They cannot find the entry. The mini program name is internal jargon. The APP wants a project code in the store. Official-account menus point at expired H5. Fix “can open” first. See What to do when nobody downloads the APP, Official-account menus without a mini program.
Two weeks live, DAU near zero, the first reaction is often “we wasted it.” Do not tear it down yet. The floor unused is rarely “people hate new things.” The path broke on a layer: cannot find the entry, login is too heavy, cannot submit on site, paper still runs in parallel, or nobody reads what was submitted. Split those layers. That beats another splash animation. The field gap with and without an entry. Compare Field work with a mobile entry versus without.
Login is too heavy. Badge plus a hard password plus a captcha: people in gloves quit. If WeCom SSO or an SMS code works, do not make the floor memorize a third password.
They cannot submit on site. No network in the basement, photos fail, scan misses. Whether you need offline is in No network on site—do you need offline. If you never walked a weak-network device, launch was not launch.
Paper is still accepted. Leaders ignore system submits and only count paper. One day of dual track and the system is extra work. What belongs on the employee app is in Employee features people will actually use.
Submitted, nobody catches it. Documents land in a back office nobody opens, or the logic is not the same as the management system. Multi-end must share APIs. See Why one back office keeps multi-end data from forking.
Another case: the features are not the work the floor must finish. Home is culture, news and a points mall. Reporting sits in a fourth-level menu. Brochure versus work tool is in Brochure mini program versus a work tool.
Split “cannot open” from “opened and still uses paper”
Without open logs every discussion is a guess. Watch a week: how many opened, share stuck on login, successful submits versus paper volume. Many opens and few submits: check the form and the network. Near-zero opens: put the entry in a container they already use—WeCom, store codes, badge QR. A template mini program with many gray buttons also looks “unused.” See A template mini program is almost as bad as none.
Management action beats another build: some document types only count as system submits. Paper is emergency only. If you do not cut the dual track, every client sits idle. When DaXi takes a “launched unused” project we walk those five steps on site. We do not redraw UI first. To look at production together, go to the Apps & Mini Programs and say who is required to use it and what they still use.
Bring a comparison sheet: paper volume, group photos, system submits, split by crew. One crew at zero and the next normal: check accounts and network first, not “frontline quality.” Everyone at zero: look at the entry and whether paper is still accepted. If a customer mini program is unused, also check whether the code is on the first screen and whether redemption needs a three-step clerk chant. See What belongs on a member mini program home.
Ask the floor in their own words. Closer than a crash rate
Ask “why did you not use it last week.” Answers are specific: SMS never arrived, camera permission off, ticket numbers do not match. Write the words down. A busy boss console and a zero floor means you built a management board, not a work entry. You need both, but phase one must let the floor submit first. If an old APP’s structure is already wrong, evaluate Rebuild the old APP, or extend it. Do not keep pasting features onto a wrong IA.
Launch-week meetings should not only score design satisfaction. Ask sales or a lead “which entry do you forward.” If they cannot answer, it is not daily work yet.