The 5-Second Rule for Data: If You Can't Explain It, You Don't Own It

Gohvio · 2026-06-03

The fastest way to find broken data.

Walk into any company. Pick any dashboard. Point at a number. Ask the person who owns the dashboard: where does this come from?

If they can answer in under five seconds -- "that's pulled from Salesforce, refreshed nightly, owned by Maya in Sales Ops" -- the data is probably fine.

If they pause, look at the dashboard, click on the cell, scroll through a settings panel, and then say "I'd have to check," you've found the problem. The data has no real owner.

Unowned data always breaks. Eventually. Quietly. Usually right before a board meeting.

What ownership actually means.

Ownership of data is not the same as access to data. It's not the same as building the dashboard. It's not the same as being the person who sends the weekly report.

Ownership means four things:

You can name the source system the data comes from.

You can name the refresh cadence and confirm the last refresh succeeded.

You can name the definition of what the metric measures -- where it starts, where it stops, what it excludes.

You can name the next person in the chain who will know if the data is wrong before it breaks.

If any of those four are fuzzy, the metric is unowned. It is being maintained by inertia. It will be wrong before the year is out.

Why this matters more than people admit.

Most data quality problems get blamed on "the system." The CRM had a sync issue. The BI tool didn't refresh. The integration broke.

That framing is comfortable, but it's wrong. Systems fail constantly. The question is whether anyone notices.

In a healthy data environment, when a metric goes weird, the owner notices the same day. They know what the number should look like. They know the input sources well enough to spot a missing import or a stale refresh. They escalate.

In an unhealthy data environment, the metric sits broken on a dashboard for weeks. It gets used in a board deck. Someone eventually notices it doesn't match the other dashboard. A reconciliation project starts. The original owner -- if one even exists -- finds out via the post-mortem.

The difference is ownership. Not tooling.

Why so many metrics end up unowned.

Three patterns cause unowned data.

First: the original owner left. They built the dashboard, knew every quirk, then went to a new job. Nobody inherited the deep knowledge, just the dashboard URL.

Second: the metric was built for a project that ended. Someone needed a custom view for the 2024 board meeting. It got built. The board meeting happened. Nobody officially retired the metric, so it kept refreshing, kept getting referenced, kept drifting from any current definition.

Third: the metric was always communal. Marketing, Sales, and Finance all use it. Each team assumes one of the other teams owns it. None of them do. When it breaks, everyone points sideways.

How to find unowned data this week.

Pick your top ten most-referenced metrics. The ones that show up in QBRs, board decks, and weekly leadership reports.

For each one, ask:

• Who is the named owner? (Not the team. The person.)
• Where does the source data come from?
• How often does it refresh?
• When did the definition last get reviewed?
• Who gets pinged if it breaks?

Run that exercise. Most companies discover that two or three of their most-quoted metrics have no clean answer to those questions. Those are your unowned metrics. They are time bombs.

What to do once you've found them.

For each unowned metric, do one of three things.

Assign an owner. Document the four items above. Put it in writing. Tell the person they own it. Make sure they have the access and visibility to actually notice when something breaks.

Retire it. If nobody needs the metric anymore, kill it. A retired metric can't lie to you. A zombie metric can.

Consolidate it. If two teams maintain similar metrics with different definitions, pick one. Standardize. Document. Assign the owner.

That's it. There's no software purchase. There's no transformation program. Just a list, a set of questions, and the discipline to answer them honestly.

The benefit when it's done.

Companies that take ownership seriously report two things.

First, far fewer surprises. When data goes weird, somebody notices it within hours, not weeks. The cost of a data error is roughly proportional to how long it sits unfixed.

Second, faster decisions. Leadership stops second-guessing dashboards. The number on the screen is the number, because the owner has been on top of it. The team can argue about what the number means, instead of arguing about whether the number is real.

The bottom line.

The 5-second rule isn't a productivity hack. It's a stress test for trust.

If the person responsible for a number can't explain its source in five seconds, then nobody is really watching it. It will drift. It will break. It will show up wrong in the worst possible meeting.

Find your unowned data this quarter. Assign or retire every metric. Document the rest. The cost is one afternoon. The savings will compound for years.

Want to find every data gap in your CRM in one upload? Sign up for your free trial today -- free.