Product portfolio management
Definition
Product portfolio management
Product portfolio management is the decision about which products get funded, which keep running and which are killed. It looks across the whole set of products a company sells, compares them on one set of criteria, and moves money between them.
Portfolio prioritization is the ranking exercise that produces an ordered list and a funding line. Product portfolio management is the standing job of applying that decision to products, year after year, including the unpopular half — stopping things.
It is also not product lifecycle management, which keeps the controlled engineering record of one product from concept to retirement. Portfolio work asks a blunter question: should that product exist at all next year?
And it is not project portfolio management. Projects end, so a project portfolio empties itself. Products do not, which is why a product portfolio only shrinks when somebody decides it should.
Product ops is a different animal again. That function keeps product teams running day to day with tooling, data and rituals; portfolio management decides which products deserve a team at all.
Key takeaways
- Product portfolio management decides which products are funded, continued or killed, using one shared set of criteria.
- The hard half is stopping things; most portfolios fail by addition rather than by bad selection.
- Score each product on contribution, strategic fit and cost to run, then draw a visible line.
- Review on a fixed calendar, because a portfolio agreed in January describes a different company by June.
How it works
Product portfolio management works by putting every product on one page, scoring each against shared criteria, and giving each a verdict: invest, hold, harvest or retire. The verdict then has to change a budget line, or nothing actually happened.
| Verdict | What it means | What changes next |
|---|---|---|
| Invest | Growing and strategically central | More funded capacity |
| Hold | Profitable but not strategic | Flat budget, no new bets |
| Harvest | Declining and still paying its way | Maintenance only, price reviews |
| Retire | Costs more than it returns | A dated exit and migration plan |
Cost to run is the number most portfolios miss. A product with flat revenue and a rising support burden is already losing money — well before the revenue line says anything about it.
The retire verdict needs the same discipline as the invest one. Software teams have a practice for exactly this: application retirement sets out how a system is decommissioned, its data archived and its users moved somewhere else.
Public sector practice formalised the same loop decades ago. GAO’s 2004 investment management framework was built on a select, control and evaluate approach to investments drawn from the Clinger-Cohen Act of 1996, which is portfolio management under a different name.
Agree the criteria before the candidate list exists. A scoring model written after the proposals are in will always flatter the product its author already favours, and everyone in the room knows it.
The output is one page, not a deck. If the portfolio cannot be read at a glance by the person holding the budget, the review defaults to whichever sponsor argues hardest on the day.
Frameworks outside product management push the horizon further out.
The National Institute of Standards and Technology runs Baldrige as the only public-private partnership improving organisational performance, and the words it chooses are resilience and long term success, not launches.
Examples
Three shapes of portfolio decision turn up again and again, and they are easiest to see where the products are visible from outside the company: in catalogues, price lists and public release notes.
Consumer software companies prune on a published calendar. A product is marked for sunset, given an end date and a migration path, and the engineers move to the products that are still growing. The announcement is the portfolio decision made visible.
Industrial manufacturers run the same exercise across machinery families, where the spares obligation outlives the sale by decades. A line can be retired commercially while remaining supported — a portfolio verdict with a very long tail.
Large enterprises split the question in two. An innovation portfolio holds the uncertain bets that are judged on learning rather than return, while application portfolio management does the same sorting for internal systems instead of saleable products.
Business services firms face the sharpest version of the question. A service line costs almost nothing to keep listed — and quite a lot to keep staffed, so the review there is really a decision about where senior people spend their week.
Related terms
- Portfolio prioritization: the ranking decision and funding line that product portfolio management applies to products.
- Product lifecycle management: the controlled record of one product; portfolio work asks whether it should exist.
- Innovation portfolio: the subset of uncertain bets judged on learning rather than on near term return.
- New product introduction: the phase that takes a funded design into production and first shipment.
- Product outsourcing: handing product build or support to a partner, often the cheapest way to keep a harvest product alive.
FAQ
How is product portfolio management different from project portfolio management?
One sorts products, which outlive their funding rounds and keep costing money. The other sorts projects, which end. The first needs a stopping rule and the second mostly does not.
Who owns the portfolio decision?
Usually an executive group with the budget authority to act on it, supported by product and finance. A review that can recommend but cannot move money is a reporting exercise wearing a governance badge.
How often should a product portfolio be reviewed?
Twice a year suits most companies, with a lighter check each quarter. Anything slower and the list on the page describes a company that no longer exists. Pair the review with the budget cycle so the verdicts have somewhere to land.
What is the most common mistake?
Adding without subtracting. Portfolios grow one launch at a time and nobody is thanked for a retirement, so costs pile up in support, patching and training. The revenue line never tells you that, which is why cost to run must be a scored criterion.
Can a small company do this properly?
Yes, and with fewer products the arithmetic is easier rather than harder. Three products on one page with revenue, cost to run and a verdict against each is a complete portfolio review.
Compare partners who can take on the build or support of a product you have decided to keep at Outsource Accelerator.
Independent
