The Market Segment Analysis Chart

Sometimes all you need to do is step back and put the right high-level data in one place so you can actually see what’s going on.

That’s the idea behind this post. The chart I’m going to show you is simple. Yet, despite spending many hours in boardrooms with executive teams debating segment strategy, I have almost never seen all the relevant data brought together in a single view. Lord knows I’ve asked for it.

What do I see instead? Seas of numbers. Live reports, spreadsheets, and dashboards. Slides with loads of dashboard clips – my favorite – especially when those clips come from different systems with different definitions and reporting periods. What should be a strategy discussion ends up a data forensics exercise combined with a quilting bee. (Hey, everybody, let's make an attractive quilt from all the dashboard clips people brought!)

So you end up spending the vast majority of your time navigating, explaining, and reconciling numbers before you can even start to discuss what they mean. You leave the meeting with your head spinning. You can’t see the forest for the trees. Heck, with some companies, you can’t see the forest for the leaves.

What’s needed is to put the relevant information together in one place, clearly and consistently, so smart people can sit down and talk about it. Note that “sitting around talking about numbers” is a concept I’ll explore more deeply in a future post – I'm a huge believer in doing that and a lot of companies don't do it enough. This post is about making a chart that enables that style of conversation.

Back to market segments, this chart helps answer two questions:

• Do we have the information we need to make the decision? If you can’t fill in some cells, leave the holes. Then ask what it would take to calculate the missing numbers and who’s going to get them. A half-completed chart tells you both what you know and what you don't know, but maybe should. Maybe the right answer is to send some ops and FP&A people off to fill in those blanks and reconvene next week instead.
• Have we presented that information in a way that helps people decide? Producing numbers is only part of the job. Helping people see the relatively small set they need to make an important decision is the other part.

I'm going to both present the chart and fill it in with some illustrative numbers. For analytical readers, I think this makes it more interesting exercise (i.e., can I unravel the mystery in the data?). But let's not get so wrapped up in the mystery that we forget the purpose of the post: if you're thinking about market segments, you should make a chart like this for your company and use it as the backdrop for a strategic conversation.

Our Example

While I’m going to use company-size segments, the same approach works for industries, geographies, countries, channels, or whatever segmentation matters to your business. Segment strategy is inherently dynamic. We start segments, invest in them, and hire leaders and build organizations around them. Eventually, we need to decide whether to keep investing, maintain them, fix them, or get out. By then, those decisions have become political.

Imagine a company with about $70M in ARR selling into three segments:

• Enterprise is the original business. The company was founded to serve large enterprises, and the segment is run by one of its earliest sales leaders. This person built a huge part of the business and still runs the segment responsible for more than half its ARR.
• SMB came later, after the board pushed for a PLG initiative. It never became truly self-service, but the company discovered a legitimate SMB value proposition, so the segment stuck around and eventually got its own sales organization. That organization is now run by a red-hot executive recruited from some super-hot company.
• Midmarket came later still. We saw an opportunity between SMB and Enterprise. The requirements, sales motion, and sales cycles were different enough that we created a separate organization. Its leader is a quieter executive whom we promoted from within.

So now we have three segments, three leaders, three histories, and plenty of reasons not to make an objective decision. Let’s look at what we’ve created.

Which Metrics to Pick?

The chart brings together fifteen measures. Most should be familiar, but a few distinctions matter:

• Size and growth. Ending ARR, percentage of total ARR, and year-over-year ARR growth show how big each segment is and how fast it’s growing. Include the percentages to save everybody the math.
• New ARR bookings and growth. How much new ARR did we sell into the segment during the period, including new-logo and expansion bookings? How fast is that number growing? This helps distinguish the size of the installed base from current selling activity.
• Customers and ARR per customer. How many customers do we have, and how much ARR does the average customer generate?
• Gross and net revenue retention. GRR tells us how much of the starting revenue base remains before expansion. NRR includes expansion. Put differently, GRR tells us how much is leaking out of the bucket; NRR tells us whether and to what extent expansion is refilling it.
• CAC payback period (CPP). How many months of gross margin does it take to recover customer acquisition cost? Calculating this by segment can involve some allocation work, but don’t let that stop you from making a useful comparison. [1]
• Average sales cycle and average deal size. How long does it take to close a deal, and how big is it? Average deal size, labeled ASP in the chart, is different from ARR per customer: one looks at recent deals, the other at the installed base. [2]
• Win rate. How often do we win? Pick a sensible definition and apply it consistently across segments. [3]
• Attainment. What percentage of reps reach at least 80% of quota? I particularly like this measure because a segment can make its aggregate plan while having an unhealthy sales organization underneath it. If three reps are carrying ten, I want to know.
• NPS. Do customers like us? Customer interviews and market research can give us a richer answer, but numerically I find GRR and NPS useful together. GRR tells us what customers do; NPS gives us a read on how they feel.

I’m not suggesting these fifteen columns come delivered on stone tablets. Depending on the situation, you might add pipeline, pipeline growth, CAC ratio, or bookings per AE. Or a dozen others. You might remove some columns, too. The objective is to assemble the relatively small set of measures to help you intelligently compare market segments.

