BCG Matrix Product Prioritization: A Small-Business Guide

The BCG matrix sorts each product or service line by two numbers: how fast its market is growing and how strong your share is compared with the biggest competitor. Stars get investment, cash cows get protected and harvested, question marks get a time-boxed test, and dogs get fixed, folded into something else, or cut. For a small business, the real value is not the labels. It is being forced to make one explicit funding decision per product line instead of spreading effort evenly.

What the BCG matrix actually measures

The Boston Consulting Group matrix is a 2×2 grid built in the early 1970s for large corporate portfolios. The vertical axis is market growth rate. The horizontal axis is relative market share, which means your share divided by the share of your largest competitor. A relative share of 1.5 means you are 50% larger than the next player. A relative share of 0.3 means the leader is more than three times your size.

Two assumptions sit under the grid:

• Growing markets consume cash. You need inventory, ad spend, hires and tooling to keep up.
• Larger relative share produces better margins. Scale tends to lower unit costs, so the leader in a segment usually earns more per sale.

Put those together and you get four boxes.

The original model uses roughly 10% annual market growth and a relative share of 1.0 as the dividing lines. Treat those as starting points. In a slow industry like commercial cleaning, 6% growth may count as "high."

Why it still works for a small business

Most owners fund products by habit, loudness or founder attachment. The oldest offer keeps its budget because it always had one. The newest idea gets attention because it is exciting. Nobody compares them side by side.

The BCG matrix makes the comparison unavoidable. It also forces you to separate two things that get blurred in daily life: where the cash comes from and where the cash should go. In a business with 5 to 15 people, those are rarely the same line, and the gap between them is where good decisions live.

It is not a perfect tool. It ignores synergies, strategic intent and capacity limits, which I cover in the mistakes section. But as a 90-minute first pass, it beats a gut-feel budget meeting.

Step-by-step: build your own matrix

You need a spreadsheet, last year's sales data and a rough view of your market. Block out two hours.

1. List your product lines

Aim for 5 to 10 lines. Group by how customers buy, not by SKU. "Annual maintenance contracts" is a line. "Contract tier B for plumbing clients" is too granular. If you have 40 SKUs, cluster them by customer and job.

2. Pull revenue and gross profit for each

Use the trailing 12 months. Include direct labor if it varies with volume. A line that makes $200K but eats 1,500 hours of staff time is a different animal from one that makes $200K in 300 hours.

3. Define the market for each line

This step decides whether the exercise is useful. Define the market the way your customer would. "Specialty coffee subscriptions sold online to US consumers" is a market. "Food and beverage" is not. If you define it too broadly, everything looks like a dog. Too narrowly, everything looks like a star.

4. Estimate market growth

You will not find perfect numbers. Use trade association reports, public company filings from larger rivals, search trend direction, supplier volume changes, and what your own vendors tell you. Sort each market into growing above about 10%, or below. A defensible estimate is fine.

If you need help sizing up rivals, the competitive analysis guide covers cheap ways to gather this.

5. Estimate relative share

Find the largest competitor in each market niche. Estimate their revenue from review counts, headcount on professional networks, pricing and volume signals, or direct conversations with customers who use both of you. Divide your revenue by theirs. You only need to land on the correct side of 1.0 and know roughly how far from it.

6. Plot and label

Put each line on the grid. Make circle size proportional to revenue. Mark profit margin next to each circle. The visual often shows something the spreadsheet hid, such as your largest circle sitting in the dog quadrant.

7. Assign a decision and a budget

Every line gets one of these: invest, harvest, test with a cap, fix, or exit. Each decision gets a number: dollars, hours, or both. A matrix without budget changes is a poster.

8. Set a review date

Put a calendar block six months out. Markets move, and a star can age into a cash cow faster than the paperwork suggests.

Worked example: a 14-person coffee roaster

Here is a realistic case. A roaster with a café has $1.8M in annual revenue and 14 employees. The owner feels stretched and wants to know where to put the next $60K of spend and a new hire.

Gross profit by line is about $274K for wholesale, $173K for subscriptions, $41K for the café, $32K for cold brew and $11K for merch. Total is roughly $530K.

