Customer Retention Cost Ratio

Live from Schenectady, New York, it's SaaS Talk with the Metrics Brothers, Growth and CAC.

And I'm Growth, better known as Ray Reich, the founder and CEO of Benchmarkit.

And I'm CAC, better known as Dave Kellogg, EIR at Balderton Capital, independent consultant, and the author of Kellogg.

And together we are the Metrix Brothers.

And we go together like autumn and leaves.

Ah, you saw my LinkedIn post on being in upstate New York last weekend, huh, Dave?

Yeah, that's one of my favorite things about fall in the Northeast. I'm going to be, I'm going to be up in Vermont in about a month, right? It'll be too late by then.

Oh yeah, I can't believe I said this, but I told my wife that this weekend in the mountains of upstate New York with all the beautiful color of the leaves. It was as beautiful as anything I've seen in California, Dave.

I might take exception to the term mountain, but I'm with you, given that the parking lot at most California ski resorts is higher than Mount Marcy, the highest point in New York. But I'm with you. It is beautiful. I'm from New York, as you know, and I love it up there. And I love— you're falling for the Northeast there? I'm from there.

Well, no, I'm falling for the Northeast. And I think we should just leave this line of discussion to another conversation and move on to today's episode, which is customer retention cost and the customer retention cost ratio. Okay.

Or we could thread the needle for the things that don't fall, the pine needles. Get it, Ray? But yeah. Okay. Well, this is gonna be an interesting topic, Ray, because this is not widely established as industry benchmarks. I know you have thoughts on I know I don't agree with all of them, so, uh, let's go mix it up.

Well, in fact, Dave, the reason I did it was I had 2 different VC firms and their operating partners reach out to me in the last 2 weeks asking if I had benchmarks for customer retention costs. And since I had 2 VCs reach out, I'm like, ah, they're starting to look at this as a metric and I don't have it.

Yeah, I think that's a very important point. It may even be the highest point of the episode, so we'll see. But if people are starting to ask about this, if you're in the audience, get ready, uh, because if they're asking Ray now, they might be asking you in a month. And it's not surprising, given the kind of frugal environment out there, that people are examining the cost of everything, including the cost of customer retention. And I presume you told those VCs there's just not a lot of benchmarks out there on this.

That's exactly what I did. But before we get into it, let's hear a word from our sponsor, and then we'll jump in, Dave.

Let's do it.

The Metrics Brothers is proudly sponsored by Benchmarkit. The B2B AI and software industry's largest, broadest, and most contextualized benchmarking index. Dave Cat Kellogg and Ray Growth Wright have put the Metrics Brothers podcast on hiatus for remainder of 2026. All episodes we have ever recorded are still available on your favorite podcasting app. You can reach Dave at Kelblog and Ray. on LinkedIn @RayReich. Thank you so much for listening to the Metrics Brothers podcast. Okay, this is going to be a fun one because it's not that well established, and you and I have both maybe have a little bit different thoughts on it. But do you want to start with what I believe customer retention cost is, Dave?

Yeah, why don't you start with that? Unless you want me to tell you what you think it is. Uh, I got it.

So For me, the customer retention cost is all those expenses that are incurred to service existing customers in the pursuit of retaining those customers and the associated ARR. And just like customer acquisition costs, people will often take that customer retention cost, divide it by the number of customers, and then that will be their total customer retention cost per customer. So, that's what it is at a high level, Dave, from my perspective.

So, are you talking about cost to renew? Like how much the renewals teams cost to renew a dollar of ARR?

That's a great question, but I have 2 different metrics, which, you know me, sometimes I don't like simplicity.

More Ray. Ray Moore's more Reich. Go ahead.

I have a customer retention cost ratio, which is all those costs that we invest to retain a customer against the total ARR base. that we're trying to retain. But then I have the customer renewal cost ratio, which is only those costs that explicitly are allocated to the pursuit of renewal measured against the total ARR that actually is available and gets renewed in that period.