What's Going in our Business?

Let’s take a quick pass through each segment. We’re looking for patterns across the measures and questions worth investigating.

• Enterprise is the largest business, but it has problems. It represents 51% of ARR, with customers averaging $837K. A six-month sales cycle seems reasonable for $720K deals, and the 28% win rate is respectable. But ARR growth is only 14%, CPP is 27 months, and 83% GRR means we lose 17% of the starting revenue base before expansion. NRR of 101% tells us expansion largely fills that hole. Only 30% of reps reach 80% of quota, which makes me want to understand pipeline distribution, ramp, quota setting; it looks like a handful of sellers are carrying the business. NPS is 21. Customers don’t hate us, but I’m not sure I'm going to market that on the website.
• SMB looks like a poor fit. It’s only 10% of the business, yet ARR growth is just 22%. The 38-day sales cycle is refreshingly short, but win rate is only 10%. More troubling is the combination of 20-month CPP and 80% NRR: we spend heavily to acquire customers whose revenue then shrinks substantially, even after expansion. GRR of 71%, NPS of 16, and weak attainment reinforce the concern. These problems extend well beyond something I’d assume our hotshot new sales leader can fix.
• Midmarket is working. It’s already 39% of the business and growing 40%, with bookings growing 35%. GRR is 94%, NRR is 115%, and CPP is 10 months. We close $221K deals in 74 days. With 88% of reps reaching at least 80% of quota, success appears broadly distributed across the sales team. NPS is 50. Customers stay, expand, and give us strong feedback, and we acquire them efficiently.

Here is the same chart with some highlighting added. Nothing changed in the underlying data; the colors simply make the differences easier to see.

My inclination would be to shift incremental investment toward Midmarket, investigate what’s wrong in Enterprise, and seriously consider exiting SMB. [4] Enterprise is still more than half the business, so I’m not suggesting we wake up Tuesday morning and fire the Enterprise team. I’m asking where the next dollar and the next headcount should go. The answer needn’t be wherever most of the ARR already happens to be.

And remember who runs these segments. The Enterprise leader helped build the company and isn’t going to love hearing that the next investment dollar should go elsewhere. Exiting SMB means questioning both the board’s PLG push and the decision to hire that hotshot executive. Meanwhile, the quieter leader promoted from within may have the strongest case for more resources and the least political clout to get them.

These are hard conversations involving loyalty, reputations, and previous decisions that people would prefer not to revisit. The chart won’t make the politics disappear, but it gives you a basis for challenging those arguments and makes it harder to allocate resources based on who built the business, who has the best résumé, or who talks the loudest.

Make the Chart and Talk About It

We could spend plenty of time debating what our hypothetical company should do. You might disagree with my inclination to exit SMB or have a different diagnosis of Enterprise. That would be a useful discussion, and the chart gives us a common basis for having it.

But remember why we did the exercise. The purpose of this post is to get you to put the relevant information together in one place. A chart like this is rarely made, even when the decisions it would support are important and most of the numbers already exist somewhere in the company.

Instead, growth is on one slide, retention is six slides later, sales productivity is in the CRO section, and NPS is in customer success. We leave the executive team to assemble the picture in their heads while debating organizational history, executive personalities, previous board decisions, and corporate identity. “But we’re an enterprise company” or “we just hired this hotshot to run SMB” can become surprisingly persuasive arguments when nobody has put the relevant numbers together.

Make the chart. Add columns, remove columns, and leave holes where you need more information. Then sit down and talk about it: what’s working, what isn’t, what you don’t understand, and what you should do. That’s the “smart people sitting around talking about numbers” part. We'll dive deeper into that in a future post. But, for now, let's give them something worth talking about.

The spreadsheet for this post is available here.

# # #

Notes

[1] Calculating CPP by segment can get messy because of marketing allocation. Corporate marketing, for example, supports multiple segments. Personally, I dislike allocation exercises, so I might use something closer to a sales CPP based primarily on direct selling costs and clearly attributable demand-generation expense. That won’t be comparable to a fully loaded CPP benchmark, but the goal here is comparison across segments. Use a consistent approach and be explicit about what’s included. You could alternatively use CAC ratio.

[2] In this example, SMB ARR per customer is $33K while average deal size is $36K. Enterprise ARR per customer is $837K while average deal size is $720K. Those differences aren’t inherently contradictory: recent deal sizes can differ from installed-base averages, and customers can expand after they land. Be clear about whether deal metrics cover new-logo sales, expansion, or both.

[3] There are several legitimate ways to calculate win rate. Pick one that suits the question and use it consistently. This exercise is about comparing segments, not achieving theological purity on win-rate methodology.

[4] Before shifting investment toward Midmarket, I’d want to know whether there’s enough market to support our growth ambitions. You could add SAM (serviceable addressable market), or better still for near-term planning, "market velocity" – an estimate of the number of relevant deals will happen over the next twelve months. There’s a big difference between 5,000 plausible transactions and 73. If that information matters and you don’t have it, add the column and leave it blank, so you'll know what homework needs to get done. Some market researchers will calculate this for you.