I was asked a disheartening question the other day:
"What do analytics teams actually do? Just make dashboards?"
And I realized this must be what a lot of folks think. So I thought I'd lay it out explicitly and correct some misconceptions along the way.
"What do analytics teams actually do? Just make dashboards?"
And I realized this must be what a lot of folks think. So I thought I'd lay it out explicitly and correct some misconceptions along the way.
The north star of analytics is quite simple:
Drive business value through data analysis.
And there are three kinds of workflows that make this happen.
Drive business value through data analysis.
And there are three kinds of workflows that make this happen.
1. Dashboards + self-service assets
This is what most folks think of when they think analytics, but it's the least uniquely valuable thing we do. We're exposing data so others can look at it themselves, *without interpretation*.
This is what most folks think of when they think analytics, but it's the least uniquely valuable thing we do. We're exposing data so others can look at it themselves, *without interpretation*.
It's scalable, but rarely high leverage. The heavy lifting (the thinking) is being outsourced, so it never quite feels that impactful when your dashboard is being used.
It's a necessary but small part of the job. it should usually not be considered our main lever.
It's a necessary but small part of the job. it should usually not be considered our main lever.
2. Ad hoc requests
This is where we directly support the decision-making process by lending our own *interpretation* of the data, not just exposing it.
This is where we directly support the decision-making process by lending our own *interpretation* of the data, not just exposing it.
This is where things start to get interesting. There's a lot of value in learning to reframe inbound questions in a way that gets to the root of what's needed. Careful alignment with stakeholders can often lead to more impactful decisions. High impact, low scale.
3. Proactive, strategic analyses
Where we don't just support, but guide decision-making with our findings. This is where even higher impact can arise. We're not only able to see things others can't, but it's high leverage, as others can draw inspiration from your work
Where we don't just support, but guide decision-making with our findings. This is where even higher impact can arise. We're not only able to see things others can't, but it's high leverage, as others can draw inspiration from your work
An example from Airbnb: our team hit half of our yearly goal after I'd proactively done an automated funnel search. It exposed a bug that had been dropping conversion rates in a particular country + device by 40% *for years*. Not hard work, but high impact + applicable elsewhere.
That said, ad hoc, dashboards, and strategic work are all part of the job - it's just about picking the right approach for the business problem at hand. We and our stakeholders need to internalize this. It'll make us all more effective.
Read more here: hyperquery.ai/blog/what-does…
Read more here: hyperquery.ai/blog/what-does…
• • •
Missing some Tweet in this thread? You can try to
force a refresh
