Determining the Number of Accounts per Customer Success Manager

This is otherwise known as, “How to Determine Customer Success Practitioner Coverage Ratios.”

Initially, the question was “how many accounts should a Customer Success Manager (CSM) handle?”

But people quickly realized that answers like “37 on the low end; 200 on the high end” weren’t actually helpful.

Then, an ex-CEO-turned-VC with a strong content marketing machine, said a different thing that has, unfortunately, stuck:

“1 CSM per $2M/ARR.”

That’s not accurate, it never was, and it needs to stop being propagated.

Here’s what to do instead…

By the way, this post assumes you have a clear understanding of Customer Success. If you have time, I’d suggest that you read The 8 Elements of Customer Success Management or, if you’re quite busy, read Customer Success Management: An Executive Overview.

Okay, so one of the main things that made that well-known-but-totally-wrong CSM coverage ratio obsolete is that…

Customer Success is not SaaS-specific

First, the fact that the Customer Success Manager (CSM) coverage ratio myth is $2M per ARR – Annual Recurring Revenue or Annual Run Rate – means that it is specific to Software-as-a-Service (SaaS) companies.

But Customer Success Management – and Customer Success in general – is certainly not limited to SaaS.

In fact, I believe the concepts behind Customer Success are even more applicable to – and important for – non-SaaS (think: transactional, non-tech) businesses that don’t have an ongoing, recurring relationship with their customers.

I’m going to work hard to ensure 2017 is the year Customer Success breaks through to the mainstream, but I digress.

So, beyond the fact that this myth is SaaS-specific, let’s go deeper by talking about how…

A Holdover from Traditional Account Management

As I laid out in my “9 things Customer Success is Not” post, Customer Success Management is not just another name for Account Management.

But this $-per-CSM model is a direct carry-over from traditional Account Management, which was all about treating customers like an account… a number. This literally means seeing customers as the revenue they give us and nothing more.

Segmenting customers based on how much they pay you is one of those traps that a lot of Customer Success organizations fall into, mostly because it seems logical, and it’s what the industry has been doing for a long time. But this, too, is a legacy of traditional Account Management.

In my piece on Logical Customer Segmentation, I describe in detail why you shouldn’t segment based on what customers pay you – which is the core of the traditional Account Management approach – and why, if you’re applying that model to Customer Success Management, you’re doing it wrong.

So if the $-per-CSM model is a carry-over from traditional Account Management, too, it should also go away.

Instead, you need to look at…

The Customer Success Manager Coverage Model Formula

Don’t try to come up with “one CSM per $xxx/MRR,” as that doesn’t fit a modern, segmented coverage model.

In fact, if you’re looking for a generic indicator of how many CSMs per Customer or per Dollar of Revenue you should have, there isn’t one.

It doesn’t – and can’t – exist since it’s 100% based on your unique situation.

Trying to normalize that way doesn’t take into consideration that you will have different types of Customer Success Practitioners (CSPs) working with different customer segments, not to mention the Customer Success Operations and Support team members you’ll need to have in place.

Rather, you need to do the following:

  1. Logically segment customers based on their Appropriate Experience
  2. For each logical Customer Segment, discover what coverage level and type of coverage is required (from the customer’s POV; not based on what you can or want to deliver)

Now, maybe you discover (discover is the key word here; not dictate) that one segment requires that you provide a high-touch, consultative experience.

You need to figure out the details of that required coverage model and the characteristics of the CSPs that will cover these customers, and from there you can see how many accounts each CSP at that level will be able to handle.

If it’s 5 accounts per CSP, and you will have 100 accounts in this segment in 6-months, you’ll need to hire 20 CSPs.

Over time, you’ll be able to optimize even the most high-touch coverage levels to eek out more efficiency, so maybe you shoot for 17 to cover 100 customers in 6-months. Maybe at some point you’ll get to a 1:10 CSP:Customer ratio, but not out of the gate.

Do the same for each of the other segments.

Also take into consideration supporting CSPs (Ops, Analysts, etc.), whether those are there at the start or you bring them in at a certain level (for example, you bring in an Ops person after you have x customers across y segments). Figure that into your fully-loaded Customer Success Management cost.

After you go through that discover process you realize you can’t afford to deliver the level of coverage for a particular customer segment – let’s say 1 CSP per 5 customers isn’t economically feasible, or the required coverage levels amount to 30% of revenue – you have two choices:

  1. Look at that spend as an investment in the process and try to get to a ratio – while delivering the appropriate customer experience – that makes economic sense at some point in the future, or
  2. Don’t do business with customer segments for whom you can’t provide an appropriate experience in an economically viable way

Many companies will go for a 3rd option – which doesn’t or shouldn’t actually exist – and that is to reduce the coverage level to a point that makes economic sense (for the company) but likely doesn’t provide the customer with an appropriate experience.

That’s something, but it’s not Customer Success.

Okay, so that’s how you determine the appropriate coverage model, and why you should look at…

Customer Success Management as a Percentage of Revenue

Just like Marketing or Sales, you should look at Customer Success as a percentage of revenue.

Industry surveys have shown that the fully loaded cost of Customer Success Management is anywhere between 15% and 25% of annual revenue.

