DIY Dashboarding Disaster

DIY Dashboarding Disaster

Paragon industries called us about a problem they had with their dashboard.  They desperately needed to avert a DIY dashboarding disaster.

They’d created a KPI dashboard using an off-the-shelf dashboard software application and were having challenges with certain types of graphs.  They needed our help in finishing their dashboard.

After meeting with the team, it became clear that this project was a major source of angst for the CEO and other directors.  It had cost them over $160,000 dollars so far, it had been going for around 9 months and they had nothing to show for it.

The vision was clear

A year ago, they decided a KPI dashboard was essential.  They planned on scrapping the plethora of Excel sheets and monthly reports they had to replace them with a live dashboard that would tell them what was happening right now.

It seemed simple. They had all the data they needed in their systems, they knew what they wanted to see on the dashboard and they could buy any number of dashboarding tools online.

They’d reviewed quite a few different tools and all of them made the complex look so easy.  Most came with a month’s free trial and had price tags at around $100 per month per user.  They liked the look of the sample dashboards and the tutorials showed how easy it was to build dashboard.

The reality was a fairy tale

What Paragon didn’t realise were all the challenges they would face in making it happen.

All their data was fairly well organised and well maintained.

They knew what they wanted to build; a dashboard that could tell them the what was happening there and then.

What they didn’t have was the technical expertise to be able to deliver the project.  They didn’t realise that it was a very technical and specialised project.

They also didn’t have the resources available.  They believed that giving the project to someone as an extra task would be fine.  It was assumed it shouldn’t take too long and it would be a little bit like using Excel.

They couldn’t have been further from the truth.

Creating a custom dashboard requires three types of expertise:

  • Data engineering
  • Business analytics
  • Graphic design

They were trying to give this project to one person who didn’t have any of these skills needed.

Over budget and very, very late

Getting the data was one of their hardest challenges.  Whilst they had all the data they needed no one knew what an API was, so they didn’t know how to access their data.  When they finally extracted it, their next hurdle was learning how to convert the JSON file to the format they needed.

Perhaps, the most unexpected challenge was in using the dashboarding tool.  The free trial was great, and they got started in no time.  But after they scratched the surface and started using the more advanced functions or started creating the dashboards they needed as opposed to the examples they were shown, that’s when the major challenge came.

They discovered they couldn’t build the dashboards they wanted using the out of the box tool so they needed to create custom graphs and custom dashboards.

What started out as good intentions, a great strategy and a simple outcome turned out to be one of the biggest problems they had.  It also became a huge distraction to the business.

They now have this dashboarding tool that doesn’t do what they need but they feel compelled to use it because they have invested so much time into it.  Even though it’s providing them with less value that they had before they started the project.

A happy ending

A project like this doesn’t happen with just one person, it requires a team.  A team of experts that know exactly what they are doing, that have made all the mistakes in the past and can spin up the dashboards you need efficiently and with little or no impact on the business.

Within four weeks and for a cost of $25,000 we built them the KPI dashboard, connected it to their data and made it accessible to everyone on any device.   We also provided them with 3 hours of monthly support for all their dashboarding needs plus the chance to discuss future projects with our data scientists and business analysts.

No Comments

Sorry, the comment form is closed at this time.