Implied ARR

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, founder and CEO of Benchmarkit.

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

And together, we are the Metrics Brothers.

And we go together like Tick and Tie, our mascots at Host Analytics back in the day.

Dave Kellogg, the script says Tick and Tock, not Tick and Tie.

Ah, old habits die hard, Ray. Every time I see tick and something, I think Tick and Tie. Okay, Ray, we go together like Tick and Tock.

All right, well, even with that, Brother Dave, I'm not so sure we want to associate SAS Talk with TikTok in the current political environment.

Excellent point, Ray. You, you don't like what's implied, do you?

Implied? Did you say implied? I think maybe we should dive into implied ARR to keep this theme going.

Okay, let's do it, Ray. I love implied ARR. Let's make that today's episode.

Okay, but before we get into implied ARR and its implications, let's hear a word from our sponsor. The Metrics Brothers is proudly sponsored by Benchmarkit, the B2B AI and software industry's largest, broadest, and most contextualized benchmarking index. Dave Kellogg and Ray Groth Wright have put the Metrics Brothers podcast on hiatus for the remainder of 2026. All episodes we have ever recorded are still available on your favorite podcasting app. You can reach Dave @Kellogg and Ray on LinkedIn @RayReich. Thank you so much for listening to the Metrics Brothers podcast. Okay, Dave, how about if we start with what implied ARR is and a little bit about the history? Sound good?

Sure. I know you did some homework here because I had maybe a different perception than you of where it came from. But let's hear what you found out.

Yeah, I did some homework and I had one of my best friends, ChatGPT, right here by my side.

Okay, so we're just going to replace you with ChatGPT. We're going to have it talking now.

Well, except ChatGPT doesn't have my sense of humor yet, Dave.

Okay, fair, fair.

Okay, so implied ARR, it represents the annualized revenue run rate for public SaaS companies. That's where it's used, and it's calculated for the most part as current quarter total revenue Multiplied by 4, trying to get to that ARR number. As an example, if a public SaaS company did $500 million in revenue in the latest quarter, the implied ARR would be $2 billion. So that's kind of what the research says. Do you want to talk a little bit about that or go into a little bit about the first time we saw it being used?

Sure, sure. So, so look, implied ARR to me is a lot like billings. And it may seem funny because they measure different things, but in spirit, they're metrics that are effectively proxies. And you never hear about these metrics in the private SaaS world because in the private SaaS world, when a VC asks a startup, what's your ARR? They answer. And when they ask how much new ARR did they have last quarter? They answer. The issue is with public companies, they don't necessarily release ARR. I think most public companies don't release ARR.

Yeah.

And very, almost none to my knowledge release new ARR, which is what billings is a proxy for. So just like billings is a proxy for new ARR in a public SaaS company, implied ARR is a proxy for ARR in a public SaaS company. And the reason we have to talk about this metric is quite frankly, simply because they don't give it to us.

No, they don't. And it would make it much easier. But even in private companies, ARR is not a standard. It's not that well-defined. But a little bit more about me and ChatGPT and what we found. The first, you know, I thought this was starting to be used by Wall Street analysts kind of in that mid-2015, 2016 timeframe. But the first S-1 filing that I could find that used the word implied ARR was Datadog. And that was in their 2019. And then PagerDuty, also in 2019, actually mentioned net new implied ARR in their S-1 filing.

Yeah, that makes sense, uh, Ray. I mean, frankly, I'm surprised that I didn't think anybody talked about implied ARR in their S-1, because to me, the first time I heard about implied ARR was, again, back to Billings, the same way I heard about Billings the first time, which was reading public company financial analyst research. And as soon as I saw it, I was like, oh, this makes sense because they don't have ARR. So they're going to basically take 4 times, hopefully subscription revenue, but I know they don't always get that. So that's why they're taking total revenue, which is, you know, proxies are proxies for something real and like what makes them worse? Well, that's the first thing that makes this one worse because it's 4 times all revenue, which is going to be including services, whereas obviously ARR doesn't. But in any case, to me, the better definition when you can calculate it, I.e., when the company breaks services and subscription revenue, is 4 times subscription revenue to get implied ARR. But good digging there, Ray. Frankly, I'm surprised that people even put it in their S-1. I thought it was purely a kind of external financial analyst metric.