So my first tendency would try to be have one benchmark for the renewal concept as opposed to 2, but, but that, that's my guess, and you've done more work here than I have. Why do I need the first one? Why do I need this one you just talked about, customer retention cost, which is all costs incurred to service existing customers in pursuit of, uh, retaining them and their ARR? Why do we— before we talk about what exactly goes in it, why are we measuring it?

So 3 reasons in my mind. Number one, I had 2 investors call me and they're looking at it, so I'm like, yeah, this is something investors are looking at. But now on a more serious note, the second one is a lot of private equity companies, and I've had a couple companies acquired by private equity. They want to know how much cash they could generate from the existing customer base, even if they didn't invest in sales and marketing, Dave. So that's all those costs that it takes to service a customer and retain it and retain it. And I'm not talking about the COGS. Not the cost to deliver the product, but all those costs to retain and then renew a customer. And the third reason is, I believe if you're going to try to make all your operations more efficient, just like we're trying to make customer acquisition more efficient and expansion more efficient, we need to know how much we're spending on retention-specific activities so we can then over time try to improve and make it more If you don't measure it, how are you going to improve it?

All right.

So, so when you— the first thing, this reminded me of an age-old exercise that people do with SaaS companies, which is they try and strip them in 2. They try to strip apart and you basically have one P&L and you pull it into 2 P&Ls and one is supposed to be the recurring business and the other is the customer acquisition business. And, and basically any SaaS company is some arbitrarily weighted, arbitrarily weighted average of those 2 that there's actually, wait a minute, there's, there's 2 businesses underneath there. When you look at a SaaS P&L, some would argue you actually need to strip it apart to understand it because it's this kind of random weighting of 2 very different businesses. One is in the business of keeping the lights on, keeping the customer base customers, and the other is in— and by the way, that business tends to be very profitable, and the other business is very unprofitable, which is in the business of acquiring customers. So I feel like you're circling around kind of an age-old topic in SaaS, which is what I'm just going to call the stripped P&L, but you're trying to do it with a single metric here, really.

That is correct. And one that operators can use to kind of evaluate how efficient they are at retaining and then renewing customers versus the private equity companies just want to know how much cash I can fly it, throw off, if I can reduce a lot of my SG&A costs and not go after growing new customers, but just retaining my existing customers.

Well, in defense of my private equity brethren and employers, because I do work with some PE firms, some of them like to expand. They definitely like a very profitable recurring business. But the ones I've worked with in general try and take one growth shot. They try and do— I say the general PE formula in my experience is maximize profits on the recurring business, try and grow it, market share in your existing market as kind of the baseline. And if that works, they still get their investment. And then for upside, if they can open one new market adjacent to it, like go from credit unions to banks or go from healthcare to pharma, if you can get one hop, Then you get the rig double bonus and you get the 5x instead of the 3x. But, but, but this is all part of the PE playbook. I agree with you.

And I just want to clarify why I have that perspective, because when I was acquired twice by PE, but once we were the tuck-in, we weren't the platform that was being used for consolidation. We were required to be the tuck-in. So guess what they were going to eliminate? All SG&A costs. They weren't going to aggressively acquire new customers, but they really wanted to hold on to our existing customers.

Yeah, and my, my guess is they're going to try and sell your stuff to, to their existing customers, right? Um, as well. So yeah, yeah, okay, that makes sense. Uh, on a tuck-in play, I'd agree with that. Okay, so Ray's trying to measure here the cost not to renew customers, but the cost to retain customers. And to do that, you're going to have to include a lot more than customer success or a renewals team, in my opinion. Well, go ahead, why don't you tell us what you're going to include, Ray?

So here's what I'm going to include. I'd love to see what you think of this. But number one, it is customer success expenses. But me, Ray, Mr. Allocation Reich, I do— if we're allocating some CS to expansion, upsells and cross-sells, I would only include the percent allocation to retaining customers and renewing customers. But if it's too hard, then I just allocate all CS expenses.

