completion-diagnostics

Completion Diagnostics

Diagnose why training programs have low completion or engagement, and recommend targeted fixes. This skill combines Docebo data analysis with L&D diagnostic frameworks.

Diagnostic Process

When a user reports low completion or engagement, follow this structured approach. Don't jump to solutions before understanding the problem.

Step 1: Gather Data

Pull the relevant data from Docebo before diagnosing:

• Use get_team_training_report to see team-level completion patterns
• Use list_enrollments filtered by course to see the full enrollment picture
• Use get_user_progress for individual learner patterns
• Use get_course to understand the course structure (type, duration, settings)

Key data points to collect:

• Completion rate — what percentage finished?
• Start rate — of those enrolled, how many even began?
• Drop-off pattern — is there a specific point where people stop?
• Time-to-complete — are completers finishing quickly or dragging out?
• Enrollment method — were people auto-enrolled, manager-enrolled, or self-enrolled?
• Course type — e-learning, ILT, blended, learning path?

Step 2: Compare Against Benchmarks

Interpret the data against training-type benchmarks:

Training Type	Target Completion	Red Flag Below
Compliance/mandatory	90-95%	85%
Onboarding (first 90 days)	80-90%	70%
Voluntary/professional dev	20-40%	10%
Certification programs	60-80%	50%
Manager-assigned development	50-70%	35%

Important: Voluntary training at 25% completion is normal — don't treat it as a crisis. Compliance training at 75% IS a crisis.

Step 3: Identify the Root Cause

Low completion has different root causes at different stages. Diagnose which stage is broken:

Low start rate (enrolled but never began) Causes: learners don't know they're enrolled, enrollment feels irrelevant, no urgency/deadline, bad first impression (confusing title or description), technical barriers (LMS access issues).

Diagnostic questions: When were they enrolled? Did they receive a notification? Is the course title/description clear about what they'll learn? Is there a deadline? Can they access the LMS easily?

Early dropout (started but quit within first 10%) Causes: content doesn't match expectations, poor opening (long intro, no hook), technical issues with content playback, immediate realization it's not relevant.

Diagnostic questions: What does the first module look like? Is there a compelling opening that explains the value? Does the content match the course description?

Mid-course dropout (engaged initially, then stopped) Causes: content too long, repetitive sections, difficulty spike, competing priorities pulled them away, no check-ins or reminders.

Diagnostic questions: How long is the total program? Where exactly do people drop off? Is there a particularly difficult or boring section? Are there reminders for people who stall?

Near-finish dropout (completed most but didn't finish) Causes: final assessment is intimidating, last section is perceived as low value, administrative friction (need to submit something, schedule a session), life happened and they forgot.

Diagnostic questions: What's the final step? Is there a high-stakes exam at the end? Does the last module feel like busywork?

Step 4: Recommend Interventions

Match interventions to the root cause. Don't recommend generic "make it more engaging" — be specific.

For low start rates:

• Send targeted launch communications explaining the "why" (not just "you're enrolled in X")
• Set clear deadlines with automated reminders at 30/14/7/3 days
• Ask managers to mention the training in their next 1:1
• Improve the course title and description to be benefit-focused
• Reduce enrollment-to-start friction (deep links directly into the first module)

For early dropout:

• Redesign the opening — lead with a scenario or problem, not a table of contents
• Add a "what you'll learn and why it matters" section in the first 2 minutes
• Check for technical issues (broken videos, slow loading, mobile compatibility)
• If the audience is mixed experience, offer a pre-assessment to skip basics

For mid-course dropout:

• Shorten the program — cut anything that's nice-to-know vs. need-to-know
• Break long modules into shorter segments (max 15-20 minutes each)
• Add knowledge checks for retrieval practice and variety
• Implement progress nudges ("You're 60% done — keep going!")
• Create cohort deadlines (complete by Friday) for social accountability

For near-finish dropout:

• Simplify the final step — remove unnecessary administrative friction
• If there's a final exam, offer a practice quiz first to reduce anxiety
• Send a personalized nudge ("You're 90% done — just one more module!")
• Make the completion feel rewarding (certificate, badge, manager notification)

For system-wide low engagement:

• Audit the training catalog for stale content (anything not updated in 18+ months)
• Balance mandatory and voluntary training (if everything is mandatory, nothing feels valuable)
• Invest in manager enablement — managers are the #1 lever for training adoption
• Build a communication plan — most L&D teams under-invest in marketing their programs

Step 5: Measure the Fix

After implementing interventions, track:

• Did the specific metric improve? (start rate, drop-off point, completion)
• Did we introduce any new problems? (e.g., shortened content but assessment pass rates dropped)
• Is the improvement sustained or was it a one-time bump from the communication push?

Set a review checkpoint 30 and 90 days after the intervention.

Quick Diagnosis Cheat Sheet

Symptom	Most Likely Cause	First Intervention
High enrollment, low starts	Awareness/communication gap	Targeted launch email + manager nudge
Starts but drops in first module	Poor opening or mismatched expectations	Redesign first 5 minutes
Steady dropoff throughout	Content too long	Cut 30% of content, break into smaller modules
Drops at a specific module	That module has a problem	Review that specific module for issues
Almost finishes but doesn't	Final step friction	Simplify completion requirements
Voluntary course, <10% completion	Content doesn't feel valuable	User research — ask why they didn't finish
Compliance course, <85% completion	No enforcement mechanism	Add deadlines, manager visibility, escalation