Growth Strategy Framework 2026: Direction, Then Route

The growth strategy framework, in three decisions

The sections below hold every part of a framework and the page never named it as one. Here it is, three decisions in a fixed order, each with the test that closes it and the thing that goes wrong when it is skipped.

The order is what makes it a framework rather than a checklist. Decision 2 cannot be answered without decision 1, because a losing position in the core changes which cells are worth opening. Decision 3 cannot be answered without decision 2, because the capability gap is what determines whether building is even available.

There is a fourth thing, and it sits outside the sequence because it is a question rather than an analysis. Before decision 3 you have to know the hurdle: a payback limit, a return on capital floor, a required market share inside a window, or a first year breakeven. The firms that publish growth cases use all four and none of them agree, which the closing section documents. A route ranking computed against the wrong hurdle is arithmetic in service of the wrong question.

Applied to the pet food client running through this page, the three decisions resolve in order. The market grows 6.7 percent and the client grows 1.5 percent, so the ratio is below one and the core is losing share. Existing brands into new channels is open and a new wet food line is open, while both at once is the fourth cell and gets named and dropped. And the three routes price at $3.20, $7.78 and 68 cents of capital per annual contribution dollar, which is a 4.7 times spread between the cheapest and the build case.

Two questions inside a growth strategy mandate

The first question is direction. Which products, which customers, which geographies, and whether the company is defending a position or opening a new one. This is where structures are strongest, because the four cell grid of existing and new products against existing and new markets is well drilled.

The second question is route. Build the capability, buy a company that already has it, or contract with someone who has it already. That choice does not change the revenue on offer. It changes the capital at risk, the time to revenue, and the margin the company keeps, and those three things decide whether the growth is worth having.

A route comparison needs a hurdle before it can start. Oliver Wyman's published case shows why. Its model answer turns down a profitable expansion because it "does not align with the client's investment criteria for a five-year payback period", checked 29 September 2026 at oliverwyman.com. The business made money and the answer was still no, because the client underwrote on time rather than on profit.

So the opening move is a question. Ask what the client uses to approve capital: a payback limit, a return on invested capital floor, or a required share of a market inside a fixed window. Each of the three produces a different ranking of the same three routes, and picking one silently is how a candidate answers the wrong question correctly.

The four cells a growth strategy chooses between

The direction question deserves a fast, disciplined pass, because it sets which routes are even available.

1. Existing products into existing markets. Share gain and category growth. Needs no new capability, so the route is almost always build, meaning more sales coverage or more capacity.
2. Existing products into new markets. Geography or channel. Capability gaps are distribution and regulatory, which are the two things partners supply best.
3. New products into existing markets. Portfolio extension. The capability gap is manufacturing or technology, which is what acquisitions are usually bought for.
4. New products into new markets. Two unknowns at once. The honest treatment is to say that the risk multiplies rather than adds, and to ask why the client is considering it.

Take those four cells to a case. A regional pet food maker turns over $300 million and its chief executive wants an extra $100 million of revenue inside three years. Ask which cells are open and the answer narrows the routes immediately. Existing brands into new retail channels is a partnership question. A new wet food line is a build or buy question. Trying both at once is the fourth cell, and it should be named and set aside.

The cell choice also sets the margin you can expect, which matters more than the revenue. Existing products into new channels usually arrive at lower margin, because the channel takes a cut and the volume is price led. New products into existing markets usually arrive at higher margin, because the brand and the customer relationship are already paid for. That asymmetry is why a revenue target alone is an incomplete brief.

Build, buy or partner, priced on capital

Now price the three routes for the pet food client. Every figure below is stated so the comparison can be rebuilt.

Build means a new wet food plant. Capital is $80 million all in, the line reaches full output in three years, and contribution margin on the new revenue is 25 percent. At $100 million of revenue that is $25 million of annual contribution.

Buy means acquiring a rival with $90 million of revenue at 1.4 times revenue, so $126 million of purchase price. The target runs an 18 percent contribution margin, which is $16.2 million a year.

Partner means a private label contract with a grocery group worth $110 million of revenue at an 8 percent contribution margin, which is $8.8 million a year. Retooling two existing lines costs $6 million.

Read the fourth column before the last one. Building costs $80 million divided by $25 million, which is $3.20 of capital for each annual contribution dollar. Buying costs $126 million divided by $16.2 million, which is $7.78. Partnering costs $6 million divided by $8.8 million, which is 68 cents, and the payback is roughly eight months.

Then read the trade the partnership makes. It delivers $110 million of revenue, more than the target asks for, and $8.8 million of contribution, which is 35.2 percent of what building delivers. A client measuring itself on revenue will love it and a client measured on operating profit will not. Naming that split, with the percentage, is the move that turns three numbers into a recommendation.

Three years is a constraint, and it reorders the routes

The brief said three years, and a payback figure ignores the timing of the cash inside that window. Run the cumulative position instead.

Building ramps. Assume the plant delivers 40 percent of target revenue in year one, 70 percent in year two and 100 percent in year three, so cumulative revenue over the window is $210 million. At 25 percent that is $52.5 million of contribution, against $80 million of capital spent up front, so the three year position is minus $27.5 million.

