The dashboard that already existed
August 22, 2026
Three weeks in, somebody mentions that Finance built this last year. The expensive part is not the three weeks. It is that there are now two of them, and by Thursday they will disagree.
Three weeks, and then the meeting
An analyst is asked for a view of monthly revenue by segment. There is nothing in the reporting tool that answers it, so she builds one. It takes three weeks between other work: pulling the source, agreeing what counts as a segment, deciding how to treat the returns, checking the totals against the ledger. It is careful work and it is good.
At the review, someone from Finance says: we have this. They built it last year, for a different director, and it lives somewhere nobody thought to look. The two dashboards do not agree. Not wildly: a percent here, a timing difference there. But enough that the meeting is now about which number is right, and the segment question that started all this goes unanswered for another month.
This is the elephant we call Duplicate: different teams quietly building the same dashboard, the same tool, the same logic. It is the only elephant in the herd that gets worse the more capable your people are, because capable people do not wait. They build.
The evidence
More than five working weeks a year, spent doing it again
209 hrs
a year — what the average knowledge worker spends on work that has already been done
Asana, Anatomy of Work Index.
Is duplicated work just a discipline problem?
It is usually diagnosed that way — people should check first, teams should communicate, someone should have asked. Asana's Anatomy of Work Index puts the figure at 209 hours a year per knowledge worker, which is more than five working weeks. A number that large is not a story about individuals being careless. It is a story about a system in which checking first does not work.
Consider what checking actually requires. You must suspect the thing exists, guess what the person who built it called it, know which of four tools they built it in, have access to that tool, and be able to tell from the title whether it answers your question or a superficially similar one. Any single link failing sends you back to building, which you know how to do, and which has a predictable finish.
So the honest framing is not that people fail to look. It is that looking is slower and less certain than rebuilding, and everyone is responding rationally to that. Duplication is a search problem wearing a discipline problem's clothes.
What does duplication actually cost?
Not the hours. The hours are the part you can put in a slide, and they are the smaller half. The real cost arrives the moment the second version exists, because two artifacts built independently from the same data will always diverge: different filters, a different date boundary, one that excludes internal transfers and one that forgot to.
Now the organization has two numbers for one thing. Every meeting that uses either one carries a quiet tax: someone has to establish which source is being cited before the discussion can proceed. Decisions get deferred pending reconciliation. And the reconciliation itself becomes a recurring job that nobody planned, staffed or budgeted.
There is a maintenance multiplier too. Each version needs updating when the source schema changes, when the segment definition changes, when someone leaves. Four versions of one dashboard is not four times the build cost. It is four times the build cost plus a permanent obligation, held by four people who each believe theirs is the real one.
How do you know the Duplicate is in your room?
Duplication is invisible from inside any one team. That is its defining property. Look for the seams between teams instead.
- Two reports with nearly the same name, owned by different departments, that do not quite agree.
- A meeting that regularly opens by establishing whose numbers are being used.
- A new hire who asks whether something exists and gets three different answers.
- The same transformation logic — a segment rule, a currency conversion, a fiscal calendar — written independently in several places.
- A tool that was built because the existing one was hard to find, rather than because it was inadequate.
Why doesn't a central catalogue fix it?
Because catalogues are written by the people who already know and read by almost nobody. The failure mode is well documented and utterly consistent: the catalogue is populated during a project, is accurate for about a quarter, and then decays into a list of things that may or may not still exist. Consulting it becomes another uncertain step, which is exactly the cost you were trying to remove.
The intervention that works is narrower and less satisfying to draw on a slide: make finding cheaper than rebuilding at the moment somebody wants to build. That means the answer has to arrive where the question is asked, in the tool people already open, at the moment they open it, not in a portal they must remember to visit.
It also means accepting that some duplication is correct. Two teams genuinely needing different definitions of revenue is not waste; it is two definitions. The problem is not that both exist, it is that nobody said so out loud, so each believes the other is wrong.
How do you get the Duplicate out of the room?
The same three moves we bring to any elephant, pointed at this one.
Map
Inventory by question, not by asset. Listing every dashboard produces a list nobody reads. Listing the twenty questions the business actually asks, and who answers each one where, produces the overlaps immediately. The overlaps are the finding.
Prove
Take the single worst collision, the one that costs a recurring meeting, and resolve it properly: one owner, one definition, the others retired or explicitly labelled as different and why. Do it with the people who built both. If they cannot agree on the definition, you have found something more valuable than a duplicate dashboard.
Scale
Put the answer where the question gets asked, and give the definition an owner who is expected to change it. What earns its keep is not a catalogue but a habit: the check that takes ten seconds is the one people will actually do.
Find out what your organization is building twice
The Elephant Safari names your herd in ten questions and ranks them by what they cost you. Or start from the business problem instead: the tool you need not existing, and buying one not fitting. Get the Duplicate out of the room. Build the thing that does not exist yet.
Every figure in this article traces to a primary source. See it in Knowledge