Fewer offers, more focus: how to simplify the portfolio

A long catalog usually hides a short strategy. When a company sells dozens of products and none of them stands out, the most likely explanation is not that the market asked for all of them. It is that nobody in the organization dared to choose. Every product has a champion, every line has a history, and removing anything upsets someone. The result is a portfolio that grows by accumulation and a sales force that spreads its time across too many fronts.

Richard Rumelt, in "Good Strategy Bad Strategy" (2011), offers a way to see this problem clearly. For him, strategy means concentrating resources on a few points where an advantage exists, which forces saying no to everything else. Simplifying the portfolio is that idea applied to what the company sells.

When nobody chooses

Rumelt argues that bad strategy is rarely a simple absence of strategy. More often it is the result of actively avoiding the work of choosing. Complex organizations tend to spread resources around to satisfy every internal interest, because concentrating means leaving someone without what they asked for. That is why, according to the author, universal consensus around a plan is usually a sign that no real choice was made.

A long catalog fits this pattern. If everyone on the commercial committee agrees with the portfolio, it is worth being suspicious. Most likely, everything that existed was kept and everything each area proposed was added.

Rumelt also lists signs of bad strategy that are easy to recognize in a portfolio discussion: fluffy language that sounds sophisticated and says nothing, refusal to confront the underlying problem, mistaking goals for strategy, and objectives presented as endless lists of things to do.

The kernel applied to the portfolio

The center of Rumelt's book is what he calls the kernel of a good strategy: a diagnosis, a guiding policy, and a set of coherent actions. Applied to a portfolio, it looks like this.

Diagnosis. Why the current portfolio does not perform. Saying that sales are low is not enough. The task is to identify where effort is being dispersed: which products consume selling time without closing, which lines demand disproportionate technical support, which items exist only because one customer asked for them once.

Guiding policy. What kind of offer and customer gets priority, and what is excluded by definition. Rumelt compares it to guardrails on a highway: they do not say exactly where to go, but they do say where not to. A guiding policy for a portfolio might read, as a hypothetical, "we sell complete solutions to mid-sized customers in one sector, not loose components to anyone." The value lies in what the sentence rules out.

Coherent actions. Decisions that reinforce one another. This is where many simplification efforts fail. Focus gets announced in one meeting and, three months later, two new lines launch because a regional manager asked for them. Promising concentration while widening the catalog is incoherence, and the sales team notices it before senior management does.

A tool: the hump chart

Rumelt proposes a practical exercise. Assign the real profit to each unit of analysis (product, line, customer, or country), correcting the cost allocations that usually hide the truth. Then rank the units from highest to lowest profit and plot the cumulative total.

If the curve rises steadily, every unit contributes. If it rises and then falls, a hump appears: some units destroy value and others are subsidizing them. The peak of the curve shows how much the company would earn if it kept only what contributes.

The author tells the story of Denton's, a chain of garden nurseries. The chart showed that the best store earned more than the entire chain combined. Closing five stores and fixing eight doubled the profit. The same exercise applied to a catalog tends to produce similar results: a few items carry the business while many others quietly erode it.

Rumelt also cites Apple in 1997, when the company went from fifteen desktop models to one and cut distribution channels and inventory.

Two questions to unblock the meeting

When the discussion stalls because nobody wants to let go of their product, Rumelt suggests one question: if we could have only one objective, and it had to be achievable, which one would make the biggest difference. It forces comparison instead of addition.

The second is a test of real priority. If a product line were truly critical, it would have a single owner, an assigned budget, and clear consequences for missing targets. If it has none of the three, it is a wish, and it should come out of the plan.

A word of caution

Simplifying is not cutting for the sake of cutting. The hump chart only works if costs are properly assigned; with accounting that spreads overhead arbitrarily, the exercise can point at the wrong culprits. In addition, some small offers sustain relationships with large customers: a low-margin product may be the reason an important account buys everything else. Before removing anything, check what it takes with it.

To start this week

1. List every active item and ask, for each one, who defends it and why. 2. Build the hump chart with the best cost information available, even if imperfect. 3. Write the guiding policy as one sentence that explicitly excludes something. 4. Review the last three launch decisions and check whether they are coherent with that sentence. 5. Apply the real-priority test: single owner, budget, consequences.

A short portfolio is usually the visible trace of a decision that someone was willing to make.