Adoption is a design and management problem, not a training problem. Remove the parallel path first: if leave approvals still work over chat, chat wins. Make the task a manager does most often the fastest thing in the tool, teach it against their real work, give every process a named owner, and read usage data as a diagnostic.
As long as an approval can be obtained by messaging HR directly, that is the route managers will take, because it is faster for them and the cost lands on somebody else. Adoption work that leaves the old path open is asking people to choose the slower option out of goodwill. Close it deliberately and announce it: requests raised outside the system come back with a link rather than getting processed. Then make certain the system genuinely handles the case, since bouncing a manager back to a workflow that cannot approve a half-day of leave loses the argument permanently. [Leave management](/leave-management-software) is usually where this gets tested first, because it is the request managers face most often.
The one a manager touches most often, which is rarely the one the project plan is proudest of. For most managers it is approving something: leave, an attendance correction, an expense, a timesheet. Make that path fast enough to complete straight from a notification without navigating anywhere, and the tool earns its place before anyone is asked to use it for something harder like a review cycle. Optimising the rare, complex task first is a common sequencing mistake, because managers form their opinion of the system during the frequent trivial one, and that opinion is what they bring to everything you ask of them later.
Because it teaches the menu rather than the job. A manager sitting through a walkthrough of every module remembers none of it a fortnight later, when they finally need to approve something. Train against the tasks they actually perform, in their own data, in short sessions timed close to when they will first need it. A one-page reference for each frequent task is worth more than a recorded session nobody reopens. Where a manager has a team, let them learn by completing one real transaction with somebody watching, rather than by observing a demonstration account that belongs to nobody.
Every process gets a named owner accountable for it running, and the manager's part is stated plainly: approvals within a working day, attendance corrections closed before the payroll cut-off, records kept current. Make the pending queue visible to the manager and to their own manager, and let escalation happen there instead of through HR chasing individuals. What to avoid is turning usage into a scoreboard, since that produces compliance behaviour such as approving in bulk without reading, which is worse than the delay you set out to fix. Aim for it being obvious what is outstanding, not for anybody being ranked against a peer.
As a diagnostic, not as evidence for a conversation about attitude. A manager who has approved nothing in a month may have notifications routing to a folder they never open, may have had no delegate configured while on leave, or may be hitting a screen that does not work on the device they carry. Go and ask before concluding anything. Patterns matter more than individuals: if a whole function avoids one workflow, the workflow is the problem. Where an [HR system](/hr-software) reports on pending items and completion, use it to locate friction, and expect some of what you find to be a design fault of your own making.
A large one, because managers get pulled in by their teams. If employees apply for leave, check balances, download payslips and update their own details through a [self-service portal](/employee-self-service-portal), the manager receives requests there and answering them in the system becomes the path of least resistance. If employees still message their manager on chat, the manager becomes a relay and the system stays empty. Employee adoption is usually easier to win, since their alternative is asking somebody and waiting, so it is worth sequencing the employee rollout first and letting demand pull managers along behind it.
Being the workaround. Every time HR re-keys a request that arrived by email, corrects a record on a manager's behalf, or accepts a spreadsheet instead of a submission, the parallel path is confirmed as viable and the next rollout gets harder. Stopping is uncomfortable for a week and decisive afterwards. Two exceptions are worth holding: a genuine system fault, which HR should get fixed rather than absorb, and a manager with a real accessibility need, which deserves an accommodation rather than a workaround. Everything else goes back with a link and an offer of help, which is a very different message from doing it for them.
Pitch N Hire is an applicant tracking system built for recruiters and hiring teams. If this answer described something you want to run properly, the ATS is where it lives.
Free for 1 user · No credit card · Talk to a real hiring expert
Get a personalized walkthrough of Pitch N Hire on your own roles and workflow. No slides, no obligation.
Prefer to talk? Book a demo · Talk to sales · View pricing
Free 1-user plan · No credit card · Talk to a real hiring expert
See your true cost-per-hire and how much Pitch N Hire could save you — our free Recruitment ROI Calculator gives you the numbers in under a minute. No signup required.
Open the free ROI calculatorPrefer a tailored walkthrough on your real roles? Drop your work email:
★ Free 1-user plan · No spam · Talk to a real hiring expert