Ray, never afraid to arbitrarily split a number in 2, Reich. Okay, so you want to try to get some allocation of customer success. And this one, I don't know, it might actually be possible in my opinion. Um, so, so I'm not— some of the allocations you do, I'm more scared of than others. So you're going to say, I'm going to take CS expense, I'm going to try and guesstimate how much of their energy is in renewal in effect versus expansion. And, and you're going to leave the expansion out, I think.

Okay.

You leave the expansion out, you're going to keep the retention renewal in. So, so this is This is the cost of retaining customers and keeping them at flat ARR, basically. Let's just argue, or keeping them at gross churn ARR, not— and keeping them at GRR, not NRR, right? Because you're not going to include the cost of the expansion, so you can't include the benefit.

That is correct.

When you think about this number. Okay.

So that's one. So the second component is the renewal team cost, whether you have a dedicated renewals team or you have an account management team. responsible for renewals, then they would be in the customer retention cost addition. But so would any onboarding customer engagement programs, technologies you use to retain customers, your customer success platform. If you're using AI tools to evaluate customer success, product utilization, their revenue trends, etc., all that would be in the customer retention cost. Any questions on those 3 things?

How about professional services?

No. So this is where I go against my allocation. I believe professional services should be a line item on the income statement from both a revenue and a COGS perspective. So I would not use that, even though I know sometimes services can be used as a customer retention strategy.

Yeah, because I think at the first order, I agree with you. If you're running a, let's say, normal kind of semi-break-even services business, Then why not just leave it out of the whole equation? But if you're losing money on— I would argue if you're losing money on services, which by the way, we'll talk about Zora a little bit because they got sold last week. Their services margins were part of it and they were bad. They were, I think, negative 20, maybe negative 30. Only CFO did a post on this. They had relatively high negative services margins. And I would argue that lost money is actually a cost to retain customers in all likelihood.

I totally agree with you, but—

But you're not going to count it. You agree with me, but not enough to put it in the definition.

I'm not going to use the definition. Once again, I don't have a black and white answer here. I just, I think this is very important, thought-provoking. If more and more investors want to know this, and operators need to become more efficient in retaining their ARR.

Okay, so what, Mr. Fearless Allocator, what about R&D? So you're okay to look at customer success and say, ah, this chunk goes to renewal and this chunk goes to expansion. I haven't heard you say R&D yet. Are you willing to make a similar guesstimate and say, ah, 20% of R&D is actually going to not only bug fixing and stuff sustaining, but also, let's just say, new features? Because if you wanna keep me renewing, I need a modicum of new features. Thoughts?

In the old perpetual software days where you had your sustaining or maintenance engineering team, it was easy. Today, no engineer wants to be viewed as a maintenance engineer. So I think it's going to be too hard even to determine that allocation. So I say no, we do not include any R&D into customer retention costs, which by the way may not be right. And if you decide to allocate 20% to maintenance, just document it, define it, and maintain its consistency. That's what I'd say to that, Dave.

All right, all depends what you're trying to measure, in my opinion. If you're trying— I'm not sure you're trying to do this because it was me who brought it up, but if you're trying to do that split P&L, I got to believe that some of the R&D belongs there in the recurring business. Because if you turn off the R&D, churn will go up. So that business won't just sit there, it'll start to shrink faster than it would have.

And we know, Dave, if you get into level 2, level 3, or severity 2, severity 3 bugs, Engineering is going to fix that, right? Because they don't want to lose customers because of it, especially if it's—

Absolutely.

Yeah.

Not just from bug fixing, but also, yeah, but incremental enhancement as well, right? But okay, you're putting a proposal out there and you're saying, hey, you want to exclude R&D. So, it's kind of the non-R&D cost of retaining the customer base.

Yeah. But now, I'm going to, what you might say, is argue against myself. Customer support, we know it's the COGS, right? I believe that customer support's primary Reason for existence is customer satisfaction and customer retention. So I include customer support in my customer retention cost calculation.

