What we learned at Tableau 2017: How do you solve a problem like a spreadsheet?

eROI sent three of our own–two analysts and a writer–on the road to Tableau 2017. They came back full of ideas and ready for action. Here are some highlights from their trip.

Data tangle/Raw data Illustration

Now with HYPER

Aaron Petrous (Performance Analyst):

Honestly the most impressive part for me was the introduction of Hyper. The big ugly databases that we have did not pair well with the previous Tableau data processing engine. I plugged those same big ugly databases into Tableau, and it ran queries in seconds that took over 10 minutes to run in the past.

The total amount of data that we’re producing is increasing exponentially. We’ve gone from aggregate, campaign-level reporting to collecting every single event, from every person, in an ongoing way. And Tableau has now developed a program that can keep up.

Cleaning/ordering data Illustration

Sharing at high speeds

Malorie Morrison (Performance Analyst):

Adam Selipsky, Tableau’s CEO, identified a real sticking place for me: that the people who have the least exposure to the data are always the ones who either have the most questions or feel the most uncomfortable with analytics. And, as an analyst, you have to stop that early and often. It was refreshing to know that this communication is a problem across the board.


On smaller teams with limited data, or on teams that aren’t super experienced with data, you might have just one person responsible for the process: getting the data, finding structure, analyzing the data, developing insights, and then acting and seeing through recommendations. And that’s a really hard thing to do for one person.

Annie Russell (Writer):

And that isn’t unique to data analytics. I wonder whether some of what we’re talking about is just the scientific method and how complex it is to run all the necessary iterations. Having just one person do that process? That’s hard.

Sharing insights Illustration

Get it together!


Thinking of this from the perspective of a writer: I cannot possibly do my whole job alone. If I start off locked in a corner writing an email–writing it in the ways that I think are best from my little locked room–I’ll send it out to the team for feedback and be shocked by how much I missed. Because I believe the reality is that I’m never gonna capture everything on my own. There’s no way to do it from my one brain. The feedback, the conversation, is crucial. And I wonder if it can feel the same in data analytics. Particularly that tension in Malorie’s point about it being tough to help people understand the data but also really difficult to catch everything if it’s just you and the spreadsheets.


An outside perspective helps, because while you can have a lot of this stuff automated, you can’t automate the insights. The hardest thing to communicate is how much work it takes between an export of our databases to getting the data usable. I think Tableau can be a good interface for sharing the data–it’s not as scary as an Excel spreadsheet; it lets people feel more comfortable.


I totally agree with that. The accessibility of Tableau–how it can be used for both user-specific and aggregate-level reporting–gives people an easy way to understand what the data is saying, which creates more ways for people to be involved with the data. That’s where this process begins.

We’ll keep talking: using our growing data resources to check our ideas and observations, to form hypotheses and test them, to look for new questions and seek out answers in fresh ways. And we’ll be exploring how Tableau can help us get, and keep, that conversation going.

About eROI

eROI crafts compelling digital experiences across email, web, and social channels. Our work has been consistently successful in driving revenue and exceeding goals for our partners.