Converting the map into a short list of fixes worth doing

4 days ago
3 min read

Updated: 2 days ago

Introduction

The map is finished, everybody agrees it was useful, and the exercise ends there. That is the normal outcome and it is a waste, because a map is a diagnostic instrument rather than a deliverable. Its entire value is in what somebody does differently afterwards, and the conversion from description to decision is a separate piece of work that nobody schedules.

It is also short. An hour, with the map and a blank page, produces three or four candidate fixes, and choosing between them is a matter of ranking rather than analysis. The difficulty is not the thinking; it is that the meeting to do it is never in anybody's diary.

1. Converting the map into a short list of fixes starts with the losses

The source of candidates.

Every point where the map shows people leaving is a candidate. Ignore the parts that are working, however much effort went into documenting them. Effort spent describing something is not a reason to work on it.

2. Add the long gaps

The second source.

Stages where elapsed time is high, even if few people are lost there. Delay produces losses that appear elsewhere and it is frequently the cheapest thing to change. Long gaps are also easier to shorten than conversion rates are to raise.

3. Add the questions you cannot answer

The third source.

Anywhere the map says you do not know. Those are measurement fixes rather than process ones, and they make the next map better. Recording a source or a date is usually the whole fix.

4. Rank by volume affected

The first sorting criterion.

A fix at a stage everybody passes through beats one at a stage a tenth reach. This ordering is arithmetic rather than judgement. Multiply the volume by the proportion lost and the order becomes obvious.

5. Then rank by cost to fix

The second criterion.

Some fixes are a sentence on a page and some are a hiring decision. A cheap fix at a high-volume stage is where you start, every time. There is usually at least one of these on any first map.

6. Pick one

The discipline.

Not four. One fix, completed, measured. Businesses that select four complete none, and the map is then judged to have produced nothing. The restraint is the hardest part of the hour.

7. Name an owner and a date

The mechanism.

A fix with neither does not happen, however obviously correct it is. This is the same discipline as any other change and it is skipped more often here because the exercise felt like the work. The map was the diagnosis, not the treatment.

8. Decide how you will know it worked

The measurement.

Which number should move, over what period. Agreed before the change, or the assessment afterwards becomes a matter of impression. Record the current value on the day you start.

9. Come back to the list

The continuation.

The remaining candidates do not expire. Once the first is done and measured, the second is already identified, which makes the second round much faster than the first.

Be careful about the fix that is most interesting rather than most valuable. Redesigning something is more appealing than shortening a response time, and the appeal is a poor guide to what the map showed.

Conclusion

Spend an hour turning the map into one change with an owner and a date.

Take your candidates from where the map shows losses, from stages with long elapsed times, and from the places where the map admits that you do not yet know what is happening, rank first by how many buyers are affected and then by what the fix costs, choose one rather than four, assign an owner and a date, agree in advance which number should move, and return to the remaining candidates once the first is measured.