Keeping Up With Data #119
5 minutes for 5 hours’ worth of reading
When investing into data and analytics, executives should not be like Alice in Wonderland. It is their raison d’être to know where they want to get to.
We all know the scene from Alice’s Adventures in Wonderland:
Alice: “Would you tell me, please, which way I ought to go from here?”
The Cheshire Cat: “That depends a good deal on where you want to get to.”
And yet, sometimes (not too often though) I feel like the Cheshire Cat when asking about the business strategy, to which a new data initiative or the whole data strategy should be aligned.
In such situation, the answer is typically either about wanting the data to tell them what to do (mislabelled as being data driven), or not knowing what the art of the possible is (but wanting to jump to the solution quickly).
I learnt that when that happens I need to take a step back and re-iterate that I’m asking about business strategy, problems, or opportunities. That I am not talking about data at all.
Once this is clarified, the answer is often clear, always ambitious, and provides the North Star for upcoming data and analytics efforts.
When it comes to data and analytics, executives do not have to know the way, but they should have a clear view on where they want their company to get to.
Today’s reading list looks at progress with data, analytics, and AI, incident management, weapons of mass persuasion, and ratios of data staff to engineering and product.
- Has Progress on Data, Analytics, and AI Stalled at Your Company? Fortune 1000 companies have been investing heavily into data, analytics, and AI. However, the reported impact of the initiatives has not improved in the last five years. The investment into data and analytics won’t go away, but in order to warrant the resources spent, the way how the companies are investing should change. The article offers four recommendations: 1) focus on culture change and its business impact, 2) start small, 3) build strong business partnership and sponsorship, and 4) don’t forget about ethics. Similarly to the findings from 2017 survey, the challenges are not related to technology. (HBR)
- Incident management for data teams: Managing data issues is data teams’ daily bread. Whether they like it or not. The way how the incidents are managed can build or damage trust an organisation has in its data. The team behind Synq admits there is no one-size-fits-all solution to managing incidents, but they offer (for inspiration) a 5-step approach anyway. The approach consists of 1) issue detection, 2) incident response, 3) root cause analysis, 4) resolution, 5) learnings. (Synq)
- Weapons of Mass Persuasion: Tracing the Story of Psychological Targeting on Social Media: A story about detecting users’ personalities at social media platforms and using them to influence real behaviours and actions. “The fact that Facebook likes just so happen to correlate, to a certain degree, with people’s personalities does not imply that targeting people with an ad based on their personalities actually causes them to do anything. Or does it?” Well, it turns out it does. And many private companies have made a business out of it — offering digital propaganda services around the world. (Behavioral Scientist)
- Data and product team size relative to engineers: A one for the fans of benchmarks and ratios. What is a good ratio of data staff to engineers? Or to product teams? “The data team size relative to overall company size ranges from 1% to 5%.” And that is for 50 tech companies analysed.(Inside Data by Mikkel Dengsøe)
Enjoy the weekend and remember that keeping up with data is easier than catching up.
