Where to play: how to diagnose markets

Where to play is the choice of which markets, segments, channels and geographies a business will compete in, and which it will not. Diagnosing a market means breaking it into segments that genuinely differ in profit and working out where you make money today. Then each segment is scored for how attractive it is and for your ability to win in it, and a choice is made. Most businesses skip the diagnosis. They discuss the market as one number, assume they can win anywhere they show up, and call the result a growth strategy.

The diagnosis is not complicated, but it is work, and most of the work is in the first step. This page walks through it in the order it should be done.

Start by defining the segments

A segment is a part of the market where the economics differ from the parts around it. Segments can be cut by product, by customer type, by channel, by geography or by application. The right cut is the one where the boundaries separate different profit levels, different attractiveness or different competitive positions. If two "segments" have the same margins, the same competitors and the same buying behaviour, they are one segment.

Two rules make the segmentation usable. The segments must not overlap, and together they must cover the whole market, so that nothing is left out and nothing is counted twice. And the cut should be done once, tested against the numbers, and redone if the numbers say the boundaries are in the wrong place.

Example: A distributor describes its market as "building products". Its invoices say it is three markets: contract supply to large builders on tender, a trade counter for small builders, and online sales to owner-builders. The three have different margins, different competitors and different reasons for winning. One market has become three, and the strategy conversation has changed.

Know your own returns by segment before you judge the market

The second step is usually skipped and is the one that changes minds. Before scoring the market, work out what the business actually earns in each segment once the shared costs are allocated properly. That means sales effort, freight, warehousing, working capital, and the assets each segment ties up. This means going back to invoices and customer accounts rather than management reports, and deciding, segment by segment, how the indirect costs are shared.

The result is nearly always uncomfortable and nearly always clarifying. Revenue and profit sit in different places. The largest segment is often the thinnest. A segment everyone treats as a sideline turns out to be funding the rest. A business that has not done this cannot say where its profit pools are, which means its where-to-play choices are effectively guesses.

Example: The distributor's contract segment is two-thirds of revenue. After allocating tender costs, freight and the working capital tied up in ninety-day terms, it is a fifth of profit. The online segment, a tenth of revenue, is a third of profit.

Score each segment's attractiveness

Attractiveness asks whether a segment is worth competing in at all, regardless of who you are. Four things decide it.

Size and growth. Current volume and revenue, past growth over several years, and what drives demand. The growth driver tree is critical. Growth from a rising population is a different kind of growth from increasing penetration of a market or take-up of new users.

Why the segment is profitable, or not. This is the part most market reports skip. Profitability in a segment is explained by its structure. Is there a real basis for advantage between competitors? How high are the barriers to a new entrant, and how intense is the rivalry? Are substitutes close in price and performance? How much power do the buyers and the suppliers have? A segment where three customers buy most of the volume will have thin margins whatever the growth rate, because those customers set the price.

Risk and volatility. How stable the returns have been and what could shift them: a regulation, a technology, a change in a large customer's ownership.

Trends. Where the attractiveness is heading, not where it has been. New capacity from competitors, changing customer needs, and the drivers of demand slowing or accelerating.

Example: A segment is large and growing at twice the rate of the rest of the market. It is also one where the top three buyers account for most of the volume, tender every contract, and switch suppliers without cost. Growth is real; the benefit will go to the buyers. It appears to be an attractive market to sell into but is an unattractive one to compete in.

Score your ability to win in each segment

The mirror question: regardless of how attractive the segment is, can you win in it? Four things decide this too.

What customers in the segment buy on. Their selection criteria and the weight they put on each: price against delivery, against range, against technical support. This is learned by asking them, not by assuming.

How they rank you against the alternatives. On those same criteria. A business that believes it wins on service and has never asked its customers to rank it is guessing.

Your relative cost and asset position. Against each competitor in the segment. Which cost drivers matter: scale, plant, network, labour, working capital? Where does each competitor stand on them, and what does that imply for their unit costs against yours?

The basis of advantage you actually have there. The assets, capabilities and relationships you can lean on in that segment specifically. And the evidence that they work: share trend, price realised against competitors, retention, win rate. What counts as a real advantage is a page of its own.

Example: A regional manufacturer serves two states well because freight from its plant is lower than any competitor's. It has been losing tenders in a third state for five years and blames the sales team. Its freight to that state is higher than the local competitor's. It cannot win there, and no sales effort changes that.

Put the two scores together, and choose

With attractiveness on one axis and ability to win on the other, every segment lands in one of four places, and each place implies a different decision.

• Attractive, and you can win: defend and grow. This is the core, and it is resourced first.
• Attractive, and you cannot win today: enter or fix only with a credible plan to change your position, and a date by which the plan is judged. Otherwise stay out.
• Unattractive, but you win: harvest. Take the cash, invest the minimum, do not pretend it is a growth segment.
• Unattractive, and you cannot win: this is the exit shortlist. Nearly every business has segments here and keeps them out of habit.

Two things make this step valid. First, the where and the how are chosen together. A segment only becomes a real where-to-play choice when there is a credible how-to-win inside it. Picking the market first and hoping for an advantage later is the most common way the exercise goes wrong. Second, where to play is also where not to play. If the diagnosis has produced a list of segments to grow and nothing to leave, it has produced goals, not choices.

The five stages of understanding the market

How finely the market is understood is one of the practices in the facts-and-foresight dimension of a stages-of-excellence strategy assessment. The five stages read:

• Lagging: The market is discussed as one number; where profit actually pools is unknown.
• Basic: Market size and growth are known at headline level, from industry reports everyone else also reads.
• Competent: Growth and profitability are mapped at segment level, and the strategy names which segments matter.
• Advanced: Granular profit-pool and share data drive where-to-play, down to categories, channels or regions.
• Leading: Market granularity is a maintained asset that spots shifts before competitors' averages do.

Most businesses are at the second stage. The move to the third is the diagnosis described above, and it can be done from the business's own invoices and a dozen customer conversations.

Frequently asked questions

How granular should the segmentation be? Likely one level deeper than the business manages itself today. If it manages by product line, cut by product line and customer type. If it manages by state, cut by state and channel. Stop when the segments no longer differ in profit or position.

Do we need market research to do this? Not to start. Own invoices, allocated costs and customer conversations get a business to the third stage. Industry reports everyone reads are the second stage. Primary research earns its place when a specific question, such as what a segment buys on, cannot be answered from inside.

How often should the diagnosis be redone? Maintained, not rebuilt: updated once a year at least, and whenever a segment's structure changes. A diagnosis done once for a strategy offsite and never refreshed becomes history.

Does this apply to a group with several businesses? The same logic applies one level up. The group asks of each business how attractive its industry is and how strong its position. It adds a third question the business cannot ask of itself: is the group the natural owner, able to extract more value from it than another owner would?

What if the diagnosis says our biggest segment is the one we should leave? Then the diagnosis has done its job. The decision is still yours, but it is now a decision rather than a default, and the momentum case will show what leaving it to drift costs.

Do you know where your profit pools are? Take the check. · The stages of excellence in strategy · How to win: understanding your competitive advantage