Blog

BLOG

Commitment, Time, People… Dashboards

PUBLISHED ON 12 APRIL, 2015

What determines the success of an Enterprise Digital Dashboard deployment project?

I would point at three factors:

1. Your level of commitment to the very specific mission of performance-driven information delivery
2. The importance you have placed on deployment times and scalability during the platform selection process
3. How much you have taken “people” and organizational constraints into account.

Let’s look at them in detail.

Commitment

The W.H. Murray quotation comes inevitably to mind:

“Until one is committed, there is hesitancy, the chance to draw back, always ineffectiveness. Concerning all acts of initiative (and creation), there is one elementary truth, the ignorance of which kills countless ideas and splendid plans: that the moment one definitely commits oneself, then Providence moves too. All sorts of things occur to help one that would never otherwise have occurred.[…]“.

So very true. And how common it is for us to avoid the hard choices and keep going in every direction at once, hoping to serve every need we can possibly encounter with the one magical set of resources. With the one tool we hope to sanctify as the definitive master of all trades.

This is the classical “swiss army knife” problem: I do not yet know whether I will need to cut some tree branches or bunch of carrots. I do not WANT to know. I demand that my knife covers both needs so that I can avoid making a decision. Surely, when the moment comes I will complain about how pathetically tiny the scissors hidden in the knife’s body are to cut some duct tape.

Hence my first point. I doubt you will get an enterprise digital dashboard deployment right if you are not ready to commit to the most essential nature of a dashboard: Serving the data consumer. Not the analyst. Not IT.

How does Providence move then in our favor (to use Murray’s words)?

• Since you decide to focus on information delivery, you can be “mobile first”. Putting other important missions aside for now, such as “visual discovery” or “analysis”, will set you free to serve powerful, native mobile apps that focus on the easy understanding of metrics, outcomes and performance against goals.
• Because you choose to commit to assisting data consumers in their access to high-level multi-channel metrics, you do not need to integrate all of your data prior to serving it on their plate. This will open the door to tapping unlimited data sources, regardless of their schemas, structure or lack of structure.

Bottom line: You will always need Tableau (or QlikView, or Spotfire!) for visual discovery and “analytical” dashboards, but that is an entirely different problem. Enterprise dashboards demand an extreme focus on cross-platform reporting, distribution and data governance.

Time

While entire corporate departments justify their existence on the never-ending drip of challenging long-term projects -this certainly being the case for traditional IT and Business Intelligence teams- business stakeholders need data today.

It is therefore a plausible thing to aim for a 360-view of every single individual in this and any other galaxy by 2078, but all the while marketing executives and CXOs are losing their patience waiting for the most basic up-to-date, actionable information.

Bearing this in mind during the platform/technology selection process will therefore have a stark impact on your ability to deliver within weeks rather than months.

I would point out two specific features of a digital dashboard ecosystem in this regard:

• Extreme flexibility does not conflict with time-to-market, rather the opposite: a platform that cannot really be adapted to speak the very specific language of the company will end up sidelined at the benefit of the very same manual Excel reports we were trying to avoid. Custom, multi-source KPIs are a must for enterprises at any level of maturity.

• Built-in scalability: a tricky feat when it is to be coupled with the previous one (flexibility), but one that various platforms have managed to pull off in different fields. In the dashboards arena this is mostly tied to “KPI/chart template libraries” that can be defined once for the entire organization, then easily instantiated across multiple departments, stakeholders, brands or countries. This ensures governance through a common language, while it avoids long deployment projects by facilitating reuse of previously defined metrics and graphs (if only for different data segments).

People

Even after showing a full commitment to information delivery as well as paying special attention to the required deployment times for a given platform, a third factor will eventually come into play potentially putting the entire project at risk:

People do not consume information by themselves.

Data consumers want to ask questions, raise objections, demand further details… complain! They are part of a team and it is absolutely essential that this is reflected in the manner in which the interact with the data they are being served.

For things take an interesting turn when people and the complexities of their daily interactions are taken into account. Think of the following consequences of bringing the “people” factor into the equation:

• People think in words, not numbers. We want to communicate insights in plain English, but such insights will not accompany status updates (immediate in nature) from the outset, as they require analysis and paused reflection
• Stakeholders at the decision-making end of the spectrum have little time for exchanges and should be provided a quick way to raise a flag or request further information (a “WTF!” button -as in “Where’s The Food?”)
• Analysts or those in charge of other data management layers (analysis, visual discovery, data integration…) should be able to use that very same communication channel to determine the priorities in their daily data exploration work.

A Digital Insight Management system (often discussed in the Sweetspot blog) provides a simple answer to these and ensures that teams make enterprise digital dashboards a success and not a last-minute failure.

In conclusion, there are many ways to approach the deployment of a corporate digital dashboard ecosystem. But, if my experience can be of any guidance; understanding the most important mission at hand and committing to its sole accomplishment, while minimizing deployment times and ensuring a positive impact on teamwork dynamics, are ultimately proven to deliver success.