For leadership teams
Executive dashboards that leadership actually uses
An executive dashboard gets used when every number on it has one written definition, every chart serves a decision leadership makes each week, and the data is fresh enough to act on before the meeting ends. Most dashboards die because nobody owns them, their numbers disagree with the finance report, and they answer questions nobody asked. Vantage Systems builds dashboards around those rules, including the one a Saudi holding company's leadership uses to follow contract value, earned, billed and collected across its projects.
Mohamed El Askary, Founder, Vantage Systems
Updated · 7 minute read
Why do most executive dashboards stop getting used?
They lose trust in one meeting and never get it back. Someone reads a revenue figure off the screen, the chief financial officer says that is not what the close shows, and from then on the dashboard is a picture people glance at while the real discussion runs off a spreadsheet. Underneath that moment are three causes: undefined numbers, no owner, and charts built because the tool made them easy.
The research on adoption is sobering. A 2022 study of 214 companies by the Business Application Research Center and Eckerson Group found that on average 25 percent of employees actively use business intelligence (BI) and analytics tools, and Wayne Eckerson described adoption as stuck around 20 percent for years. Meanwhile, 87.9 percent of respondents to Wavestone's 2024 survey of data and artificial intelligence (AI) leaders said investment in data and analytics was a top organizational priority. Companies keep spending on the reporting layer; most people keep not opening it.
The fix is not a better chart library. It is going back to what Stephen Few wrote in 2004, that a dashboard is a visual display of the most important information needed to achieve one or more objectives, arranged so it can be monitored at a glance. The word that matters is objectives. A leadership screen exists to support specific decisions, and everything that does not support one should leave.
How do you get one definition per number?
Write the definitions down before building anything, get finance and operations to sign the same page, and display the definition one tap away from every figure. A metric dictionary entry needs six fields: the name, the formula, the source system, the grain (per project, per month), the owner, and the exclusions. If two departments cannot agree on a formula, that disagreement is the first finding of the project, not a detail to settle later.
Money is where this bites hardest. In the operations dashboard we built for a Saudi holding company, the money view rests on four figures, and each one hides a judgment call that has to be settled in writing.
| Figure | What it measures | Questions the definition must answer |
|---|---|---|
| Contract value | Total value of awarded work | Are approved variations included? Is value added tax (VAT) in or out? |
| Earned | Value of work completed to date | Measured from site progress, certified progress or cost incurred? |
| Billed | Value invoiced to the client | Counted on invoice issue or on client acceptance? How are credit notes handled? |
| Collected | Cash received against invoices | Is retention held back counted separately? Which bank date counts? |
The gaps tell the story
Once the four are pinned down, the differences between them become the most useful lines on the screen. Earned minus billed is work done but not yet invoiced, which usually points at a stuck certification or a slow project manager. Billed minus collected is the receivables problem, which belongs to finance and sometimes to the chief executive's phone.
Which weekly decision does each chart serve?
Every chart on a leadership screen should map to a decision someone makes on a regular cadence, with a named person who acts on it. Run that test on an existing dashboard and most tiles fail it. The ones that remain are the dashboard.
| Chart | Weekly decision | Who acts |
|---|---|---|
| Earned versus billed by project | Which projects need an invoice pushed out this week | Project directors |
| Collections aging by client | Which clients get a call, and from whom | Finance lead, sometimes the chief executive |
| Equipment location and utilization | Move, rent or release equipment between sites | Operations manager |
| Crew allocation against plan | Reassign crews before a site falls behind | Operations manager |
| Projects off schedule | Which project gets a leadership review | Chief operating officer |
How should drill paths work?
Every number on the top screen should open, in one or two taps, into the rows that produce it. That is how trust gets built: when a director doubts a figure, they can see the twelve invoices behind it instead of emailing someone to ask. A dashboard without drill paths forces every doubt into a meeting.
A good path follows the organization, not the database. In the research impact dashboard for a university, spending is set against research output at three levels, faculty, department and researcher, so leadership can move from a faculty that stands out to the departments and people behind it. Each level answers the question the level above raises.
- Top: the handful of figures leadership reviews every week.
- Breakdown: the same figure split by the unit that owns it, such as project, region or faculty.
- List: the records that make up one slice, sortable by size or age.
- Record: the single contract, invoice or researcher, with a link to the source system.
How fresh should the data be, and who owns it?
Fresh enough for the decision it serves, and labeled with its age on every tile. Collections reviewed in a Monday meeting need data from Sunday night, not from the last month-end; crew allocation may need same-day data; board metrics can be monthly. A visible as-of time stops arguments about whether a number is current before they start.
Ownership has two halves. A business owner answers for each metric's definition and for acting on it; a technical owner answers for the pipeline that feeds it. When a figure breaks, both names are on the tile. Most freshness problems are really data infrastructure problems: unreliable exports, manual spreadsheet steps and systems that cannot be read automatically. Fix those first, or the dashboard inherits them.
- Show the refresh time and the source system on every figure.
- Alert the technical owner when a feed is late, before leadership notices.
- Review usage monthly and retire tiles nobody opens.
Does an executive dashboard need to work on a phone?
Yes, and it should be designed for the phone rather than shrunk onto it. Senior people check numbers between meetings, in cars and before walking into a site visit, and a desktop layout squeezed onto a small screen gets abandoned on the first try.
The phone view should carry the top figures, the exceptions that need attention this week, and the drill path, nothing else. Wide tables become lists, hover tooltips become taps, and every chart gets a one-line text reading of what changed. Our operational dashboards are built with the phone layout designed first and the desktop view expanded from it.
Frequently asked questions
Should we use Power BI or Tableau, or build a custom dashboard?
It depends on how much the screen must do beyond charts. Off-the-shelf BI tools suit standard reporting on clean data. A custom build earns its place when the dashboard needs workflows, such as assigning follow-ups, tying into operational systems, or a phone experience leadership will actually use.
How many metrics should an executive dashboard show?
As few as the weekly decisions require, which is usually under a dozen on the top screen. Everything else belongs one level down in a drill path. If leadership cannot read the top screen in a minute, it is carrying too much.
What if our data lives in spreadsheets?
Start there, but plan to move. A dashboard can read structured spreadsheets for a first phase, provided each one has an owner and a fixed layout. Over time the figures that matter most should come straight from the systems where the work is recorded.
Who should own the executive dashboard?
A named business leader, usually in finance or operations, owns the definitions and the agenda it serves. A technical owner keeps the data flowing. Without a business owner, a dashboard drifts into a collection of charts nobody is accountable for.
How do we know whether leadership is really using it?
Measure it. Log who opens which view and how often, and compare that with the meetings it is supposed to support. Tiles that nobody opens for a quarter should be removed.
Can the same dashboard serve the board and the leadership team?
Yes, if both read from the same definitions. The board usually needs a monthly, smaller cut of the same figures; leadership needs the weekly detail. One metric dictionary behind both prevents two versions of the truth, which is the approach we take on portfolio dashboards for private equity.
Sources
About the author
Mohamed El Askary, Founder
Mohamed builds the systems Vantage ships, from data pipelines and models to the screens people use. He is a United States Air Force veteran and a Chapman University graduate in data science and business finance.