Top 7 Ways to Add Value to Dashboards for All Users
Before you even break out SAP’s Lumira Designer to create that perfect dashboard for your company’s executives, take a minute to really think about what’s important to make this effort provide value.
Reporting dashboards are analytic visualizations that allow rapid analysis of processes and efficiencies, keeping stakeholders in control of their business’ performance. Like the dashboard in your vehicle, your data dashboard should provide the most important data in real time and allow instantaneous analysis at a glance. You must be able to make quick decisions without taking your eye off the road (or your business). You need to be able to monitor your speed, temperature, and fuel as often as you need and be able to trust the gauges and make good, split-second decisions based on their readings. However, if you live in my home, you may simply wait for the low fuel warning light to come on, then tell me the car is almost out of gas. Speaking of that, your dashboard could also set up triggers to alert you if a measure falls above or below a certain threshold. This is the “check engine light” of your data world.
The value of a reporting dashboard is in its ability to quickly monitor and change behavior to drive incremental, continuous improvements. When used to its fullest with the proper research and thought, Lumira Designer can create game-changing dashboards, making cumbersome spreadsheets and report data easier to monitor and immediately actionable.
If you haven’t had time to properly open the box and read the directions on Lumira, are new to SAP dashboards or still using Xcelsius, give us a call. You may wish to chat about Lumira Designer before you set expectations, or worse yet, set commitments on delivery with the latest evolution of SAP’s dashboard technology. We recently gave someone a free demo who said he wished he’d reached out two years ago, as it would have saved him a lot of work. If you have a specific question, shoot us a note and we’ll try to answer it.
Three Types of Dashboards
An operational dashboard monitors business processes that frequently change and tracks the current performance of key operational metrics that are constantly changing, for instance as shipments go out in a warehouse environment. Going literal with my auto analogy, Kingfisher created a dashboard for Honda that tracks metrics to look for potential quality issues as cars make their way down a production line. Below are metrics that may be tracked in a warehouse environment.
Source: Newcastle Systems
Like the speedometer on your vehicle’s dashboard, this is data that is most often monitored frequently, sometimes even minute by minute. You’ll see these in supply chain, production, logistics, retail, etc. These are all about monitoring constantly and making small, meaningful, continuous changes that add up over time. These could be production numbers, sales in real time, machine yield, etc. Remember, a smaller change over the long haul can mean real money. We have a client who made a 5% change in logistics and that percentage now means $3M annually in savings.
Strategic (or Executive) Dashboards
A strategic dashboard is used to monitor the status of Key Performance Indicators (KPIs) and is typically used by executives. Like the temperature gauge on a car’s dashboard, this health check data is reviewed by executives to take the temperature of the company (or part of the company) from a higher-level view. Strategic dashboards assist executives staying on top of KPIs throughout the business, such as sales, finance, marketing or human resources data. Measuring to forecast, budget or prior year gives decision makers a reference to current performance. The dashboard may show visuals of monthly, quarterly, annually and allow the user to select different views.
An analytical dashboard is used to analyze large amounts of data, allowing users (usually business analysts) to investigate trends, predict outcomes and discover insights. Like your vehicle health report or your 30K mile checkup for your vehicle, this data is analyzed at a deeper level and is used to see if any issues/trends need to be addressed. Since all of the data may tie together at such a granular level, the data behind an analytical dashboard needs to be accurate, trusted and up-to-date. Analytical dashboards often include advanced Business Intelligence features like drill-down and ad-hoc querying, allowing the user more freeform exploration to uncover insights.
Why Dashboards Have Failed
How can some organizations get so much value out of dashboards and some yield no value? Some view their dashboards as simply charts on a page. Others see them as key business tools. To be clear, a true dashboard is much more than static charts on a page. Excel sheets creating charts to paste somewhere represents many of the issues with Business Intelligence initiatives in general. Ideally, you’d like to avoid simply wrangling spreadsheets when possible by pulling data from your single, trusted warehouse or mart into a report or dashboard and avoiding the possibility of spreadsheets being updated differently by multiple users outside the system. You also want to make the process automated, avoiding the time spent on manual updates, as well as creating a tool that is monitoring in real time. Finally, to get a true 360-degree view of the company, expand your dashboards’ reach across your entire business, utilizing all three kinds of dashboards for all types of users.
