Discovering behaviours at Songkick

The team

In this phase, we were 2 product designers working together to discover users' jobs to be done and translate our findings into archetypical behaviours that we could share with the rest of the team.

‍
Analysing data

Since July 2013, there was a survey running on Songkick.com that people could fill after buying tickets. In this survey, they were telling us things like how they knew about a concert, what makes them decide to buy tickets, when was the last time they missed a concert and why, or how excited they were about going to see an artist.

We started analysing all this qualitative feedback, trying to extract the main users' jobs to be done. For instance, the following example is someone's answer to "Describe your journey from the moment you knew about this concert to buying the tickets. How excited were you? Was there something that almost prevented you from buying?"

‍

‍

We started highlighting every piece of the answers that shows a need, a decision timeframe, or barriers. This type of answer was a very common, people were missing on concert tickets because their friends didn't want to go, or they didn't know they have similar tastes. From that, we can extract the job "When I'm interested in a concert I need to know if my friends can come with me, since I don't want to go alone." And then start thinking how Songkick could solve for this job better.

We did the same with almost 500 answers (out of 2,000).  And got a pretty massive spreadsheet full of needs.

‍

Mapping out journeys and jobs to be done

Using all this qualitative feedback, MySQL and Mixpanel, we were able to see users' journeys in detail. We started mapping needs and pain points to the four main steps of the purchasing journey: 

- Awareness: How did they find out about the concert (Songkick, Spotify, Facebook, radio, other concerts...)

- Consideration: How they decide whether to go to the concert (Social, money, logistics, excitement, quality, venue...)

- Advanced consideration: Which factors influence people reconsidering wether to go to the concert (Friends going, cheap tickets, change of venue...)

- Purchase: What are the pain points and frustrations they face, how many tickets they buy at once... (Online queue, scalpers, input errors, credit card details...)

After this process we got to some key learnings: 

- Identified 100+ jobs to be done (JTBD) which provide innovation opportunities for Songkick and competitors/disruptors.

- 25% of Songkick Ticket Purchasers go on to repeat purchase with Songkick.

- We weren't activating Ticket Purchasers effectively.

- Songkick was satisfying only 30% of customers needs.

- Identified 10 archetypical behaviours.

‍

‍

Sharing the knowledge

This phase was key. We needed everyone in the company to know about these needs, to be able to prioritise our work more effectively and make sure we're working solving the right needs. (It was also a great excuse for me to do some illustrations, which fills me with joy).

It was very important to make them self-explanatory, so instead of creating a poster full of descriptions and characteristics I decided to illustrate the behaviours and let them speak for themselves. And to make it very clear that we based our personas in actual research, I included on of the most representative answers from each category.

‍

Why behavioural archetypes instead of "personas"?

Behavioural archetypes are built based on research, both qualitative and quantitative. And they are representations of how a person can react in a particular context with a specific job in mind. This means that the same person can exhibit different behaviours. So the discussions goes around how to fulfil users' needs in a particular scenario, while they are behaving in a certain way.

For instance, in my particular case, I became a concert junkie when I moved to London. If I see something that sounds remotely interesting, I get a few tickets. Sometimes I don’t even know the band, or I have just listened to a few songs on Spotify. I go to at least one concert per month, and sometimes it's one or more per week. So I'd say I'm a concert junkie in most situations.

But. Back in 2014. The day I knew Portishead was on tour and coming to Madrid, where I was at the time, I was the loudest screaming superfan in the country. I queued for 5 hours to buy the presale tickets, I queued in front of the only shop that was selling the tickets, in the physical world, not in front of a computer. Then, the day of the concert I queued for hours as well to get to the first row. Totally worth it.

The great thing about these archetypes is that you help people empathise with behaviours rather than stereotypes that are limited and unvarying.

‍

Ideation sessions

After sharing all this knowledge with the rest of the team, these archetypes were present in everything we were working on. For example, we would run an ideation session to help "social shepherds" go to more concerts. Since ~15% of the mobile app users said they would like to connect with other users, share experiences in social media platforms, get organised with friends...

Using these archetypes to think about an experiment or feature was eventually a habit within the whole team. It became another piece of data that we could use to push ideas forward and make sure we were building the right thing.

Key leanings

- Don't come to conclusions too quickly when analysing behaviours. Try and get to "the why" of things many times.

- Be patient, even if it takes time, you'll start seeing patterns.

- Choose a methodical way of start analysing your results and stick to it.

- Each person can exhibit more than one of these behaviours depending on the context. Human behaviour can't fit in just one pattern.

Thanks for reading!