You know, Dave, you just mentioned something really important, and that is subscription versus total revenue, which includes the services revenue. I think it'd be a good time to double-click on that. You good with that? Okay, I'm jumping ahead here in the script, but this was one of my biggest concerns of implied ARR is if it was subscription revenue times 4, I'm pretty comfortable with this as that proxy, as you talked about. But because— and I was talking to Alex Clayton at Maritech, and it was the first time I really heard how at least they were calculating this. They take total revenue times 4. It's like, okay, but if one SaaS company has 90% subscription and 10% services, and another has 70% subscription and 30% services, that's 2 very different stories, right, of what implied ARR is telling you.

Yeah, and I get it. I mean, look, I do believe if you look at Maritech's public comps table, uh, they'll have implied ARR as one of the columns. Uh, they're definitely— they're probably the second place I heard about this metric. The first is just reading financial analyst reports. But because they have a database of public comps, I'm sure they include it as one of the columns. And look, it's one of these, it's bad, but you can't do better. If you want to create a single metric that works across all public SaaS companies, it's going to have to be implied ARR inclusive of all revenue because not everybody breaks it out. And again, the kind of the first order correction that maybe needs to be applied, but you don't really have any data to correct it. But the first thing that's going to make it wrong, how about that, is that it's including services.

Yeah.

So, if you happen to know that one Company, you know, Company A has a big services business because they're enterprise and Company B is a transaction volume driver where it implements for free, that then you're going to know that you're kind of mixing apples and oranges when you compare their implied ARR.

Yeah. And I tried to do a little bit more digging, Dave. So I looked at Workday and ServiceNow as 2 examples. And they both actually report subscription revenue and services revenue. By the way, only 35% of public SaaS companies do that. 65% just show total revenue. But back to Workday and ServiceNow. If you look at the Workday net implied ARR growth, it's around 3.2%. But if you just look at the subscription growth, it's 2.5%. So it's slower. Where ServiceNow is just the opposite. Their total implied ARR growth was 6%. But if you look at just their subscription business, It was growing at 6.8%. So to me, if I'm an analyst, and I know we never give investment advice here on SaaS Stock, but ServiceNow looks even better when you just look at this subscription implied ARR growth rates.

Yeah, so, so let me just try and recap where we are at this point. So first, there's a metric that maybe you've never heard of called implied ARR, because if you're a private SaaS company vendor, odds are you've never heard of this unless you read financial analyst So there is this new metric, it's implied ARR. Why does it exist? Because companies don't release ARR. So basically Wall Street has created a proxy for ARR called implied ARR. And then what's its first limitation is that it's based on all revenue, not just subscription revenue. Striking, Ray, that you said that only 35% of people break it out. That surprised me. I would've guessed that was inverted the other way, that 65% broke it out and 35% didn't. and you're saying it's 35-65. So, so it makes sense, even more sense, why they have to include everything, because 2/3 of companies don't break it out. So now back to you. I know there's one thing I want to talk about, but, but back to you.

Well, as I was, you know, doing my research, you know, the first thing I thought about was your session at SaaSmetricspalooza '24, where you talked about one of the important things about having the ARR, or in public companies, implied ARR, So many of the SaaS metrics are based upon it. Churn rate, gross net revenue retention, CAC payback period, ARR per FTE. So to me, that's a reason why implied ARR is so important is we can still use it to do these SaaS metrics calculations.

Yeah, I mean, kind of the question we didn't ask yet is why does this thing exist? And I said, because people want to create a proxy for ARR, which begs the question, why do they want a proxy for ARR? And the answer is, yeah, ARR is kind of the atlas holding up the world, right? It's the underpinning of CAC ratio, all churn metrics. It's— ARR is a fundamental ingredient to many, many important SaaS metrics. So that's why you need one. If you want to calculate SaaS metrics, you need ARR. And there's really 2 cases that make ARR hard. One is a public company where they don't release it. So you have to guess using this implied mechanism, i.e., 4 times revenue run rate basically, or 4 times revenue most recent quarter. other cases where it's usage-based pricing. Because I actually think that ARR, I mean, first order, ARR is implied ARR in a usage-based pricing model. Not always, but a lot of companies basically say spend is truth, which is my view in a usage-based model. Don't try and do some baseline versus variable. Don't try and bust it up. Just say, hey, here's how much people spent last quarter. And then effectively implied ARR just multiplies that by 4. So the interesting thing, the most interesting thing about implied ARR to me, as much as I know you don't like it for some things, is that implied ARR for usage-based really kind of is ARR if you calculate it in my mind the right way.

Yeah. Well, then we could take it to the next level though, Dave, because then we have net new ARR. And traditional world, right, says net new ARR includes all your new ARR minus your Churned ARR. And as you, I believe you say, and I want to speak for you, new ARR is the new logo ARR plus existing customer expansion ARR. And the churn can be from downsales or lost customers, correct?