I mean, look, you didn't call it customer retention operating costs because customer support is not an operating cost, it's custom consult. So I'm okay with you including it if you want to. I would include it, by the way, in that split P&L. So I actually feel like, Ray, you're very close to this notion of split P&L, but you're reluctant to pull the trigger on certain items like R&D or the PS loss. But in the split P&L, there's no question in my mind customer service COGS would be in the recurring business.

I actually think an episode— we should dedicate an episode to that split P&L because I think it's a very important concept you come up with.

I was thinking the same thing, Ray.

So, and then the last thing, as companies are starting to invest more marketing resource dollars into customer sat and customer retention, especially user group conferences or customer conferences, I believe you should look at what percentage of marketing programs are being allocated to existing customer satisfaction retention and include those in the customer retention cost calculation.

Yeah, I mean, in general, I would not do that. I mean, on the flip side of this, I spent a lot of my time reminding marketing teams that they can do customer marketing, um, because by default they tend to lean very heavily towards new business. My other objection is you use 2 words that sound similar but are in my mind super different, which is user group, which means letting the local users use the conference room in the New York office and buying them 2 dozen Dunkin' Donuts for a grand incremental cost of $12 versus a user conference, which is a multimillion-dollar CEO pet project at many companies. And if you just want to slide that into the cost of retaining customers, I'm not sure I agree at all. I think that that's—

Yeah.

Look, existing customers go to those conferences, but I think they're primarily about generating new business, either with new customers or even new business within the installed base, which you are excluding anyway in this calc.

That is correct. And since I'm excluding expansion, I don't know, I, we've all done these customer conferences. And for me, I was amazed at how many economic buyers would attend and that our salespeople got to spend time with the economic buyers they don't on a regular day-to-day basis. And we just found that our customer sat went through the roof if they attended the conference versus they didn't.

Yeah, I get it, Ray. It's very hard to allocate these things. So, but in general, I would, on this one, I have a definite point of view, which is I'd exclude marketing because I'm excluding the actual only cost I think is spent in retention is the donuts and the newsletter. And those are kind of de minimis. And I think the big conference is much more about expanding. either expansion or new ARR, and you get some renewal as a byproduct.

Now, what I'd like to do is, you know, that's just the customer retention cost. The metric that's most important to me is the customer retention cost ratio, where we take all of those components of customer retention cost in the numerator, and then we divide it by the total ARR that we're trying to retain and renew. It's the same ARR you're using in your gross revenue retention. So that ratio says I'm spending X cents to retain every dollar of ARR. So that's the metric I really want to get to.

Okay, go for it.

So the question is, do you think the denominator should be total ARR that I'm trying to retain Or only the ARR that we actually end up renewing? Because then it'd be closer to a renewal cost ratio versus a retention cost ratio.

Yeah, so I— look, this is a distinction. Let me just make it clear what Bray is asking. If you want to do the cost of retaining or renewing, should you divide by the pool of stuff you're trying to renew or the stuff you actually got renewed? Um, and if you want to be literal, like, well, I only renewed 50 units of 100, then I see that you just say I spent $50 to renew 50 units, I spent a buck a unit. But, but I— but the flip side, I spent $50 trying to renew 100 units. I, I would argue I would always do the pool size, frankly. And here's the reason, because I think it's a slippery slope. I mean, first, ideally you're getting 85% of it anyway, so there shouldn't be that much difference. And if you're only renewing 50% of the pool, you've got a whole bunch of other problems than how to define this metric. So, so in general, it should be only a 10 or 15% gap. So that's the first reason. The second reason, in a perfect world, is I want it to be 100. I want to get everybody, um, and I never want to forget that. The third is, um, you avoid this up for renewal, the ATR issue, the available to renew. So say there's a 100-unit pool out there to renew, but half of it's renewing next year, half of it's renewing this year. Do I count it or not? And they got called from the CSM They went to the user conference, right? They called tech support, but they're not up for renewal this year. But all of those acts were spent in trying to renew them next year. So they are in the pool of stuff that I am trying to renew. And therefore I don't like the approach that says, oh, for this metric, there might be other metrics where maybe I'd get sold by it. But like staffing, I don't know, ARR per CSM is another interesting one. Do you do that against the whole pool or only what's up for renewal?

