Short answer: Measure first, then act: if fewer than 60% of intended users log in weekly within 30 days of launch, intervene. Most causes aren't code but unchanged processes, old methods still running and too little training. An "adoption health check" finds the blockers, and adoption usually climbs above 80% within 4–6 weeks.
A 60-person building-materials wholesaler spent NT$450,000 on an order and inventory system. A month after launch, sales still send orders to an assistant over LINE for re-keying, and the warehouse still ships from paper. Only 11 of 30 accounts log in weekly, and the owner starts doubting the vendor.
Myths busted: low usage doesn't automatically mean a bad system
Most "nobody uses it" systems fail in the rollout, not the build.
- Myth: a good system gets used naturally. Reality: as long as the old way still works, most people won't switch; you need a firm cutoff date.
- Myth: one training session is enough. Reality: the change-management ADKAR model separates knowing how from being able to do it; a single 2-hour lecture is mostly forgotten in two weeks.
- Myth: lots of complaints means features must change. Reality: complaints often reflect resistance to process change; separate "truly hard to use" from "not used to it."
The core framework: three adoption health metrics
Adoption = people who log in and complete the key action each week ÷ people who should use it.
- Login rate: weekly logins ÷ intended users; target 60% by day 30 and 80% by day 60.
- Key-action completion: for example, the share of orders created by sales themselves; target 70% by day 60.
- Shadow processes: the number of workflows still handled in Excel, LINE or paper; target zero by day 90.
All three low means a rollout problem; high logins but low key actions means design blockers; many shadow processes means the old way never stopped.
Three scenarios: different sizes and maturity need different fixes
Small companies lack resolve; large ones lack owners.
| Scenario | Common blocker | Fix first |
|---|---|---|
| 15-person studio, first system | The owner doesn't use it | Owner leads and reviews only system reports |
| 60-person wholesaler moving from Excel | Sales still order via LINE | Set an Excel cutoff; assistants stop re-keying |
| 200-person manufacturer, cross-department | No one owns cross-department flows | Appoint one champion per department |
Hidden costs: low adoption burns money every month
An unused system often costs more than its development.
- Double entry: re-keying LINE orders takes 1–2 hours a day, about 250–500 hours a year.
- Incomplete data: reports are unreliable, so decisions go back to Excel.
- Training: 3–6 hours per employee, about 90–180 hours for 30 people.
- Maintenance still billed: monthly fees continue while value doesn't arrive.
KPI scorecard: rating your vendor's rollout support
Score 10 dimensions 1–3 each; 24+ means the vendor met its rollout responsibilities.
- User testing before launch
- Role-based user guides
- On-site or live support in launch week
- Adoption reports provided
- Issues answered within 1 working day
- Distinguishes bugs from usage questions
- Proactively suggests process adjustments
- Small changes delivered within a week
- 30/60-day review meetings
- Docs updated with each release
How ScriptWalker handles it, and who we're not a fit for
We treat rollout as part of the project and run user tests with at least 5 people before launch; Nielsen Norman Group research shows 5 users uncover most usability problems.
- Engagement models: Project for build and rollout, Retainer for post-launch tuning, Advisory for process diagnosis; no full outsourcing with on-site staff.
- Starting prices: custom system development from NT$80,000, maintenance from NT$2,000/month.
- Not a fit: owners who won't change processes and just want "Excel on the web."
- Not a fit: no internal contact to join testing and training.
- Not a fit: demands for a same-day company-wide switch with no pilot.
Rescue playbook: 80% adoption in 6 weeks
Stop the old way first, then add training.
- Week 1: run adoption reports and interview the 5 least active users.
- Week 2: sort issues into bugs, design blockers and habits; send the first two to the vendor.
- Weeks 3–4: 30-minute role-based sessions; appoint champions.
- Week 5: announce the cutoff for Excel and LINE orders; the owner reviews only system reports.
- Week 6: recheck the 3 metrics and repeat for lagging departments.
Decision checklist: 5 or more ticks means fix the rollout before the features
The problem is likely the rollout, not the code.
- ☐ Weekly login rate below 60%
- ☐ Some people still use Excel or LINE for the same task
- ☐ No cutoff date for the old method
- ☐ Training held only once
- ☐ The owner rarely opens the system
- ☐ No champion users per department
- ☐ Complaints are mostly "not used to it"
- ☐ The vendor provides no adoption reports
- ☐ No user testing before launch
- ☐ No 30-day review meeting
FAQ
These are the questions owners ask most; the first sentence of each answer is the answer.
What adoption rate is normal?
Around 60% weekly logins at day 30 and 80% at day 60 are reasonable targets. Below that, find out why.
Is low adoption the vendor's fault?
Not necessarily. Separate bugs, design blockers and habit problems; the first two are mainly the vendor's responsibility.
How long should Excel run alongside the new system?
About 4–8 weeks covering one month-end close, then set a cutoff date to avoid permanent double entry.
What training works best?
Role-based groups, 30-minute sessions repeated several times, plus a champion user in each department.
Next step: run an adoption health check
System launched but nobody uses it? ScriptWalker (a Taiwan-based Laravel/Flutter custom development studio) can run adoption reports and user interviews to tell you whether to change the code or the rollout.
- Email: [email protected]
- Phone: 0916-224-047
- LINE: @ufv9089p