- What does product portfolio strategy actually decide?
- Why don’t the standard portfolio matrices work here?
- The unit of decision is the line in a segment
- When should you fund a line?
- When should you harvest a line?
- When should you exit a line?
- How to run the decision
- Frequently asked questions
Product portfolio strategy is the set of decisions about which of your product lines get new money, which get run for cash, and which get closed. Three verbs: fund, harvest, exit. Everything else filed under the phrase, the roadmaps, the innovation funnels, the planning templates, is either an input to those three decisions or a way of postponing them.
You will say you already make those decisions every year in the budget. You do not. The budget rolls last year’s allocation forward with adjustments, which is a decision to keep the portfolio you already have, made by default and filed as planning.
If your immediate problem is too many products rather than the wrong allocation, start with the pillar guide on how to cut a product line without losing revenue. This article covers the investment side of the same decision.
What does product portfolio strategy actually decide?
It decides where the next dollar, the next engineering hour, and the next hour of bottleneck capacity go, line by line and segment by segment. A portfolio strategy that does not change at least one of those three allocations is a description of the portfolio, not a strategy for it.
Three scarce things, not one
Most businesses treat portfolio decisions as capital decisions, because money is the resource with a formal approval process. But money is rarely the binding constraint in an industrial business. Engineering hours and bottleneck capacity usually are, and they get allocated informally, to whichever line has the loudest advocate or the most urgent customer this month.
A portfolio strategy worth the name allocates all three on purpose. If the capital plan funds a line and the engineering queue starves it, the line has not been funded. It has been given a budget it cannot spend.
Isn’t that what the capital plan is for?
The capital plan answers a narrower question: which projects clear a return hurdle. It does not ask whether the line those projects serve should be growing at all, and a harvest line can produce a perfectly good project that should still not be funded. Portfolio strategy sits upstream of the capital plan and tells it which lines it is allowed to invest in.
Why don’t the standard portfolio matrices work here?
Because they score a whole product line on one position, and industrial lines rarely have one. The same line can lead in one segment and trail in another, and the classic matrices average those into a single dot that describes neither, then attach an instruction to the dot.
A line is not one position
The growth-share matrix, the one with stars, cash cows, question marks, and dogs, places each business on market growth and relative market share. Applied to an industrial product line, it asks for a single share figure. But a line of pumps, valves, or conveyors typically sells into several segments at once, and its position in each can be completely different. A line can be the default choice in one industry and an afterthought in another. One dot cannot hold both facts, so the matrix quietly discards one of them.
Share you cannot measure
The second problem is the input. Reliable market share data exists for consumer categories and for large industrial markets. For most niche industrial segments it does not, and the share figure that goes into the matrix is an estimate made by the people whose line is being scored. The output looks analytical. The input was an opinion.
The dog that holds up the account
The third problem is the one that costs money. A line the matrix labels a dog can be the product a top customer needs you to supply in order to keep buying everything else from you. Divest it on the matrix’s advice and you can lose far more than the line’s revenue.
You will object that these matrices are still taught for a reason, and they are. They were built for diversified companies allocating cash between whole businesses, and they are useful at that altitude. The trouble starts when they are pushed down to the product line, where the averaging does its damage.
The unit of decision is the line in a segment
The fix is to change the unit. Fund, harvest, and exit decisions should be made for a product line in a specific segment, not for the line as a whole, because that is the level at which a customer actually chooses you or does not.
The method I use for this is the Right-to-Win Matrix. It arranges product lines against the segments they serve and asks, for each cell, what advantage a customer in that segment could name for buying from you rather than from the next supplier.
Building the grid
Put your product lines down the side and the segments you sell into across the top. Fill only the cells where you actually have sales. For each filled cell, record three things: contribution over the last year, the direction it has moved over the last three, and the advantage, written in the words a customer in that segment would use. A cell that needs a paragraph of explanation from your own sales team does not have an advantage written in a customer’s words.
The one question per cell
If a customer cannot name the advantage, you do not have one. You have a price. That single question sorts most cells quickly. Cells with a named advantage and room to grow are funding candidates. Cells with contribution but no named advantage are harvest candidates. Cells with neither are exit candidates. What remains are the borderline cells, and that is where management attention should go.
When should you fund a line?
Fund a line in the segments where customers can name your advantage and where more investment would widen it. Profit today is not the test. A profitable cell with no named advantage is a harvest cell, and funding it buys volume a competitor can take back.
The funding test
Three conditions, all required. The advantage is named by customers rather than claimed by you. The segment has room: either it is growing or you hold a share small enough to grow within it. And the specific investment you are considering widens the advantage rather than simply adding capacity or variants. Money that fails the third condition is maintenance spending labeled as growth.
Funding is not always a new product
You will object that competitors are all launching new products and that funding a line means doing the same. It does not have to. At Illinois Tool Works I grew a U.S. equipment business from $42 million to $67 million in revenue without launching a single new product. The investment went into how the product was sold, moving customers from buying a machine when the old one broke to having their whole fleet of machines managed. The advantage customers could name was the relationship, and that was what the money widened.
When should you harvest a line?
