CEDX Resourcing · Projects & Delivery
Resourcing plans people against weeks: 1,124 hours planned for W31 against 932 hours of capacity, ten people over 100%, the bench priced at $107K of idle cost. The over-allocation is on the grid before it is in the exit interview.
The grid does the thing spreadsheets hide: one consultant's W31 cell reads 320% — four jobs, one person, same week. You cannot fix what you cannot see, and this you can see from across the room.
Runs on demo data. The 24 people in the captures are the seeded demo roster.
What it is
Eight weeks of schedule, person against week, with tentative hours hatched and leave in a diagonal pattern. Jade's W31 reads 320%; an engineering manager on a 20-hour week reads 280%, 300%, 220%, 160%, 260% across five weeks. Leave and work in the same cell is visible, not discovered.
Twenty-four people with role, practice, utilisation, free hours this week and over four weeks — and skills on the row: “Inventory, QA, React”, “Tax, Checkout, SQL”. The bench-ready count of 14 is computed from ≥40 free hours over four weeks, not guessed.
Six open placeholders spell out the role, skill, weeks and load — senior checkout engineer, 24h/week, P0, W32–W36 — 129h/week of unfilled demand in total. Each has a Find match button that reads the roster's skills and free hours instead of starting a email thread.
Product tour
Not a mockup and not a concept deck. This is what opens at /app/resourcing.
01 — Overview
Planned 1,124h against 932h of capacity, average utilisation 149% among people with load, 178 person-week conflicts scored with 20 rated high. The alerts name names: one senior consultant at 128% — 52h planned on 40h — and leave colliding with a cutover, 24h confirmed plus 18h tentative while away.
02 — Capacity
Person against week for W31–W38: utilisation per cell, tentative hours hatched, leave diagonal. The over-capacity count of 10 and the 424 over-hours to rebalance come straight off this grid — and so does the fix, because the free cells are on the same screen.
03 — People
Twenty-four rows of who can do what: role, practice, this week's utilisation, free hours now and over four weeks, skills. The 180% platform engineer has zero free this week but 76h over four; the 70% mobile engineer has 100h — the rebalancing shortlist writes itself.
04 — Allocations
320 allocations with hours, percentage of week, state and match score on every row. Above the ledger, the six placeholders of unfilled demand wait for a person, each with role, skill, weeks, priority and a Find match button.
Who runs it
Roles, not references. We have no named customers yet, so nobody in these photographs is quoted, credited or claimed as one.
Owns the person-by-week grid. A 320% cell is fixed with a trade in week 31, not with an apology in week 38.
conflicts · 178 scored
Reads the placeholders: six unfilled roles at 129h/week is the hiring-and-borrowing agenda, with priority P0–P2 already attached.
unfilled · 6 roles
Sees their own week as it is actually planned — confirmed, tentative and soft hours separated — instead of hearing about the fourth project at the fourth kickoff.
states · confirmed / tentative / softThe shape of it
Every figure below is legible in the captures above. Nothing here is a projection of your estate — it is the state of the demo data.
How it runs
Work arrives as placeholders: role, skill, weeks, hours and priority — six currently unfilled at 129h/week, waiting for a person rather than a hope.
Find match reads the roster's skills and free hours. The match score is printed — 36% on the current top rows — so a weak match is visible as weak.
The allocation lands on the grid as confirmed, tentative or soft. 320 allocations make up the week: 1,124h planned, 424h over capacity to rebalance.
Conflicts are scored per person-week — 178 of them, 20 high — and the grid shows where the hours can go: the bench-ready 14 with ≥40 free hours over four weeks.
One record
Resourcing plans the week; the rest of the estate lives it. The grid, the timesheet and the roadmap are three readings of one roster.
Planned hours on the grid become logged hours on the sheet; the utilisation forecast is where the two are reconciled.
CEDX ServicesThe placeholders are engagement demand — a senior checkout engineer for a rollout — and the engagement margin depends on the match.
CEDX RoadmapAn initiative's effort in eng-months is real only if the people exist; the roadmap's capacity tab reads this roster.
CEDX LeaveThe diagonal cells on the grid are leave records. Planning against a week someone is away is exactly the collision the alerts name.
Limits
Finding this out on the third call is worse for you than reading it here, and worse for us.
Start
Open it right now
Your roster, your skills
Resourcing with the rest of it
Questions
Yes. Every screenshot is a capture of the running build and you can open the same build at /app/resourcing. It runs on demo data.
Four allocations landing on one person in one week — 320% of their capacity, rendered in red on the W31 cell. The grid total computes the damage: 10 people over 100%, 424 over-hours to rebalance.
Each unfilled placeholder carries role, skill, weeks and load. Find match reads the roster for people with matching skills and free hours in those weeks, and prints a match score — the current rows read 36%, which is the product saying the fit is weak, not hiding it.
Computed, not declared: bench-ready means at least 40 free hours over the next four weeks — 14 of the 24 demo people qualify — and the bench-idle card prices the free hours at $107K of cost.
Yes — leave is marked diagonally, directly in the person-by-week grid, and the overview alerts on the collision: “24h confirmed + 18h tentative while on leave.” The records themselves live in CEDX Leave.
No certification has been issued. What we can evidence about hosting, encryption, tenant isolation and production access is written up on the security page.
Live build, demo data, no card. Then count how many of your people would read over 100%.