Attending the 2016 Data Insight Leaders Summit in Barcelona proved to be a very wise decision. Unlike so many buzzword-focussed events, this one actually did attract speakers & delegates who were data insight leaders.
It also featured plenty of variety & interactive formats, to help delegates get fresh insights into their own data leadership challenges.
My role was to chair Day Two of this event, together with the Sales & Marketing track on Day One.
I’m pleased to report that being chair helps focus your attention on the key points & what best to bring out as lessons to remember. So, for the benefit of you readers who didn’t manage to make it to Barcelona, let me try & summarise the main lessons from this conference.
As a group, of about 50-60 leaders, mainly from across UK, Europe & USA we discussed topics as diverse as the best ’technology stack’ to use with your Data Lake & whether you should centralise your data science team. Much of our content focussed on best use of data scientists, but there were also case studies from data governance & BI leaders.
Another interesting dynamic was the discussions between those attending from an IT background (including a couple of CTOs) & those working within business (in roles from Data Scientist, to Head of Analytics, BI or CDO). There was much talk of the relative benefits of going around IT (or ‘undercover’) versus working with your IT team & it’s required processes.
Throughout this event there were a couple of formats of panel debate (some mainly on stage & some with questions from audience). Popular data insight leader ‘hot potatoes’ from these sessions included:
- Can you build data infrastructure needed ‘in-house’ or do you need to 3rd party help?
- Should you centralise of decentralise your analytics team?
- How can you put together the perfect data science team (rather than needing to hire complete individuals/unicorns)?
- Should you recruit the talent to ‘grow your own‘ or buy in services, to get experienced analysts who can add value now?
Most of those topics have no one ideal answer, it depends on your business context & the data maturity of your organisation. However, the discussions were useful in helping highlight things to consider & options available.
It was also really useful to have a cross industry event focussed on this skill set. That gave opportunity to hear data leaders from businesses as diverse as Coca Cola, Linked In, GE Capital, Money Supermarket, Trip Advisor, Deutsche Telecom, BT and British Gas (amongst others).
Sadly, I don’t have time in this blog post to share a full debrief from the event. But, I’ll share excerpts & ideas prompted over the coming weeks, as relevant to different topics.
For now, I thought it might help to focus on the key insights for data leaders that I noted for my ‘closing remarks’. These were under the following themes:
- One obvious difference between those organisations driving real change & value from analytics (verses others) is they have a clear strategy.
- In the best cases this is also clearly aligned to the strategy & goals of the organisation as a whole.
- This kind of alignment & clear purpose helps both collaboration to drive acting on insights & retaining analysts who see purpose in their work.
- If there was one theme from speakers, it was that the key differentiator is people.
- It was also interesting that not just technical skills, but attitudes like empathy & humility came out as important.
- A key need was for the training of softer skills (better questioning, communication, influencing & visualising to name a few).
- Beyond the immediate team, we also covered need to change the decision-making culture in organisations.
- One approach to this was educating the C-Suite, especially the CEO, on what Data Science can do & its limitations.
- Once again there was a call, for the humility to collaborate with other teams & let them sometimes take the credit.
- This was not the chief focus (which was a pleasant change), but many leaders have Data Lakes as ‘sandpit’ solution for analysts.
- It was interesting to see how many actually ‘dual tracked’, also having a structured Data Warehouse for better performance.
- The majority of teams benefited from using R (to a lesser extent Python) & had found benefit in a move away from SAS, SPSS et al.
- It was disappointing on first day to have few mentions of customer; some Data Science teams are too internally focussed.
- This improved on Day 2, especially from those organisations who also used research & market insight to prompt ideas/requests.
- Too few organisations seem to have yet built a strong link from Data Science capability to how they give customers what they want.
Applying to your role
I hope those few thoughts help you. It was a very encouraging event, not least because one saw collaborations beginning to emerge between what have too often been competing teams. CDOs & Data Science leaders working together. CTOs & IT teams seeing the need to collaborate with developments in business analytics teams. Heads of Analytics or BI seeing the need for one another & even how Statisticians & Machine Learning experts can cooperate.
As ever, there are those who believe only in their own approach (e.g. Machine Learning will take over the world). But, the most encouraging future, looks to be greater cross-discipline cooperation. That is beginning to reveal that the more creative, caring & communicating data leaders may well drive the greatest change.
So, keep developing those people & leadership skills – as well as your technical knowledge.