Harvest a line that still earns but has no advantage a customer can name and no realistic path to one. Harvesting means you stop investing, keep serving well, and price for cash. It is a deliberate choice, and it fails most often when it happens by neglect instead.
What harvesting actually means
Freeze engineering on the line except for cost reduction and compliance. Stop adding variants. Test price increases, because the customers who remain on a line with no named advantage are often there for reasons other than price. Keep quality and delivery exactly where they are, because those are what keep the cash coming.
But isn’t harvesting just giving up?
No. Harvesting is funding the lines that can win with the cash from a line that cannot, and done well it can run profitably for years. The failure mode is harvest by neglect: nobody decides, investment dries up by default, quality drifts, customers notice, and the line collapses faster than any plan would have allowed, sometimes taking the customers’ other business with it. The difference between the two is whether someone wrote the decision down.
When should you exit a line?
Exit when a cell has neither contribution nor a named advantage, and the capacity or attention it consumes has a better use. Exit a segment before you exit a line. Most lines have one or two segments worth keeping, and a whole-line exit throws those away.
Exit the segment first
At the tank liner manufacturer I owned, the business concentrated on premium products where the margins justified the attention, left lower-value segments to competitors willing to work harder for less, and added a floor liner product aimed at larger, more profitable jobs. The product lines survived. Some of the segments did not, deliberately.
At Unarco, the Berkshire Hathaway shopping cart business I ran as president, the move was the same in a different form: capacity shifted toward remanufacturing, the most profitable segment, and away from manual lines with weaker economics. In both cases the decision was made at the level of the segment, and the line was reshaped rather than closed.
What an exit really costs
Count the obligations before you count the savings. Spare parts and service commitments often run for years after the last unit ships. Customers who must qualify a replacement need time and help. Tooling and inventory carry write-offs. And a line that shares components or capacity with others will shift cost onto them when it goes. If the line in question is large enough to be a business in its own right, the decision changes character, and the triage for that is in fix it, sell it, or shut it down.
How to run the decision
Run it as a fixed annual sequence with one owner, not as a debate that restarts whenever a line has a bad quarter.
The sequence
- Build the grid. Product lines against segments, filled only where you have sales.
- Put contribution and trend in every cell. Use the best allocation you have and note where it is weakest.
- Ask customers for the advantage. Ask them directly, in each segment, why they buy from you, and write down their words rather than your sales team’s.
- Mark every cell fund, harvest, or exit. Borderline cells get a named owner and a date to resolve them.
- Move one scarce resource this quarter. Reallocate engineering hours or bottleneck capacity from a harvest or exit cell to a funding cell, so the grid changes something real.
- Review on a fixed date each year. Refresh the grid, and review sooner only when a cell’s named advantage changes.
Who decides
One person. Not a committee, and not the owners of the individual lines. You will object that the people closest to each line should have a say, and they should: they supply the evidence, the customer conversations, and the cost figures. They should not vote on their own lines, for the same reason nobody grades their own exam. The decision needs an owner whose interest is the whole portfolio.
Frequently asked questions
How often should the portfolio be reviewed?
Once a year for the full grid, on a fixed date, plus an immediate review of any cell whose named advantage changes, such as a competitor launch in that segment or the loss of the customer who defined it. More frequent full reviews turn strategy into reaction.
Does a small manufacturer need a portfolio strategy?
More than a large one does. With fewer engineers and less capacity, each misallocated hour is a larger share of what you have. The grid for a small business can fit on a single page, and it should.
What if every line is profitable?
Then profit is not telling you anything, and the advantage test has to do the work. Profitable lines with no named advantage are harvest lines, and their cash should fund the lines that can win. Being profitable everywhere is often a sign that nobody has looked closely at the segment level.
Where does R&D allocation fit?
Downstream of this. Engineering hours are one of the three scarce resources the grid allocates, so the R&D plan should be built from the funding cells rather than argued separately. An R&D budget spread evenly across lines is a portfolio decision made by default.
One limitation of this method, stated plainly: it depends on attributing contribution to segments, and many cost systems cannot do that well. If yours spreads overhead by volume, the contribution in each cell will flatter the complex, low-volume segments and understate the simple ones, and the grid will recommend funding the wrong cells with confidence. Fix the allocation roughly before you trust the grid, or treat contribution as the weakest of its three inputs and lean on the advantage test.
About the Stagnation Assassin
Todd Hagopian is a Fortune 500 transformation executive who has generated $3B in shareholder value across Berkshire Hathaway, Illinois Tool Works, Whirlpool, and JBT Marel, where he serves as Global VP of Product Strategy and Marketing. Known as The Stagnation Assassin, he is the author of two published books: The Unfair Advantage: Weaponizing the Hypomanic Toolbox and Stagnation Assassin: The Anti-Consultant Manifesto. His blog is published in 15+ languages and read by operators worldwide. Bring him to your stage via his speaking page or connect with him on LinkedIn.
One question to answer yourself: which product line would you exit tomorrow if one customer did not buy it? If a name came to mind in under ten seconds, you already have your first portfolio decision, and the grid will tell you whether that customer is worth the line. Send what you found through the contact page and I will tell you what I would check next.