January 2018

Digital Ethnography

Experiment with digital methods to define your personas faster
Starting point

The amount of information generated by users online is constantly increasing, but traditional research techniques don't offer quick and effective ways to collect and analyse that data. How can designers use conversations happening on social networks to better understand behavioral patterns?

Clusters of Twitter users discussing the topic of nature in Paris
What we did

Digital methods allow to approach the amount of data generated on social media and help designers formulate assumptions around existing personas. Digital methods require to identify one or more relevant social media where to conduct the exploration, and some research queries that often recur in the online discourse around a specific topic. The extraction of all the content where the defined keywords appear generates a dataset, that can be used as a starting point to create visual maps. These maps guide towards the identification of specific clusters of people who share the same attitude or vocabulary around the defined theme. Different clusters represent the initial step into the formulation of a set of personas.

Why it is valuable

Applying digital methods to elaborate initial assumptions around personas is a powerful way to take into consideration the online world as an important dimension where people express their opinions more and more and manifest their mental models. Digital methods have the advantage of being economically-effective, and quickly generate a hypothesis of who the personas are in the given context, and what drivers define their behaviour. Those hypotheses can be used to streamline the preparation and execution of a field-research study, or as input to an ideation session.


The integration strategies to conduct ethnographic research across online and offline world could be further evolved to develop insights that reflect a deep conjoint understanding of both spaces. Besides social media, digital methods can be applied to existing dataset such as web analytics or blog archives, giving the opportunity to leverage available knowledge before approaching any discovery phase.

Related tips and stories

Explore similar best-practices