Do not create dashboards that rely on manual data updates. It’s a sleeper cost and a huge risk. Sooner or later, the employee(s) in charge of manually keeping the dashboards up to date get busy and forget to update the dashboard, or leave the company and no one knows the “secret voodoo” to update the dashboard. Unfortunately, the stakeholders could continue using it, assuming everything is current. Out-of-date dashboards can be worse than not having dashboards at all, as you’re now making decisions based on inaccurate data.
to Building the
1. Make the dashboard easily accessible
- Whether it’s delivered right to an inbox or a direct link that can be bookmarked, make it super easy to access, not buried deep in a folder structure that you need a map to find.
- Lack of adoption is one of the top reasons to doom dashboards.
2. Display reliable data
- Seek a single, trusted truth.
- Do not pull from a manual process, as errors will occur.
- Explain where the data comes from, when it was last updated, and what the business definition of the metrics are.
- Making decisions from a bad dashboard is worse than not having one.
- Ideally, pull directly from a data warehouse or data mart.
3. Make it easy to digest
- The point of a dashboard is not to go through huge spreadsheets looking for negative numbers to trying to find where to start looking.
- Make charts and other elements extremely easy to read and process, then allow for drill-down as necessary.
- For instance, consider simple, clean type of elements where at first glance you can tell you’re either in the green or in the red for that KPI.
4. Show only the most important metrics (They tend to be the metrics related to your time or money)
- Include the most important items to track for the users the dashboard is intended.
- Too many KPIs make a dashboard that’s cluttered and hard to read.
- Split screen views by domain, don’t mix unrelated content on a single page.
5. Use timely data
- The point of a dashboard is to continually monitor and make adjustments.
- Old data makes for dashboards that are more harmful than not having a dashboard at all.
- Particularly with operational dashboards, you’ll want to continually update with accurate, timely data and make small, incremental changes that add up efficiencies over time.
6. Include reference points
- I remember creating a chart years ago and asked my coworker if 2,647 was a lot or a little. He said, “It’s 2,647.” I asked him if 10 wins was a lot or a little. He said, “It’s 10 wins. That’s exactly what you need to know.” I told him if it’s the tenth game of the season, it’s great. If it’s the last game of an 80-plus game season, it’s pretty terrible. Our baseline in this instance was about 325 and we were performing high above normal (unsustainable, in fact). Good in this instance, but short-term and something we had to plan for retraction.
- A trendline may be what you need to show how current levels fit into the landscape.
- Year-over-year comparisons, goals or acceptable levels (merchandise in stock, for instance) can all be reference points.
7. Be customizable
- If you really want to kick it up to the highest level, provide customizable dashboards. Give each user the rights to adjust their dashboard to meet their own specific needs. You should be able to do this with login rights.
The bottom line is, if a dashboard doesn’t get used, it may be invaluable information with no value. That said, you must have trusted data with an automated process, to ensure dashboards are accurate and up to date. Inaccurate dashboards are worse than not having any, as decisions will be made based on inaccurate data.
Remember for whom you’re creating the dashboard. Think about how many times you get asked the same question each week. Talk to your users and get their thoughts on data that they really need. Give them the info they need, the way they’re likely to use it when they need it. After you deliver a dashboard and your users come back with lots of requests for changes, know it’s a good thing. You’ve started them down the road and they’re seeing the value of what you’re doing. Embrace their changes as they’ve embraced the change you’ve begun.
I’m sure there are more ideas out there and I’ve likely missed some. I’d love to hear any “steps” I may have missed or how dashboards have helped you and your company. If you have any other ideas for blog items, please drop me a line at firstname.lastname@example.org.