Nobody Has a Formula for Your Portfolio Split

How to Make the Call Anyway

You've built a defensible platform roadmap. Your VP wants three new product launches on the roadmap. Manufacturing wants a cost reduction on the legacy chromatography instrument before margins erode further. Finance wants one number, this quarter, for what percentage of the R&D budget belongs to each. You don't have a formula that produces that number. You have judgment, some data, and a strategy document that says "grow" without saying how much of the budget belongs to growth versus, say, keeping the chromatography instrument running with maintenance spend.

This is a different problem from the one your roadmap already solves. Deciding what makes the roadmap doesn't tell you how much budget each part of it actually gets, and that's before you even ask whether the strategy behind it was designed top-down or simply emerged from what the business was already doing.

This isn't a gap in your training. Robert Cooper, whose Stage-Gate® framework much of the industry already runs on, states it plainly in Winning at New Products1: no formula tells you what percentage goes where. If the person behind the industry's own portfolio playbook admits that, the absence of a formula isn't your failure to find one. It's just the problem itself.

The framework: strategic buckets, sized from the top down

Cooper's strategic buckets framework starts from a simple premise: strategy only becomes real when you decide where the money goes. Until a business allocates spend to specific projects, strategy is still just language in a slide deck.

The framework works in four moves, and the first two have to happen before the last two make sense.

1. Start with buckets, not project lists. Before you touch a single project, management decides how resources should split along a small number of buckets, broad categories like project type (new platform, product line extension, maintenance and cost reduction, sometimes called sustaining engineering or product care), product line, market segment, and sometimes geography. Cooper's research on Honeywell shows one business simplifying this to three buckets entirely, using the three points of the Mercedes star as a memory device for platform, new product, and maintenance spend. Fewer buckets forces sharper thinking about what actually matters.

   Across Honeywell's own product line split (Sealants, Roofing Membranes, Flooring Coatings, Deck Coatings) with its market segment split (Industrial, Residential, Automotive, Institutional), that's sixteen tiny slices, things like "7 percent of Deck Coatings spend goes to Institutional." Each slice is too small to fund or defend on its own. Two dimensions stay manageable. Four, crossed together, turn into decimal points nobody can act on.

2. Resist the urge to slice every dimension at once. Cooper's own writing warns that splitting by project type, then product line, then market, then geography becomes "too complex and onerous" for most businesses to run in practice. Pick two dimensions, three at most. If you can't defend a split along a fourth dimension in one sentence, you don't need that dimension yet.

3. Run the gap analysis before you argue about percentages. Categorize your current, already-approved projects by bucket and total up what's actually being spent today. That's your "what is." Compare it against the strategic split leadership says it wants, your "what should be." That gap is what you bring into the room: if maintenance is running at 40 percent of spend and the strategic split says it should be 25, someone has to explain the extra 15 points or agree to cut it.

4. What happens once the buckets are sized? Projects compete only against other projects in the same bucket, using a weighted scoring model (a checklist of criteria like strategic fit, technical risk, and market size), expected commercial value (ECV), or another ranking method your business trusts. A maintenance project never has to out-argue a new-platform project for funding. It only has to beat other maintenance projects. That keeps political horse-trading from deciding your whole portfolio.

This works at smaller scale too. A ten-person IVD supplier can run the same logic with two buckets, growth and maintenance, tracked on one spreadsheet tab, settled once a year in a single conversation with the general manager instead of a formal scoring committee. The mechanism, decide the split before arguing project by project, holds regardless of company size. The paperwork is what shrinks.

The devil's advocate case: isn't "no formula" just an excuse?

Here's a fair challenge to all of this. If there's no formula, what stops "strategic judgment" from becoming a polite name for whoever argues loudest in the room winning the split?

Cooper's benchmarking research answers this directly, and the answer isn't comforting for anyone hoping structure alone solves it. His study of 205 U.S. businesses found that financial methods, the ones that look most rigorous on paper, are used most often but produce the worst outcomes for product portfolios: too many approved projects, poor strategic fit, and pipeline gridlock. The businesses that actually performed best used strategic approaches and scoring models together, and used them consistently across every project, not selectively when it suited someone's argument. Discipline, applying one method the same way every time, mattered more than which formula was used.

None of that tells you which method to pick. It tells you what to do once you have one.

The takeaway

Sizing your strategic buckets isn't a calculation you file away once and forget. It's an agreement negotiated under real ambiguity, and it only holds if someone keeps reviewing it through the year, not just at planning time.

The PM who handles this well isn't the one who finally finds the formula. It's the one who picks a defensible method, applies it the same way to every dimension, and owns the fact that judgment, and the work of maintaining it, was always the job.

Sources:

1. Robert G. Cooper, Winning at New Products (4th/5th ed.), pp. 272-273. Robert G. Cooper, Scott J. Edgett, and Elko J. Kleinschmidt, "New Product Portfolio Management: Practices and Performance," Journal of Product Innovation Management, 1999: https://onlinelibrary.wiley.com/doi/10.1111/1540-5885.1640333