Trust Doesn’t Come with the Dashboard
A mandate can require use. It cannot require belief.
This is the second post in a series. What began as a single follow-on to No Facts Inside the Building earned more space than one post could hold. This post takes up the trust problem. Post 3 will take up the rational actor argument: why even a trusted solution may not be enough.
When we closed the first post in this series, we noted that what happens when a discovery problem goes unaddressed is not hard to predict. We said we would take that up next, and we will, but not all at once. The argument split into two distinct problems, and collapsing them would let each one off the hook. So we are handling them separately.
This post is about trust. Not trust in an abstract sense, but trust in data specifically: what it requires, what undermines it, and why no dashboard, however well designed, can substitute for it.
The Numbers Already Disagree
Start with what is already true in the situation we described. The underlying systems of record are producing different numbers depending on which system you are reading. This is not a minor discrepancy. It means that before anyone opens a dashboard, before any visualization renders, the foundation is already contested. Two people reading from different systems will arrive at different pictures of the same reality.
This is the trust problem in its clearest form. It does not live in the interface. It lives underneath it, in the data pipeline, the collection methods, the definitions, and the governance processes that generate the numbers in the first place. A new dashboard built on top of a contested foundation does not resolve the contest. It inherits it and makes it visible in a new format.
We explored the relationship between dashboards and the systems beneath them in an earlier post. The insight holds here: the dashboard is only as trustworthy as the data that feeds it. When operators already know the numbers don’t reconcile, a cleaner interface is not reassuring. It is cosmetic.
Mandate Is Not Trust
Here is where hierarchical organizations tend to make a predictable mistake. A solution is directed from above, the directive carries authority, and somewhere in the implementation logic, authority gets mistaken for trust.
It does not work that way. A mandate can require use of a system, but it cannot require belief in what the system says. Operators who have watched numbers shift depending on which product they open do not suddenly trust those numbers because a dashboard now displays them attractively. Trust in data is built through a different process entirely. It requires consistency: the same numbers, from the same source, over time. It requires transparency: visibility into where the numbers come from and how they are produced. And it requires a track record of answering operational questions accurately, which means the system has to have been designed around operational questions in the first place.
None of those conditions are met by a solution designed without walking the process. In The Spring and the Aqueduct, we described this pattern directly: the issue in those situations is not information. It is conviction. Leadership has already decided what the solution looks like. Trust is assumed to follow from deployment.
It doesn’t.
What Trust Actually Requires
Closing the trust gap is not a communications problem or a training problem. It is a data integrity problem, and it has to be addressed at the source. That means examining the systems of record, surfacing the places where definitions diverge and numbers disagree, and doing the unglamorous work of reconciling them before any visualization is built on top.
That work is slower and less visible than building a dashboard. It does not produce something you can demonstrate to a higher echelon in a briefing. But without it, the dashboard remains a representation of a contested reality, and operators, who live inside that reality every day, will know it.
If your organization is working through a data trust problem, Storm King Analytics is glad to help. The gap between what your data says and what your people believe is often wider than leadership realizes, and it rarely closes on its own. Reach out at info@stormkinganalytics.com.
There is still a second problem waiting. Even if the trust gap were closed, even if the data were clean and the numbers reconciled, operators may still build their own tools. Not because they distrust the dashboard, but because it was never designed to answer the questions they actually need answered. That is the subject of Post 3.