Your product is your people's time. The system either measures it or it does not.
Project kickoff and field HR-systems training across professional services firms, and the same tension every time: a system that has to capture time accurately from people whose incentive is to spend as little time as possible capturing it. Walk the five decisions, find your seat, then run the two-minute check.
Five calls that decide a professional services program
Made in design, felt in the first billing cycle after go-live. Pick a decision.
Time capture gets designed by finance for accuracy and reviewed by nobody who has to do it fifty times a week between client meetings.
Design time entry for the biller, not the reviewer, and test it with real fee earners before design locks.
Late, bulk, from-memory time entry is not a training problem, it is a design outcome. It degrades revenue recognition, utilization and project margin at the same time, and none of those errors are visible individually. The firms that get this right test entry with actual fee earners early and treat every extra click as a cost.
Has a fee earner tested time entry end to end, and how long did it take them?
Client, engagement, project, phase and task all have to carry billing rules, revenue treatment and reporting. The structure gets set early, often by whoever is available.
Set the hierarchy with the people who price work and the people who report on it in the same room, before conversion.
This structure decides what questions the firm can answer later. Get it too shallow and you cannot analyze profitability where it varies. Too deep and fee earners misfile everything. Either way it is close to immovable after conversion, because every historical record inherits it.
What profitability questions will we be able to answer after cutover, and which will we lose?
Pricing has drifted over years into a mix of fixed fee, time and materials, capped arrangements and milestone structures, and the contract terms live in documents rather than in a system.
Inventory the live pricing constructs before design, and make the revenue model fit them rather than the reverse.
The design usually assumes a cleaner commercial model than the firm sells. Every unmodeled construct becomes a manual adjustment at close, and manual adjustments at close are what auditors write up. Getting the inventory first is cheap; discovering it at the first quarter close is not.
Which of our live pricing constructs are modeled in the design, and which are manual?
Compensation structures in a partnership carry a level of confidentiality and complexity that generic HCM security models were not designed around.
Design the security model against the compensation reality first, and have someone senior test what each role can see.
Security gets configured to a role matrix that looks correct and is tested by administrators who can see everything. The first time a manager sees something they should not, the trust cost across the partnership is far larger than the technical fix.
Who has tested, as an ordinary user, what each role can see?
The plan gets built around a technical readiness date. Billing runs on its own cycle and clients expect invoices when they always arrive.
Sequence cutover so the first billing cycle after go-live has slack in it, and staff that cycle as if it were part of the program.
A delayed or wrong first invoice run reaches clients, which makes it a commercial problem rather than a systems one. It is also the moment every design decision about time, rates and revenue gets tested at once, so it deserves the treatment a parallel run gets in payroll-heavy sectors.
Who is staffed to run the first full billing cycle after go-live, by name?
What those five mean for the chair you sit in
Firms judge programs by realization and utilization, not by milestones. Each seat's exposure, the early sign, and the question worth asking this quarter.
Adoption decides everything downstream
Every number the firm manages by comes from data that fee earners enter between client commitments. If entry is slow, the data arrives late and approximate, and no amount of reporting investment fixes a bad input. The other exposure is political: partners who dislike the system will say so loudly and early.
The design has been reviewed by finance and not by anyone who bills.
Has a partner tested time entry and told us honestly how long it took?
Revenue quality depends on constructs you may not have inventoried
Fixed fee, capped, milestone and time and materials arrangements each carry different revenue treatment. Anything unmodeled becomes a manual close adjustment, which is both a control weakness and a monthly cost. The project hierarchy set at conversion then decides which profitability questions you can answer at all.
Close depends on spreadsheet adjustments nobody has scoped out.
What percentage of revenue is recognized without manual intervention?
Confidentiality is a design requirement, not a setting
Partnership compensation, performance data and staffing plans carry sensitivity that a generic role matrix does not respect by default. Security tested only by administrators is untested. And the firm's people move between roles constantly, so the model has to survive change rather than a point-in-time review.
Security was signed off from a role matrix rather than from user testing.
Who has verified role visibility as an ordinary user rather than an administrator?
Four leaks, and only one of them is visible in a status report
Illustrative weighting from what these programs surface in practice, not a benchmark. The point is the ranking, not the heights.
Only the last one looks like a defect. The other three look like normal operations, which is why they survive hypercare and become permanent.
Five questions worth more than a readiness assessment
Answerable from memory, scored on this page, nothing captured and nothing emailed.
These five are the start of the instrument. A full review also covers rate structures, resource planning integration, WIP and revenue controls, and the first-close plan. Or skip the tooling and book the program review.
Putting a firm on a new platform?
Pre-SOW, mid-build, or stabilizing after a rough first close. I sell no software and staff no builds, so these questions get asked out loud. Tell me where the program is and I will tell you what I see.