Most reporting transformations are planned around the destination. The licences are bought, the design looks good in the demo and the timeline is agreed. Then the team starts on the existing reports and finds that nobody knows how many there are.

These are the questions to have answered before you commit budget and scope. None is hard on its own. The trouble is that most programmes answer them halfway through, when every answer means rework and a revised plan.

1. Scope: what actually exists?

Ask your team for a count they can defend. Not the list on the wiki, but what is really in use across shared drives, SharePoint, report servers and Power BI today.

  • How many reports are there, and of what kind?
  • How many are copies of each other, saved under a new name or date?
  • How many has nobody opened or refreshed in the past year?

If the answer is 'around two hundred, we think', that is your first risk. Cost, effort and timeline all hang off this number.

2. Accountability: who answers for each report?

Every report that carries forward needs someone who can say whether the new version is right. Many have an author who has left, a requester who has moved on and users who assume someone else owns it.

  • Is there a named business owner for each report, not just the last person to edit it?
  • Which reports depend on one person who knows how to run them?
  • Who will sign off that the new version is correct?

Make it a rule: a report without an owner does not carry forward by default. It goes on a list for someone to claim or retire.

3. Duplication: how many versions of the same report are there?

Duplicates are rarely identical. A team copies the board pack, adds a tab, changes a filter, and now two versions agree most of the time. Carrying both forward doubles the cost and keeps the confusion.

  • Which reports serve the same audience with the same measures?
  • Where two versions differ, which one do people actually use?
  • Who decides which version survives?

4. Definitions: is each KPI calculated once?

This causes the most pain after go-live. If Gross Margin is calculated three ways today, the new reporting has to pick one, and someone will notice their number has changed, usually in front of the board.

  • Which KPIs appear in more than one report, and are they calculated the same way?
  • Where are rates or thresholds fixed by hand rather than taken from the source?
  • Who has the authority to agree each disputed definition?

Settle these before the rebuild, not during user testing. A definitions argument late in the programme can stall a whole release.

5. Knock-on effects: what relies on what?

Reports feed each other. A consolidation workbook draws on an extract, the board pack draws on the consolidation and the forecast draws on the board pack. Change one piece without the others and something downstream quietly stops updating.

  • Which reports feed other reports, and which feed the board?
  • What will stop working when the old systems are switched off?
  • Where can the team not confirm what a report relies on, and who can?

Ask for the gaps, not just the findings. Something the team cannot confirm is not the same as something that does not exist.

6. Manual work: where do people touch the numbers?

Many reports only work because someone pastes in a figure each period, saves a file from an email or updates a table by hand. These steps are invisible in a report list, and they are the first to fail when the process changes.

  • Which reports rely on manual steps or emailed files?
  • Which of those steps will the new reporting remove, and which need a new process?
  • Do the people doing them know the programme affects them?

7. Retirement: what are you not carrying forward?

The cheapest report to modernise is the one you switch off. Most estates carry reports nobody reads, built for a project that ended or a manager who left. Decide what retires before the estimate, not after.

  • Which reports are stale, archived or superseded?
  • Which have no owner willing to claim them?
  • How will users be told, and how long will the old copy be kept?

8. Sign-off: how will you know it is done?

Define done before you start. For each report in scope, agree who signs it off, what they compare it against and what difference is acceptable.

  • Who signs off each report or group of reports?
  • What must the new output match, and over which periods?
  • Where are decisions recorded so they survive staff changes?

Have the answers before you approve the estimate

None of this needs the new platform. It needs a clear view of what you already have. Programmes that answer these questions first are scoped on evidence, and the sponsor can see where the risk sits.

A plan built on a guess about the estate is a plan to discover the estate the hard way, on the programme's budget.