How do you see which lab projects are slipping?

Moving one project's end date by three weeks bought our example lab about twelve days of green on its roadmap. It didn't buy a single passage of cells.

That's the awkward thing about tracking slip. You measure the project against its plan, and the plan is the one thing you're allowed to change. In control terms, you're reading the error against a setpoint you can move. So a roadmap needs readings that remember the original plan, and people who know which reading answers which question. Here's one re-plan, followed through the portfolio view, and what the view can't tell you.

What does it mean for a lab project to slip?

A project slips when a milestone's due date moves later than the date first committed to. The first committed date is the baseline, and slip is the gap in days: a milestone first due on 14 August and now due on 4 September has slipped 21 days.

The baseline has to be written down at the moment of the first move, by the system, and never edited. If it can be typed in or changed, it becomes the date people wish they'd promised. In CCA Labs the baseline is set the first time a milestone's due date moves, and it's write-once. A project's slip is the largest slip across its milestones, shown as +21d. A milestone that never moved shows a dash in its Slip column, not zero: never measured isn't the same as on time.

A re-plan, step by step

The example lab is a small biotech adapting HEK293 cells to grow in suspension without serum. The adaptation is its own project, PRJ-ADAPT-26, running from 1 June to 30 September, with the serum stepped down from 5% to nothing.

At 0.5% serum it stalls. On 3 August viability is 82% and the cells are clumping. On 10 August the cell culture scientist and the head of R&D allow three more weeks:

Milestone	Baseline	Due now	Slip
A3 Growing at 0% serum	14 Aug	4 Sep	+21d
A4 Three stable passages at 0%	4 Sep	25 Sep	+21d
A5 Adapted bank frozen, 20 vials	25 Sep	16 Oct	+21d

The project's end date has to move too, from 30 September to 21 October, because CCA Labs won't let a milestone fall after its project ends. A1 and A2 were done on time and never moved, so they have no baseline.

The first 2 L bioreactor run in the scale-up project needs that adapted bank. Nothing moves it automatically, so the process scientist moves it by hand the same day: 27 November to 18 December. The final milestone, three runs that agree within 15%, can't move the full three weeks. That would put it on 5 March 2027, after the grant paying for it ends on 28 February. So it moves two weeks, to 26 February, and the runs are squeezed a week closer together. Two days of margin are left.

How does the roadmap decide a project is at risk?

By three checks, any one of which is enough: a milestone past its due date, progress behind the elapsed schedule, or projected spend over the allocation. A blocked milestone overrides them all and makes the project Blocked.

The middle check is the interesting one. Elapsed is the share of the project's time used, from start date to end date. Progress is the share of milestones completed, unweighted. If elapsed minus progress is more than 0.2, the project is At Risk, "Progress is behind the elapsed schedule". It's relative, so it scales: a three-year grant half gone with a fifth done is at risk, and a two-week project on its first morning isn't.

On 31 August the adaptation is 91 days into 142, so elapsed is 0.64. Two of five milestones are done, 0.4. The gap is 0.24, so it's At Risk, and it crossed the line in the last week of August. The owner gets an alert when a project moves into At Risk.

Now work the same check against the old end date, 30 September. Elapsed on 31 August would be 91 / 121 = 0.75, the gap 0.35, and the project would have gone At Risk around 12 August, two days after the re-plan. Moving the end date held the alarm off for about twelve days. Nothing about the cells changed.

Why can a project slip and still be on track?

Because slip and health measure different things. Slip is how far the dates moved; health is whether the work keeps pace with the time it now has.

The scale-up, PRJ-SCALE-26, shows +21d and reads On Track. It runs from 3 August to 26 February 2027 and is 14% of the way through. One milestone of four, "Vessels and probes received", was done on 27 August, so progress is 25%, ahead of elapsed time. The health check is right that nothing is late. It can't see that the grant ends two days after the project does. That's knowledge a person has to bring to the review.

What the portfolio shows on 31 August

The Plan draws each project as a bar from start to end, with its milestones as diamonds. A milestone that moved keeps a dashed ghost diamond at its baseline, joined to its live diamond by a red line. The adaptation's A3, A4 and A5 each trail a three-week red line, so the slip is visible at a glance. The programme gets a group bar spanning its projects.

The Table flags the adaptation Behind Schedule, and the project's own page gives the reason in full: "Progress is behind the elapsed schedule".

Next 30 Days shows every open milestone due in the window, with anything late in a pile of its own. On 31 August that's A3, due 4 September, and A4, due 25 September. Owners and members get a reminder once a milestone is three days out.

Team Load adds up each person's allocation across open projects. The process scientist is at 120%: 70% on the scale-up, 20% on the adaptation, and 30% on a project in another programme.

Where it goes wrong

Milestones don't depend on each other. Runs can: on the Plan, a run can depend on a run in another project, a broken order shows in red, and moving one run moves the runs after it. But the first 2 L run is a milestone here, so it had to be moved by hand. Forget to, and the Plan shows a bioreactor milestone before the cells it needs exist, with no warning. It's a reason re-plans need a person who knows the whole chain.

Team Load ignores dates. It adds 70, 20 and 30 and says 120% for every month. It can't say that the real crunch is December, when the first 2 L run and the other project's deadline land in the same fortnight.

Progress is unweighted. Five milestones means progress moves in steps of 0.2, the same size as the threshold. One completed milestone can flip a project from At Risk to On Track overnight.

Overrides hide as well as explain. A manager can set a project's health by hand with Set Health, with a required reason shown to everyone and a hand icon on the badge, "Set by Hand". It's the right tool for "waiting on ethics approval". It would be the wrong tool here: the check is telling the truth.

How CCA Labs lays it out

The Projects page has two views, Plan and Table, grouped by Programs, Funds, Owners or Health, or flat. The Plan zooms from days to years, with a line for today and a legend that counts projects On Track, At Risk, Blocked and Planning; click one to filter.

A project whose start date hasn't come yet is drawn as Planning. The Table shows each project's slip as a chip, and each project's Milestones list has a Slip column, with the original due date on hover. Health is worked out when the page loads, so it can't go stale. Equipment clashes don't feed it.

You can lay your projects out this way when you create your lab. The planning tools are on the all-in-one lab page, and the goals these milestones count towards are in how to set research goals and milestones that show when you're behind.

The roadmap counts days. The bench counts money as well: the extra passages at 0.5% serum, and the agent that rescued them, cost the adaptation £371.10. What goes into a number like that, and why the programme page says £138.28 a run when a passage costs £61.42, is the next part: how much one lab experiment really costs.