Done is not the same as fixed
Action tracking in customer experience usually stops at "done": the extra cashier was hired, the policy was reprinted, the app was updated. Whether customers noticed is rarely checked. Tracking to outcome closes that gap, and it is what turns a list of tasks into evidence that the CX programme works.
What to record for each action
- The issue, in customers' words, with the count that prompted it.
- The owner and their team.
- Where it applies: locations, regions, app version.
- The start date.
- The measure that should move, such as negative share for billing waits.
- The check date.
- The outcome: improved, no change, worse.
Set the baseline first
Before the change, record the measure for the affected locations and for similar locations that are not changing. Without a baseline, any later number is just a number.
Did the fix work? Measuring change after an action.
Did the fix work measuring change after an actionChecking the outcome
- Wait long enough for your review volume, often four to eight weeks.
- Compare changed locations with their own baseline and with unchanged peers.
- Read reviews from after the change, not just counts.
- Record the result and a sentence on why.
Free review snapshot
Want this done for your own reviews?
No commitment. We use your details only to send your snapshot and follow up. Privacy policy
Statuses that mean something
- Planned.
- In progress.
- Done, awaiting check.
- Checked: improved.
- Checked: no change.
- Stopped.
The "awaiting check" status is the one most programmes are missing. It keeps finished tasks visible until someone looks at the result.
Reviewing the log
Go through the log in the monthly action review: actions awaiting check, results, and anything overdue. Report to leadership how many actions were checked and how many improved the measure.
Governance: a monthly CX action review.
Governance a monthly cx action reviewWhen the outcome does not move
Check the measure, the cause, the rollout and the timing before giving up, and keep the record either way.
When a fix doesn't move the numbers.
An example
A restaurant chain logged an action to add a second cashier at weekend lunch in twelve outlets, with billing-wait negative share as the measure. Six weeks later the share had fallen in nine outlets and not in three. Store visits showed the three had not staffed the counter after 2 pm. The log kept the action open until they did.
Common mistakes
- Closing actions when tasks finish.
- No baseline.
- No comparison locations.
- Deleting actions that did not work.
Common questions
Which tool should we use?
A shared tracker is enough to start; what matters is the fields and the check.
Who owns the log?
The CX team keeps it; each action is owned by the team making the change.
How many actions at once?
Few enough that each gets checked.
Should leadership see it?
A monthly summary: checked, improved, no change.
Checklist
- Issue, owner, scope and measure recorded.
- Baseline taken.
- Check date set.
- Outcome recorded.
- Failures kept.
In short
Record every CX action with its issue, owner, scope and measure, take a baseline, keep it open until the outcome is checked, and keep the actions that failed.
Starting the log
Begin with the actions already in progress: add a measure and a check date to each, and take a baseline today.
Doing this with Falcon
In Falcon, recurring complaints are clustered into specific issues, owners are suggested from your taxonomy, and reviews can be assigned and tracked as tickets until they are done. Negative share by topic, location and month shows whether an issue fell where a change was made, with the reviews behind every number.