Yeah.

Because sometimes it can be a 2x difference in the size of your CS headcount. So, but putting that aside for a sec, I like talking about the whole pool because that is what I'm trying to renew, whether it happens to be up for renewal this year, next year, or the year after.

I totally agree with that, Dave. I, I would use total ARR, the whole pool. We're in total alignment.

Amen.

So, you know, we're running out of time, so maybe we just summarize what we talked about.

Did you want to get to your other metric? Didn't you have another metric for today?

I do have the customer renewal cost ratio, and the difference there is in the numerator I only look at those expenses 100% directly applied to the renewal process— your account managers, your CSMs if they're involved, if sales is involved. So that's where I actually have all the renewal-specific expenses in the numerator and then the ARR renewed in the denominator.

Now, I like this metric. The first metric I don't love because I, I like it— I'd rather do an episode on how to split the P&L in 2, and that metric would kind of not be necessary anymore if there was a standard way to split it in 2. And then everyone could look at their business and say, ah, here's a recurring business, here's a non-recurring business, and you can kind of see how much you're spending to acquire customers and if it's worth it and what happens to them once you do. This metric, however, the cost to renew as opposed to the cost to retain, I like a lot for a very simple reason. It's the old an ounce of prevention is worth a pound of cure. Because if we can calculate— do you have a typical benchmark for that, Ray? Like what it costs? What is a typical cost to renew using your math and your definitions?

I, I just did it with a couple companies and one was 7% and one was 9%.

Hey, that was exactly my guess. It was gonna be a nickel or a dime to renew a dollar of ARR. And I like this metric a lot because I actually think it helps the customer success team team enormously because it says you can spend a nickel or a dime now to renew something, or you can spend $1.50 to backfill it, right? So I like it because it's normalized to ARR, right? It's not really a CAC ratio because it's the cost of keeping, not acquiring ARR. But, but gosh, I can save that, you know, if I take the 10 cents down to 5 because I'm cheap. If I screw up and lose renewals I could have got otherwise, it's going to cost me $1.50 to backfill every dollar I lost. So, I think of this as kind of the ounce of prevention is worth a pound of cure metric. And that's why I like it. To me, the most number one important thing about churn ARR is it costs CAC times CAC ratio times churn ARR to backfill it. And the more we say that to people, the more they're going to love customer success.

That's a great way, I think, to end this episode, is that one of the reasons it's so important is if you know how much it costs to renew it, For every percent you don't renew, you know how much money you're going to have to spend to replace it.

And that's a whole lot more.

Well, there you have it. You have the customer retention cost ratio and the customer renewal cost ratio. And I think the most important thing with all metrics, Dave, as you and I talk about all the time, make sure you define it, everyone understands the definition and calculation, and maintain the consistency over time.

Bingo.

Okay. All right. I enjoyed this episode. I got to get back to looking at some of these beautiful leaves.

Some of these beautiful leaves. Okay. Uh, well, awesome, Ray. Have fun in the forest this weekend.

Okay. Bye-bye now.

The Metrics Brothers AI, SaaS, and More is a production of the Metrics Brothers Growth and CAC and a member of the Benchmarkit Podcast Network. By accessing this podcast, you acknowledge that the Metrics Brothers make no warranty, guarantee, or representation as to the accuracy or sufficiency of the information provided Or the humor content of the jokes told.

Ahem.

Ray. Information, opinions, and recommendations presented are, according to our spouses, probably wrong and provided for information purposes only. This podcast should not be considered professional, or for that matter, unprofessional advice. We disclaim any and all liability for any direct, indirect, indirect, misdirect, incidental, special, ordinary, consequential, inconsequential, or other damages arising out of any use of, or God help you, reliance upon the information presented here. Ray Grothreich is based in New York City and available on X @rayreich. Dave Kellogg is based in Silicon Valley and available on X @kelblog. Schedectady, which is French for unspellable, is not our actual production location. Thanks for listening.