CEDX Time · People & HR
Time tracks hours like a time clock and reads them like an analyst: deep versus shallow work, a fragment score for every person-day, and context-switch storms priced in the recovery minutes they burn.
The most honest number on the overview is 121.5 — hours of recovery cost, the re-entry minutes burned after meetings and ping storms. Not tracked time. The time tracking never captures.
Runs on demo data. Timers use fixed demo durations — the screen says so itself.
What it is
Sixty person-days scored on switches, short-block percentage, meeting minutes, deep minutes and drift — with an expected loss in hours on every row. Forty-one score 55 or higher and land in the protect queue.
Ninety-six clusters with names like “Meeting → deep failed re-entry”, “DM ping cascade” and “On-call alert spray” — each with switch count, recovery per switch and a total cost. Eighty-seven read high risk.
The timer form is labelled “demo form, does not persist”, and the running timers say “fixed demo durations — no live wall clock (SSR-safe)”. Draft your day auto-groups twelve captured minutes into a suggested block — “review before posting”.
Product tour
Not a mockup and not a concept deck. This is what opens at /app/time.
01 — Overview
Tracked 252.3 hours, deep work 49%, fragment score 60, switch cost 121.5 hours, focus health 59 and 853 switches. Hours split Deep 124, Meeting 41, Shallow 37, Comm 32, Admin 19 — and the trend chart splits deep from shallow plus comm, weekday by weekday.
02 — Fragmentation
The board scores each person-day — 88 at the top with 12 switches and 56% short blocks — and puts a loss figure beside it: 3 hours 2 minutes for that day alone. Protect levers and the scoring model are one tab away.
03 — Switch cost
Individual switches are noise; clusters are the problem. The board groups them — 14 switches with 12 minutes of recovery each is 2 hours 55 minutes gone — and ties each cluster to a project, so “Checkout rebuild” carries its own switching bill.
04 — Timer
Timers run against projects with a category — 1h 24m of Deep on a payment refactor, 41m on a cart accessibility pass — and the day drafts itself: twelve captured minutes grouped into one suggested block, review before posting.
Who runs it
Roles, not references. We have no named customers yet, so nobody in these photographs is quoted, credited or claimed as one.
Watches the protect queue: 41 high-fragment person-days is a scheduling problem before it is a productivity one.
41 high days · loss 61h 11m
Reads deep share against meeting load — 124 deep hours against 41 meeting hours — before agreeing to another recurring slot.
deep 49% · target 50%
Checks switch cost by project before the retro: which work is paying a 121.5-hour switching tax, and why.
121.5h recovery · 96 clustersThe shape of it
Every figure below is legible in the captures above. Nothing here is a projection of your team — it is the state of the demo data.
How it runs
Timers and log segments capture the week — 320 segments this one — each with a project and a category from Deep to Admin.
Every person-day gets a fragment score from switches, short blocks, meetings and drift; 55 or higher joins the protect queue.
Switch clusters are priced in recovery minutes — the difference between 853 switches as trivia and 121.5 hours as a bill.
Levers move the simulated average from 60 to 34 and cut switch cost to 70.1 hours; the saving is printed as 51h 28m, on screen.
One record
Time reads the estate's own records: projects to log against, people whose week it is, seats that need covering.
Drafted day blocks — accept, edit, merge — are where captured time becomes a submittable sheet.
CEDX ProjectsTimers run against projects like “Checkout rebuild”; the switch-cost board prices storms per project.
CEDX ShiftsThe people whose fragment scores climb are the same people Shifts seats; understaffing shows up here as fragmentation.
CEDX PerformanceWeekly check-ins and goals sit on the same record the focus data describes.
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 projects, your week
Time 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/time. It runs on demo data, and the timer says so on screen.
A per-person-day score built from switches, short-block percentage, meeting minutes, deep minutes and drift — the columns are visible on the Fragmentation board. In the demo, 41 of 60 person-days score 55 or higher.
Recovery minutes after context switches, grouped into clusters like “Meeting → deep failed re-entry”. The demo week prices 96 clusters at 121.5 hours — the re-entry time ordinary time tracking never sees.
Not in the demo — and it says so: “demo form, does not persist”, “fixed demo durations — no live wall clock”. The screens show the mechanics; live capture is part of a pilot.
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 ask what last week's switching bill was for your own team.