Yeah. So first, just taxonomy check. Yes, that's my taxonomy. A lot of people don't use it, but I love it because it's symmetric, which is I always say there's new ARR and churned ARR. New has 2 flavors, right? From new customers and from existing customers. The latter also known as expansion. Right? And then churn ARR has 2 flavors, lost, i.e., they stopped doing business with us, and downsell. That's kind of my taxonomy. I love it. A lot of people, I think you sometimes say, oh, churn is only lost, I think in my world. And then there's shrinkage. But I like the symmetry of the 2-level hierarchy. So, back to you.

But the story about this was we have implied net new ARR. Which for public SaaS companies, it takes current quarter revenue, total revenue, minus previous quarter revenue, and then they multiply that by 4, and that gives them net new implied ARR. And then a lot of the analysts use net new implied ARR for SaaS metrics calculations such as CAC payback period. In fact, Dave, what they do is they And Meritech, I'll use Meritech as an example. They took previous quarter sales and marketing actual expense and they divided that by the implied net new ARR for the current quarter. And I have lots of maybe issues with that, but that's a use of implied ARR for a CAC efficiency metric.

Sure. So what you're saying is basically in a place where— what I'm hearing, Ray, is in a place where you would like to do it Purely on net new. Wait, you're saying net new here? Uh, now I got myself confused, right? Hang on. You're saying you want to— ah, I know what it is. You want the metric to be based on new ARR, but they're basing it on net new ARR. And as usual with public companies, why is that? Answer: because they have no choice. Um, and, and that does explain, by the way, you have just explained why Maritech's payback periods are so high. I'm on their website right now, right? And it's, what is it, 25 months? I just lost it. But whenever you look at Maritech comps, the single metric that most strikes out to me is like, wow, these are high, is their CAC payback period. And that's because they're basing it on net new ARR, i.e., net of churn, rather than new ARR, which is really the right way to calculate the metric.

Yeah, which is why when, you know, a year ago it was up to 29 months at median and my private SaaS company benchmarks had a median of about 16. I'm like, what's the dissonance between the two? And it was because they're using net new ARR, which includes the churn. You're exactly right.

Yeah, I'm trying to find it right now. It looks like their mean right now is 22 months, but they're showing lots of great companies above 30. Like, they got BlackLine at 29, they've got Ping at 32, uh, they've got— who else up there? They got Domo, a favorite to talk about, at 41. New Relic 51 box at 66, right? Um, and I think in your benchmarks, right, what's the median again for CAC payback period for private SaaS?

For all companies, it's at 16 months right now.

Yeah, and they're showing 22. So that's a— well, that's a 33% difference. Uh, so that's roughly grosso modo what you'd see if you included churn, right? Because GRR on average is 15 to 20%. So you think it would be a 15 to 20% correction. So maybe some of that is public versus private SaaS as well.

Could be. But, you know, we've spent a lot of time talking about implied ARR, what it is, what are some of the challenges. But I once again go back to your presentation at SaaSpalooza. Why wouldn't we just use trailing 12 months, not only for, you know, talking about how a company's growing over the last 12 months, but for the SaaS metric calculations?

So, look, you can. And I think this is what I call spend is truth, right? And I think in a world where ARR varies every month, and that might be usage-based pricing, but just so you know, I work with companies where the monthly bill varies and they're not. They're user-based. And the reason is, oh, 8 guys used it this quarter. We'll charge you for 8. 10 people use it this quarter. We'll charge you for 10, right? So, just because it's user-based doesn't mean it doesn't vary because some people don't sell know, 10 people for a year, i.e., every month is 10 people regardless of who use it. Some people base the monthly price on usage in the sense of number of users. So sorry we're so deep in the weeds here, but these things matter. So, and to summarize that whole concept, I call it monthly varying SaaS just to make it simple. And the definition is wherever the monthly bill isn't the same. And it might be because you have a usage-based model, it might be because you have a value-based model, it might be because you just charge per users per month, like per So there's a bunch of reasons you could have it, but as soon as you enter that world, the ARR concept starts to lose meaning. And the question is, what should you use instead? You're smiling, so I want to go back to you.

Yeah, well, I hate introducing new metrics here in the podcast, but you mentioned about the variable. I've got this category now called variable recurring revenue. So all usage-based or outcome-based GAAP revenue, I translated into variable revenue. So then I add subscription revenue to variable revenue. And I get annual recurring revenue.

