CLWM study materials
CLWM study resources

Low Adoption of a Matter-Opening Workflow

CLWM · Put your knowledge to work in context.

All case studies

Low Adoption of a Matter-Opening Workflow

Meridian Legal rolled out a new matter-opening workflow in its practice management system across 130 users in four offices over six weeks. The workflow made client identification fields, a conflict-check confirmation and an engagement-letter date mandatory before a matter number could be issued. At week six, adoption - measured as matters opened entirely through the workflow - is 28%. Partners in the two largest offices use an 'urgent matter' override that an administrator can apply, and 41% of last week's new matters used it. There are now 1,400 duplicate client records, and the monthly realisation report built from matter data is being challenged by two practice group heads. Training was a single recorded 45-minute webinar with 70% attendance. The project sponsor, the managing partner, says the system 'isn't working' and wants to revert to the old process.

1. Diagnose why adoption is low. Which explanations would you test first, and what evidence would distinguish between them?

Read the worked answer

Reframe the problem. The system is functioning; people are choosing not to use it. A 28% adoption rate with 41% override usage is a behavioural and design signal, not a technical fault. Reverting removes the data problem without solving the cause.

Test these explanations, in this order:

  1. Process fit. Do the mandatory fields match how matters actually arrive? If a partner takes an urgent injunction by phone at 8am, the workflow may be physically incompatible with that moment. Evidence: walk five recent overridden matters end to end and time how long a compliant opening takes versus the old route. If compliant opening takes 11 minutes and the matter is a genuine emergency, the override is rational.
  2. Reinforcement. The override is available, unmonitored and faster, so the environment rewards bypass. Evidence: override reasons by office, and whether any override has ever been refused.
  3. Sponsor and peer signals. The managing partner calling the system broken in public tells 130 people that non-use is acceptable. Evidence: what practice group heads say in team meetings, and whether any visible partner opens matters compliantly.
  4. Knowledge and ability. A 45-minute webinar with 70% attendance leaves roughly 39 people untrained and everyone without practice. Evidence: support ticket themes and error messages by field.
  5. Data quality loop. 1,400 duplicate records suggests the client search step is poor, so users create a new record rather than fight the search - which then corrupts the realisation report and confirms the group heads' complaint.

Use a simple framework such as ADKAR to separate awareness, desire, knowledge, ability and reinforcement. This case is heavy on ability and reinforcement, light on awareness. Train more, and you will fix the smallest cause.

2. Design the intervention plan for the next 90 days. Cover stakeholder handling, the override, the duplicate records and the reporting challenge.

Read the worked answer

Fix the design before pushing adoption. Recruit six to eight users, including one partner from each large office, to open five real matters - including one genuinely urgent matter - through the workflow while you watch. Record every friction point, rank by minutes lost, and fix the top five. If urgent matters genuinely cannot comply, design an urgent variant with the minimum data required at opening and the rest completed within 24 hours.

Govern the override rather than abolishing it. Some safety valve is necessary or people will invent a worse workaround. Restrict it to defined categories, make it time-limited with automatic expiry, require a reason code and partner sign-off, and publish the override rate by office weekly. The aim is visibility and shrinking volume, not a zero target in month one.

Clean the data and fix the cause. Deduplicate the 1,400 records with a published rule - match on legal name plus identifier plus date of birth or company number - merge with an audit trail, and make the client search forgiving before blocking new record creation. Freeze the creation of new client records without a search step.

Rebuild trust in the report. Sit with the two practice group heads and finance to reconcile the realisation report against the billing system line by line. Publish the definition, the source and the known limitations. A report that cannot be reconciled should not be circulated monthly; fix it or label it provisional.

Stakeholder plan. Segment: partners in the two large offices (key blockers), practice group heads (data credibility), secretaries and administrators (highest volume of actual opening), fee earners. Give each a different message: partners get time saved and fewer billing queries, administrators get a shorter form and fewer duplicate emails, heads get a report they can defend.

Training. Replace the webinar with 20-minute role-based sessions in each office, a one-page job aid at the point of use, and floor-walkers for two weeks after each go-live.

3. What would you measure over the 90 days to know whether this is working, and how would you reduce the risk of the metrics being gamed?

Read the worked answer

Outcome and behaviour metrics, reported weekly by office:

  • Adoption: percentage of matters opened entirely through the workflow. Report volume as well as percentage, since a falling matter count can flatter a percentage.
  • Override rate and reason mix by office and by partner. Watch for the concentration of 'urgent' reasons in the same practice group.
  • Duplicate record rate on new records created, plus the closing balance of the 1,400 backlog.
  • Matter-opening cycle time, median and 90th percentile, from instruction to matter number issued.
  • Completeness: percentage of new matters with all mandatory fields populated and no later manual correction.
  • Reconciliation variance between the realisation report and the billing system, published as a single number.
  • Support tickets per 100 openings, and a two-question user pulse monthly.

Reducing gaming:

  • Audit a sample of 20 matters monthly, tracing the data from the workflow into the report. If people open compliantly and then edit fields afterwards, the completeness metric will not catch it; the reconciliation check will.
  • Cap and review overrides rather than banning them, so urgency is documented and comparable across offices.
  • Do not set individual adoption targets; they invite dummy matters. Set office-level adoption plus a data quality measure, so speed cannot be bought by cutting corners.
  • Publish the same numbers to everyone. Shared visibility with a named owner per office is more effective than private dashboards.

Decision point. Agree at the outset what result at day 90 means continue, adjust or stop, and what specific evidence would justify reverting. Committing to that in advance stops the sponsor from pulling the plug on a bad week and gives the team a fair test.