“How can I export a csv from that?” What do you do when users won’t use your dashboards?

4 minute read

You’ve automated the ETL and built beautiful BI dashboards but the users won’t touch them.

Instead, they just keep asking for csv files so they can import them back into Excel.

They seem to want to do exactly what they’ve always done – even though your new solution produces their exact same reports, faster, better and (finally) with some security.

You’ve gone above and beyond to build something that is really great for them. Investing hours of effort and energy creating something amazing and no-one is interested to use it.

“I just need a csv I can import into Excel. Thanks.”

Whilst everyone on the internet is seemingly talking about the latest data visualisation tools, you are stuck in the dark ages making csv’s!

What is this nightmare!?

The first thing to consider is it even your problem?

Many would argue that it isn’t.

They might say the users don’t get it. They never will. Just give them the csv and don’t ask. In other words, give up caring.

Or they might argue that the users should not be “wasting” company time (and money) gathering data, consuming information or building their own reports. No. These people should be “doing what they’re paid for”, “doing analytics” and “making decisions.” In other words, they should leave the report building to the experts …

In these cases (and sadly in others) the result is that the failure to use technology is being made out to be the users fault. Well it isn’t.

If you’re here, reading this, then my guess is that you know that, you do care and maybe you’re looking for a more constructive and positive solution?

So what is really going on here?

People only want csv files to do something that your tool is not enabling them to do.

It is almost impossible to predict all such use cases so to succeed you need a way to find out what’s really going on.

Unfortunately, this is not easy.

Sometimes people feel they have good reason not to tell you important things. Reasons might be politically or confidentiality motivated. Equally they might be trying to avoid some kind of personal or team embarrassment.

Alternatively, maybe they do not know what is possible with technology. They therefore never considered your tool could help so never thought to put it on the requirements spec.

It is a long potential list.

How could you have navigated things differently to ensure you were not wasting your time?

What could you have done to ensure you had total clarity on these (hidden-until-now!) user needs?

What if you could have taken that knowledge to determine precisely how to employ your tools & technology so that they would actually get used?

Wouldn’t it be great to be safe in the knowledge that whatever you are producing will be used and will make a difference?

How great would if feel to have total clarity on what is really needed right from the outset?

What is needed is a good discovery process.

Here is what has worked well for me.

Start by getting out from behind your screen.

Go and sit with your users. Not in a meeting room but actually at their desks.

Get them to take you through their work.

Observe what they’re doing, with the csv’s, with everything.

Ask “Intelligent Naive” Questions.

For example, ask

  • Why they are doing things a particular way?
  • What is the history behind their processes?
  • What hinders them?
  • What helps?
  • Where do these processes fit in with other people?
  • What are they fundamentally trying to achieve?

Explain you are there to understand them and their processes first, not to judge or “sell” them technology they don’t want.

Often what might seem trivial to you, will be a big deal for them (and vice versa!)

Don’t be tempted to jump straight into technical solutions. We can’t help it sometimes!

Listen & learn.

Then categorise each piece of information they are telling you into ways technology could help them.

Think human-first not tech-first.

Try this checklist and ask yourself:

What could …

  • … save them time (i.e. less copy and pasting, easier formatting, short scripts)
  • … buy them flexibility (i.e. filters, scenarios, what ifs, combining data)
  • … help them better communicate, make a point or tell a data story (i.e. report, chart)
  • … give them reassurance (i.e. improving their security, backups, accuracy)
  • … help them discover something new (i.e. a combined metric, a trend)
  • … help them understand (i.e. a metric, a visual, a pivot)

All the time consider them as a person, the impact it will have on them, their job and their life.

How can you make this a positive for them?

By interacting with your users in this way, you will learn an order of magnitude more, build trust and be better prepared than even to know you’ll be working on something that your users will actually want.

I’ve put this checklist together into a free PDF for you to print out and use. Complete the details below to get it.

If you try it, let me know how you get on ([email protected]). Would be great to hear what works and what doesn’t for you.

Getting this right is so important for you and for your users.