The Lifecycle Dashboard

Four modules, one screen

Nothing on this dashboard is computed here. Each card reads a module you can also open on its own, and that’s the first thing to know when a card looks wrong — the fix is almost always in the module, not on this screen.

On an account that hasn’t started using off-boarding, four cards read No off-boarding cases / No exits recorded / No analyzed exit interviews yet and the 90-day attrition tiles read zero. That’s an empty module, not a broken dashboard.

“Open” means two different things here

This is the one to understand before quoting anything off this screen, because two cards sit side by side and count the same cases differently.

Grievance status splits cases into Open and Closed by their status: Resolved, Dismissed and Withdrawn count as closed, everything else is open. Five cases, three open, two closed.

Grievance aging lists the oldest open cases — but it doesn’t filter on status at all. It filters on whether a case is final, and a case only becomes final when its appeal window shuts. So the two cases the card next to it just called closed can be sitting at the top of this list, oldest first.

Two rules follow, and they’re worth knowing before this misleads someone:

• Read Grievance status for counts. It’s the one that matches what the Grievances list shows you.
• Read Grievance aging as “cases not yet finalised”, not as a work queue. A case appears there until the appeal window lapses — which is deliberate, since a dismissal can still be appealed — but the heading doesn’t say so.

Withdrawn cases never leave that list. A withdrawal doesn’t open an appeal window, and nothing else marks it final, so a case somebody withdrew months ago stays under oldest open cases indefinitely and, being the oldest, stays at the top. If the aging card is showing you something ancient and irrelevant, that’s usually why.

The Avg resolution tile is narrower still: it averages only cases whose status is Resolved, and only those resolved in the last 90 days. Dismissed and withdrawn cases contribute nothing, so an account that closes most cases without a formal resolution shows a dash here however busy it’s been.

Attrition is measured twice, two different ways

Attrition (90 days) and Attrition rate trend sound like the same measurement at two zoom levels. They aren’t.

• Attrition (90 days) counts exit cases started in the last 90 days, by the date the case was opened. Its tile says Exits (90d), but somebody serving three months’ notice is already counted.
• Attrition rate trend counts exits by last working day, divided by that month’s headcount, as a percentage.

So the trend is people who have actually gone; the 90-day tile is people on their way out. During a notice period the same person appears in one and not the other. For “how many people have we lost”, read the trend. For “how much off-boarding is in flight”, read the tile — and its Active exits neighbour, which is the honest version of the same number.

Attrition rate trend needs monthly headcount snapshots, and BeeHR takes one per month. Until two exist, the card explains itself rather than looking broken:

There’s no backfill and no setting to switch on. Same mechanism as the two snapshot-fed cards on The Workforce Dashboard.

Hires vs exits

Six months of arrivals against departures, and the fastest read on this screen — lines crossing is a story, whichever way they cross.

Its own footnote carries the caveat: exits before the off-boarding module are invisible. Hires come from joining dates on employee records, which go back as far as your data does; exits come from off-boarding cases, which only start when you begin using the module. A company that has just adopted BeeHR sees hires with no matching exits and looks like it has never lost anybody.

The hires side reads the first 100 employees. Above that headcount, treat the card as indicative and reconcile against the employees list.

Promotions, and what “never promoted” means

Promotions gives four tiles: this quarter, pending, total, and average raise.

The average is over percentage-based raises only — it says so in the footnote — so a company that grants fixed-amount increases sees a dash no matter how many promotions it has run. It also averages every percentage promotion on record whatever its status, so a pending or rejected one still moves the number. Read it as “the size of raise we typically propose”, not “what we paid”.

Longest since promotion ranks people by how long it’s been, and it counts only promotions that have been applied — not approved. Approval and application are separate steps: a promotion applies on its effective date, by a scheduled job.

That gap is visible in normal use. Approve someone’s promotion today with an effective date next month and this card still says Never promoted against their name, counting from their joining date, until the date arrives. Nothing is wrong; the raise hasn’t happened yet. See Creating, Approving and Reverting a Promotion for the full sequence.

Rows showing Never promoted against a long tenure are what the card is for. Six rows only, longest first, from the first 100 employees.

Exit interviews are deliberately blurred

Exit-interview sentiment is the one card on this dashboard built to tell you less than it knows.

It shows an average NPS out of 10, a count of detractors (anyone scoring 6 or below), and up to five recurring themes. What it won’t show is a theme that only a few people raised: a theme is reported only once enough interviews mention it — five by default — and everything below that threshold is counted and hidden, with a footnote saying how many were suppressed.

That’s a re-identification guard, and it’s the point of the card. In a twelve-person department, “manager” appearing once is a named accusation; the suppression is what makes it safe to put exit-interview data on a dashboard at all. If a footnote says themes are hidden, the answer is more interviews, not a setting.

The same instinct runs through Exits by type & reason: confidential reasons are excluded server-side. A reason marked confidential when the exit case was created is counted in the exits donut but left out of Top reasons — so the reasons list can legitimately add up to fewer than the total above it.

Tenure and span of control

Both cards also appear on The Workforce Dashboard, where they answer “who do we have”. Here they sit beside attrition and answer “who might we lose”.

Tenure distribution bands people by time served. People with no joining date are excluded rather than shown as unknown, so its total can be below your headcount.

Span of control gives three numbers: how many managers, the average number of direct reports, and the largest team. A manager here is anyone at least one person names as their reporting manager — a structural fact, not a job title. The Max reports tile turns amber at eight, which is a prompt to look rather than a policy.

Both read the first 100 employees.

Reading it well

Three cards earn their place on a monthly review: Hires vs exits for direction, Longest since promotion for the people most likely to be next out of the door, and Grievance status for whether anything is accumulating.

The rest is context. And when a number here disagrees with the module it came from, the module is right — this screen is a summary, and every summary on it is a definition somebody chose.