Reading the grid:

• Subscriptions are the star. Market growth is 22% and the roaster leads its local niche. Only 20% of revenue, but 33% of gross profit.
• Wholesale is the cash cow. Biggest line, mature market, strong position. It funds everything else.
• Cold brew is the question mark. The market is hot, but at 0.2 relative share, this roaster is a small player against beverage brands with distribution.
• The café and merch are dogs. Slow markets, weak share, thin margins.

The decisions:

1. Subscriptions: invest. Move $40K of the $60K here. Spend it on a retention fix (the box arrives late 1 in 8 times) and on a second paid acquisition channel. Target: 22% to 30% of revenue within 18 months.
2. Wholesale: harvest and defend. No new spend. Raise prices 3% at renewal, trim the three biggest discounts, and protect the two accounts that make up 35% of the line. Expect about $20K in added margin.
3. Cold brew: capped test. $15K over 90 days with explicit exit criteria: at least 3 retail accounts reordering and gross margin above 30%, or the line stops. Without those numbers, the roaster drifts into funding it forever.
4. Café: reshape, don't cut. On paper it is a dog. In practice it is the showroom that sells 40% of new subscribers their first bag. Cut Monday and Tuesday hours, saving about $28K in labor, and keep the rest.
5. Merch and equipment: cut. Stop reordering, sell down the $22K of inventory, and free roughly 4 hours a week of the manager's time.

The new hire goes to subscriptions customer support, not the café. Total plan cost: $60K of spend plus the hire, offset by roughly $48K in margin gains and savings. The owner had been about to split that $60K evenly across all five lines.

Common mistakes and how to avoid them

Treating the label as a verdict

The café above shows the problem. A dog can be a loss leader, a customer acquisition channel, or the reason your best clients stay. Before you cut anything, ask three questions: Does it feed another line? Would key customers leave without it? Does it carry fixed costs other lines would then absorb? If the answer to all three is no, cut it without guilt.

Drawing the market too wide or too narrow

If you call yourself a "marketing agency" competing against global networks, every line looks like a dog. If you define a market as "vegan gluten-free wedding cakes within 10 miles," every line is a star. Use the market as your customer sees the choice. A buyer comparing three local roasters is not comparing you with a national grocery brand.

Using your own growth instead of market growth

A line growing 30% in a market growing 35% is losing ground. A line growing 5% in a market that is flat is winning. The vertical axis measures the market, not your trend. If you only have your own numbers, you are guessing about the wrong variable.

Starring everything

Owners tend to see the whole portfolio as promising. Real capacity forces trade-offs. If three lines are all labeled stars and you can fund one, you have an unresolved argument about positioning, not a good problem. Rank them by margin and capacity fit, and pick one. The cognitive biases article explains why we resist this, especially sunk-cost attachment to older lines.

Filing it after one meeting

A matrix is only useful if budgets, calendars and hiring plans change. Attach every decision to a number and an owner. Then check it on the date you set.

Pairing the matrix with other tools

The BCG grid answers "where is each line relative to the market?" It does not answer "what should we do about it in detail?" For that, pair it with something else:

• For an expensive bet like the cold brew test, run a short cost-benefit analysis before committing.
• For choosing how to grow a star, such as new customers versus new products, the Ansoff matrix gives you the options.
• For a line that is shrinking and you don't know why, a root-cause method like 5 Whys is better than guessing.

Where the matrix falls short

Be honest about the limits. It assumes market share drives profitability, which is weaker in service businesses where reputation and relationships matter more than scale. It struggles with new categories where no market data exists. It says nothing about your team's skills or your capacity. And it treats lines as independent, when most small-business offers are tangled together.

Use it as a prompt for a decision, then apply judgment. If a line's position on the grid surprises you, investigate. That surprise is where the value is.

Key takeaways

• The BCG matrix sorts product lines by market growth and relative share, which gives you a default decision for each: invest, harvest, test, or fix and cut.
• You do not need perfect data. Define each market the way your customer sees it, and estimate whether your relative share is above or below 1.0.
• Look at gross profit and hours consumed, not just revenue. A line that is 20% of sales can be a third of profit.
• Cap every question mark with a budget, a deadline and exit criteria, or it will absorb cash indefinitely.
• Check whether a dog supports other lines before cutting it. Reshape it if it earns its place as a channel.
• Attach dollars and owners to each decision, and review the matrix every six months.