PortfolioPlane

Proof, not status.

Why your PMO tool computes a health score nobody trusts

Every major portfolio tool ships a red-amber-green verdict. Almost none of them agree on what produces it, and two of them will tell you so in their own documentation.

Published 2026-08-15Vendor documentation read 2026-08-14Every competitor claim links to its source

Ask a program director how they actually run a portfolio review and you get a version of the same answer. The tool says amber. Beside it they keep a spreadsheet listing which workstreams are late, by how many days, and who owes the next thing — and when the steering committee asks how the program is going, that is what they read from. The amber never comes up.

The tool is not broken. The number is doing exactly what it was built to do; it just cannot survive one follow-up question, so people route around it and rebuild by hand the thing it replaced.

Here is the thesis: a blended health score is the only figure in a portfolio system that cannot be traced back to a fact, and every vendor that ships one has quietly documented why. You do not have to take that on faith. All five of these formulas are published by the vendors themselves.

Five products, five different things called “health”

Sciforma computes a Global Score out of 100 as a weighted average of Time, Cost and Quality, where the Quality component is itself derived from planning problems, resource problems, transaction problems, status reports, work items and risk-completion ratios.3 It is the purest blend on the market. Ask what a 61 means and the honest answer is that it means the weighted average of those inputs came to 61.

Broadcom Clarity does not compute from data at all. A project manager picks a status for schedule, for scope, and for cost and effort. Each choice is scored 10, 20 or 30. The overall gauge is the sum: green below 40, yellow from 40 to 89, red at 90 or above.2 The arithmetic is deterministic and the inputs are opinions, which makes the output a very precise rendering of how three people felt on a Tuesday.

Celoxis goes the other way and computes RAG automatically every day, extrapolating a projected finish date and a projected cost from current progress.10 This is the most defensible of the five — right up to the per-project setting, in the product’s own documentation, called “Always assume this project is on time and budget”, which forces the indicator to On Track regardless of the data.

Jira Align tracks five dimensions of health — budget, people, quality, risk and issues, and schedule — each set by hand. Atlassian’s own guide notes that no history is kept, describing it as “a point in time view”, and observes that practitioners work around this by typing the date into the reason field.5 A health indicator with no history cannot answer when a project turned, which is the only question anyone asks a health indicator after the fact.

ServiceNow SPM now predicts RAG with AI, with admin-tunable calculation logic — available to customers holding the Now Assist and SPM Pro Plus licenses.4 The prediction may well be good. It is also a per-seat upsell on the answer to “how is this going”.

The objection, and why it does not hold

The obvious defense is that an executive needs a summary — that nobody will read eleven workstreams, and a color is how you get attention in a forty-slide pack. That is a real problem and it deserves a real answer.

But notice what the summary costs. To average schedule against cost, you must decide how many days of delay equal how many dollars of overrun. Every one of these products makes that decision for you and none of them shows you the exchange rate. When the amber turns red, the first question in the room is always the same — why — and the number cannot answer it, because the number is where the information went to be destroyed.

Wellingtone’s annual survey of PMO practice keeps finding the same thing from the other end: organizations struggle to trust the project data they hold.57 The tools are not failing to compute. They are computing something nobody can defend in a meeting.

What we do instead, and what it costs us

PortfolioPlane publishes no health score. A portfolio reads as counts that keep their denominators: 119 requirements mapped of 119, 98 resolved of 119, 0 verified of 119. Nothing is averaged, so nothing has to be un-averaged when somebody asks why.

The cost is real and worth naming. A count with a denominator does not fit in a status column, and there is no single cell to paste into an executive pack. Our answer is that the sentence you would have to write instead — zero of a hundred and nineteen requirements have been verified by a named person — is the sentence the meeting was going to arrive at anyway, twenty minutes later, after somebody asked what amber meant.

The spreadsheet beside the dashboard had already worked this out. It is a register of facts, kept by hand, because the tool would only give a verdict. That spreadsheet is the product spec.

Sources

Read on 2026-08-14. The full numbered register across all eleven products is on the comparison hub.

  1. 1Broadcom Clarity — Project Status Summary report (dimension statuses summed 10/20/30; green under 40, yellow 40–89, red 90 and above)vendor documentation
  2. 2Sciforma — Project Health Score (weighted Time, Cost and Quality scores, out of 100)vendor documentation
  3. 3ServiceNow — AI Status Reports (RAG predicted per dimension, behind Now Assist)vendor community article
  4. 4Atlassian — A Project Manager Guide to Jira Align, Part 3 (five-dimension manual health, no history retained)vendor community article
  5. 5Celoxis — RAG indicators (computed daily; per-project “always assume this project is on time and budget” setting)vendor documentation
  6. 6Wellingtone — The State of Project Management 2026, press release (72% collate reports for half a day or more each month; 44% dissatisfied with PMO reporting)practitioner research