sometimes in various languages according to the listening perimeter

Data hybridization: cross-checking the points of views. .

Photo credits : Tobias Fischer – Unsplash

If today collecting data comprehensively has become a hassle due to constraints imposed by GDPR and social networks, implementing a methodology to drive your analysis will increase your studies’ reliability.

Data hybridization, i.e. usage of different data sources, gives context and perspective to the first insights.

Search data (what people look for) generally complements in an acceptable way Social data (what people say): this choice can be discussed depending on the type of question or the study’s aim.

Diversify data sources to make your analysis reliable

This choice can be made by the Data Analyst in charge of your study, the availability of resources will then be taken into consideration.

What you need to know before launching your in-house Social Listening service ?

Beyond the choice of the tool which essentially involves technical and financial considerations, the main points to settle are:

  • Data governance: within which framework data management from Social Listening takes place?
  • What about the target organization in the insights’ generation? It needs to be sized up depending on the internal clients’ typology, the linguistic perimeter, and the anticipated volume of demands
  • The resources which will be allocated to the analysis and the data distribution to reach your targets with readable and relevant contents.

These questions need a deep reflection on how to organize the companies’ data models to integrate social data in your decision-making

Find here a presentation of Actulligence Consulting’s services in social listening.

Julien Bontempi – Consultant in digital strategy and Social Data Intelligence with Key Accounts, SMEs in Aeronautics and Defense, Dermocosmetics, and NGOs

Published by Julien B

Consultant en stratégie digitale & Social Data Intelligence auprès des Grands Comptes et PME du milieu Aéronautique & Défense, Dermo-cosmétique et ONG.

Leave a comment

Your email address will not be published.