My biggest problem with industry surveys focused on newer and rapidly evolving areas of innovation (like Customer Success Management) is this: those who are doing it best may not even be represented in the survey!

And if they are, their results are either removed for being outliers or are offset by the poor results of everyone else.

So fully-loaded Customer Success Management costs falling between 15% and 25% of revenue sounds plausible with a few caveats.

  1. No one can agree on what “fully loaded” means (Salaries? Infrastructure costs? What else?)… the more you figure into this calculation, though, the closer you’ll get to reality. By the way, I assume the rather large range of 15% to 25% of revenue is due, in part, to the lack of agreement on what “fully loaded” means.
  2. Don’t include too many things, though… fully loaded Customer Success Management costs likely shouldn’t include Customer Service or Support costs (those often sit under Operations or a different part of the business and are – while important to the success of the customer – not direct Customer Success Management costs).
  3. Should you include the cost to support fee-based Customer Success Management services, such as expedited onboarding, implementation, consulting, etc.? If those resources also contribute to the included Customer Success Management services, then yes; if not, it depends.
  4. For a lot of companies – and this will likely vary across different customer segments, too – Customer Success Management costs are heavily front-loaded (Onboarding, Implementation, Training, etc.), so the percentage of first-year revenue will be higher than the percentage of revenue in the ensuing years. Of course, the heavily front-loaded nature of Customer Success Management is often attributed to the fact that companies simply haven’t operationalized across the rest of the lifecycle!
  5. If you’re taking advantage of true Customer Success-driven Growth, your customers should be expanding their relationship with you over time, meaning your customers will pay you more (through upsells, cross-sells, etc.), but the required human coverage levels may actually go down, meaning the percentage of revenue you’ll spend on Customer Success Management will decrease over time.
  6. For companies that are just getting into Customer Success Management, you should probably over-invest right now; it’s about learning, not about profitable unit economics. If you’re anywhere close to 15% of revenue at this point, you’re WAY too low… increase spending to increase learning
  7. Maybe 15% to 25% is accurate, so keep that number in mind, but know that it’s likely to evolve as the industry learns more. Also, remember that it may never, ever apply to your business.

Every other part of a (mature, well-run) business applies a “fully-loaded-cost-as-a-percentage-of-revenue” model, not a staff-per-customer or staff-per-revenue number.

Why should Customer Success be different?

About Lincoln Murphy

I am a Customer Success Consultant focused on Customer Success-driven Growth. I wrote the Customer Success book which you can buy at Amazon. If you need help applying Customer Success-driven Growth principles in your company or would like me to speak at your event, please contact me. Also, connect with me on LinkedIn or follow me on Twitter or Facebook.

Comments

  1. Brooke Goodbary says:

    I think a major constraint you overlook is limited budgets and headcount. Most SaaS companies (which, for now, is where Customer Success is most commonly found), have limited resources and will struggle to implement a cost as a percentage of revenue model because they haven’t realized the ARR of these customers yet. For this reason I think the critical first step is for a company to establish the level of service they expect to offer their customers. It’s a simple question that leads to much broader discussions around what are realistic expectations of the Customer Success team. Where can we comfortably accept a 1: many experience or leverage automation to maximize the efforts of a small team? Once a company can agree on the experience CS will provide, the team can get started weighing different account distribution techniques. I agree that segmentation is certainly the starting point to determining what level of service you can reasonably expect to offer each customer.

    CSMs should also be able to prioritize their accounts on a few different axis- such as which customers have a strong potential to upsell in the next 3 months, stakeholders who might be able to write a good customer story, etc. Having this priority list in the back of a CSM’s mind will help them better manage clients’ expectations and make sure they’re dedicating the appropriate time to their highest value or highest potential accounts. I would argue that a customer that is only paying $1000/mo, represents a low upsell potential, and has a well-established usage record writes in asking for a 1:1 training for someone new on their team, CSMs should feel confident directing them towards public training resources.

    The final factor in account segmentation/ prioritization is being able to identify which customers are at risk of churning. CS teams should consider creating a customer health score to highlight which accounts have exhibited troubling behavior that make them a good candidate for a personalized outreach campaign.

    • I appreciate your efforts to influence me, but this isn’t a battle of opinions… I’m working from experience across many different companies in several countries. And blasting your thesis at me across multiple channels doesn’t make the things you say true.

      If you start with what you (as a company) can or want to provide the customer, you’re starting in the wrong place.

      Start with the customers, logically segment, understand each segment’s Appropriate Experience, and then you’ll know the level and type of coverage you need to provide.

      If you cannot provide the Appropriate Experience for those customer segments because you can’t afford it, don’t do business with that customer. They’re a bad fit for you until you can provide them with their Appropriate Experience.

      That said, I do agree with you that there is further segmentation required based on the customer’s status, health, Success Vector, etc. But that’s a different type of “segmentation” (I refer to those as cohorts for clarity vs. Logical Segments) and is therefore out of scope for this particular post.

  2. Awesome post! We’ve been using this model on this year’s budget. We’re aiming to invest 16% of the revenue on CS. The devil is on the details, doing all the math over what the CS cost means to our different customer segments isn’t easy, but doable. Sometimes CS investment is a CSM, sometimes is an expensive LMS system for instance.

Speak Your Mind

*