Buying delivers immediately. Three years at $90 million is $270 million of revenue, and at 18 percent that is $48.6 million of contribution against $126 million of purchase price, so the position is minus $77.4 million.

Partnering also delivers immediately. Three years at $110 million is $330 million of revenue, and at 8 percent that is $26.4 million of contribution against $6 million of retooling, so the position is plus $20.4 million.

Inside the stated horizon only one route is cash positive, and it is the one with the worst margin. That is an uncomfortable answer and it is the right one to give, because the constraint the client imposed is what produces it. If the chief executive will extend the window to six years, building wins comfortably, and saying so gives the client something to decide.

The general lesson transfers to any route comparison. A payback number ranks routes on steady state economics, and a cumulative cash number ranks them on the horizon the client actually has. Ask which one the decision is being made on before you present either.

What the combination would have to deliver to rescue the acquisition

An acquisition case is usually defended by promised gains from combining the two businesses, so put a number on what those gains would have to reach.

The build route pays back in 3.2 years. For the acquisition to match that, $126 million of capital needs $126 million divided by 3.2, which is $39.375 million of annual contribution. On $90 million of target revenue, that is a contribution margin of 43.75 percent.

Nothing in packaged pet food operates at 43.75 percent of contribution. So the acquisition does not fail because the integration is hard or the culture is different. It fails on arithmetic, at the price offered, and the only two variables that could change the answer are the multiple paid and the revenue base bought.

Work those instead. At 1.0 times revenue the purchase price is $90 million, and lifting the target's margin from 18 to 25 percent gives $22.5 million of contribution, so payback is $90 million divided by $22.5 million, which is 4.0 years. That is within reach of the build case, and it tells the client exactly what to negotiate: a multiple at or below 1.0 and a credible path to the acquirer's own margin.

This is a more useful discussion of combination benefits than a list of cost lines. It states the threshold, names the two levers that reach it, and leaves the client with a walk away price. A candidate who instead lists procurement savings, distribution overlap and shared overhead has said nothing a chief financial officer can act on, because none of it is sized.

Share loss inside a rising revenue line

One branch belongs in every growth structure and is usually missing. Whether the market itself is growing.

The Census Bureau reported retail and food services sales of "$773.9 billion" for August 2026 against "$764.5 billion" in revised July, a rise of 1.2 percent on the month, checked 30 September 2026 at census.gov. The same release states that it "no longer contains the most up to date estimates", with revisions issued on 28 September 2026, so it should be quoted with that caveat rather than presented as current.

Use a market growth rate of 6.7 percent a year, which is the total retail figure published alongside the online series. Our client grew 1.5 percent. Relative share moved by 1.015 divided by 1.067, which is 0.9513, so the company lost 4.9 percent of its share while reporting growth.

That calculation reframes the whole mandate. The chief executive asked for $100 million of new revenue, and the diagnosis says the company is already losing position in what it has. Growth that comes from a partnership at 8 percent margin will not fix a share problem in the core, and it may hide one for two years.

Say this before the route comparison. A candidate who opens with market growth against client growth has bought the right to spend the rest of the case on routes, because the direction question is now settled with a number instead of a guess. One honest caveat belongs with it: the Census release covers total retail and food services, and no pet food series appears in it, so the 6.7 percent is a stand in until the client supplies its own category data.

Using the growth strategy framework when no numbers are given

Every figure on this page arrived because someone asked for it. In a live case the first ninety seconds have no numbers in them at all, and that is the part a framework has to survive.

Say the three decisions out loud, in order, before asking anything. It takes about twenty seconds and it does two things: it tells the interviewer the route branch is coming, which is the branch that usually goes missing, and it earns you the right to ask for data rather than guess at it.

Then ask for the three numbers that close the three decisions, one question each.

1. For the market. What has the category grown at over the last two or three years, and what has the client grown at over the same period. Two numbers, one ratio. If the interviewer has no category figure, ask whether the client has gained or lost shelf space, which is the same question in a form a client always knows.
2. For the direction. Which products and which markets are genuinely available, and what the client already has in each. The useful follow up is what capability is missing, because the answer to that is what decides the route.
3. For the route. What the client uses to approve capital. Ask it plainly: is there a payback period, a return threshold, or a share target. This is the question candidates skip and it reorders the entire answer.

Three more questions earn their place once the routes are on the table, and each one is a number that moves a ranking: the contribution margin on the new revenue, the ramp profile if anything is being built, and the horizon the decision is being made over. The pet food comparison here flips on that last one. On steady state payback building wins against partnering, and inside the three year window the client actually imposed, only partnering is cash positive.

If the interviewer supplies nothing, say what you would need and what you would assume, then carry the assumption visibly through the arithmetic. An assumption named out loud and used consistently scores; the same assumption used silently reads as a number you invented.

Where a route answer loses its points

The weights here are ours, lifted from the grading code in src/lib/grading, and no consulting firm publishes a comparable scheme. The argument for each weight is on how an answer is assessed.