A mobile system for 5G site operations, stuck at 30% adoption. The fix wasn't more features — it was going to where the work actually happened.一个 5G 站点运维的手机系统，采用率卡在 30%。解法不是加功能——是去到工作真正发生的地方。
We'd built a mobile system for 5G site operations — approvals, safety checks, the works. It was, on paper, a good tool. And yet adoption had stalled below 30%. Field teams just weren't using it. The instinct in the room was to add more features. I didn't think the problem was features.我们给 5G 站点运维做了个手机系统——审批、安全检查，一整套。纸面上，它是个好工具。可采用率却卡在 30% 以下。一线团队就是不用它。当时屋里的直觉是再加功能。但我不认为问题出在功能上。
Not "more adoption" as a number on a slide — but a tool the field would reach for because it fit how they actually worked. If they weren't using it, the tool was wrong, not the people.不是 PPT 上"采用率更高"那个数字——而是一个一线愿意主动去用的工具，因为它贴合他们真实的干活方式。如果他们不用，那是工具错了，不是人错了。
None of it was a missing feature — it was a reality gap. So we rebuilt around those three constraints: offline-friendly, far fewer fields, webhook-driven mobile approvals, and automated safety-credential checks before a permit could issue. I also gave management a cross-party SLA dashboard that ended the blame games between vendor, operator and TowerCo. Adoption went to 100%, approvals dropped from days to under a day.没有一个是缺功能——是现实的缺口。所以我们围着那三个约束重做：离线可用、字段大幅精简、webhook 驱动的手机审批、出许可前自动校验安全证件。我还给管理层做了个跨方 SLA 看板，结束了供应商、运营商和铁塔公司之间的甩锅。采用率涨到 100%，审批从好几天降到不足一天。
A tool nobody uses is a failed project, no matter how good it looks in a demo. The answer usually isn't in the backlog — it's on site, in the gap between how a system was designed and how the work really happens. I don't trust an adoption number until I've stood where the work is done.一个没人用的工具就是失败的项目，不管 demo 里多好看。答案通常不在需求列表里——在现场，在"系统被设计的样子"和"活儿真实发生的样子"之间那道缝里。没站到干活的地方，我不相信任何一个采用率数字。