Moving a client off spreadsheets without losing the plan

Spreadsheets are a reasonable place to start a strategic plan and a bad place to keep one. The question is not whether they are adequate in principle — it is whether you are already paying the cost of them in labour.

The four signals it is time

1. History is being asked for. A board wants an eight-quarter trend and nobody can produce it, because the cells were overwritten each period.
2. Two versions are circulating. Someone emailed a copy, someone edited the copy, and the reconciliation is now a task.
3. Assembling the pack takes more than a day. Every quarter, by hand.
4. Owners cannot update their own items. One person becomes the transcription service for everyone else, and updates arrive late or not at all.

None of these? The spreadsheet is fine. Two or more? The cost is already being paid; migrating just stops paying it.

What actually needs to move

Less than people expect:

• Goals, with one owner each.
• Initiatives and actions, with owners and dates.
• Measures with definitions, baselines and targets.
• Any history worth keeping — usually the last four to eight periods per measure, not everything.
• Open risks.

Deliberately leave behind: dead initiatives, measures nobody has updated in a year, and formatting.

A one-pass migration

Step 1 — import the plan you already have. Import the existing document or spreadsheet directly rather than retyping it. A plan that comes in as structured goals, initiatives, actions and measures gives you something to correct instead of something to build.

Step 2 — fix ownership first. Every goal, initiative and action gets exactly one name. This is where most of the value appears, and it is a 30-minute conversation, not a data task.

Step 3 — repair the measure definitions. For each measure: formula in words, exclusions, unit and direction, source system, update frequency, baseline with a date, target with a date. Anything that fails this test either gets defined or gets dropped.

Step 4 — load only the history you will cite. Four to eight periods per measure covers almost every board question. Loading five years of a measure whose definition changed twice creates a trend nobody can defend.

Step 5 — set the review dates. Put the next four quarterly reviews in calendars before you finish. A migration that ends without dates produces a tidier version of the same drift.

Step 6 — let owners update their own items. Give each owner access and have them make one real update in the first week. The migration is complete when the person responsible for a number is the person who enters it.

Keep the old file for one quarter

Read-only, clearly marked superseded, with the date. It resolves the inevitable "the spreadsheet said something different" question and costs nothing. After one full review cycle, archive it.

Setting client expectations

Half a day of your time and about two hours of the client's, spread over a week — not a three-week data project. If a migration is being scoped in weeks, too much history is being moved. Trim it.

The measure of success

Ninety days later, ask one question: can the sponsor see the current status of every goal without asking anyone? If yes, the migration worked, regardless of how the reports look. If no, the plan moved but the practice did not, and the cause is almost always that owners are still not updating their own items.