Yeah, and this is where we get to fight like the metrics brothers that we are because Ray and I are firmly on different sides of the line here. But my answer is as soon as it starts to vary, the whole notion of ARR starts to lose meaning and therefore just use spend as truth. And you can decide if ARR is trailing 12 months or 4 times trailing quarter or 12 times trailing month. I've seen companies do it, but they're all applying the spend as truth thing. Personally, I think I like 4 times the trailing quarter, which by the way is implied ARR. That's what I mean. In a usage-based world, implied ARR is ARR. But not everybody does it that way. The other thing, Ray, that I think is very important is that ARR was supposed to be a forward-looking metric, right? That it was, I just signed a contract for 100 units, and therefore over the next year, I'm going to get 100 units. And as soon as you go to this spend is truth, this backward-looking trailing spend model, These metrics start to mean different things. We've talked about it before with the famous Snowflake 2-year lookback, where when they do NRR, they look at last year's spend over 2 years ago's spend, right? And you're taking a metric that in an AR world has actually got one foot in the future and one foot in the past, right? Next year versus last year. And you're comparing it to 2 feet, one a little bit in the past and the other a lot in the past. So these are some of the nuances based on this, because Again, this is all driven by the fact that we're using ARR as an ingredient for a bunch of SaaS metrics and we're interpreting those SaaS metrics, but how you interpret it should be a function of how you define ARR.

Okay. So anything else you wanna add or should we try to summarize this very deep and detailed conversation we just had?

I don't know, Ray. I had a line in here. I can't remember why I had it in where I was saying you're tilting at windmills and we had the whole Don Quixote theme. Where was it? 'Cause when Ray and I in the prep, We talked about this. I'm like, Ray, you're tilting at windmills again. And why was that? Do you remember?

Yes, it was because I was talking about I don't even like the CAC payback period equation of, um, Maritech because they're including churn. And I said, hey, at SaaS Metric Standards Board, we use CAC payback period, which looks at new logo ARR, and we divide that into sales and marketing expenses for the time period and Times 12, that's how you get CAC payback period.

Got it. And that's why I called you Ray Quixote, uh, Ray Quixote, because, because you're— in my mind, you're tilting at windmills in some sense because you, you want the world to be different. And I understand that. Look, the SaaS metrics world in private companies is a way better world than public, uh, because you're deliberately blinded to certain metrics. But I don't think you can do any better, Ray.

And, and because I love the book Don Quixote from La Mancha, right? And by the way, the author, Miguel de Cervantes, you know, he's a Spaniard from the late 1500s, early 1600s. And he wrote that book because he wanted to reintroduce chivalry into society. And I always like to think that I'm a chivalrous person. So I actually like being called Ray Quixote.

There's no place for chivalry in SaaS metrics, Ray. I don't know about this. Ray of La Mancha, Ray Quixote. I don't know what your new nickname is going to be, but it's pretty soon we're going to change you from growth to something else.

Yeah, I think I'm going to stick with growth. So hey, let's summarize. So implied ARR is taking the annualized revenue run rate, you multiply it by 4, and that gives you implied ARR. Companies are using it because they need a proxy for ARR in private companies to do all those SaaS metrics. And most importantly, net new ARR is implied, and it does include all the churn And the downsell, and that can impact the SaaS metrics calculations. I think that's kind of the summary, Dave.

Yeah, I mean, my summary is pretty simple. We got this metric called implied ARR. If you're a private company SaaS, you probably never heard of it. It's probably most relevant to companies entering the IPO transition zone because you're going to have to start reporting on this metric and people, whether or not you report on it, people are going to calculate it for you. So I do think for the audience, if you're a small private SaaS company, you probably don't need to worry about implied ARR. If you're getting into that IPO zone, couple hundred million, good growth rate, you need to start thinking about not only implied ARR, but billings. Because whether you report it or not, they're going to get calculated for you. And people are going to ask questions about them. So, so that's my take on this metric. Go ahead, Ray.

And you know what they say, if you're going to use implied ARR, understand the implications.

Got it. Oh, ouch. Ray, now I have one last pet peeve because we have a minute. One of the things you say, and I think Maritech said it as well, is that they talk about They say it's annualized run rate, right? And this is a total hair split, but to me, that's redundant. That's like saying ascend up the stairs, right? Because I think the definition of run rate is you take the current period, the current quarter, and multiply by 4. So, I think run rate is inherently annualized when you say it. So, I just wanted to call that out since we had a minute. And I know our audience has to speak precisely. I would just say it's the run rate. Revenue, uh, not annualized run rate, or not run rate multiplied by 4. I would say the run rate revenue, i.e., the most recent quarter multiplied by 4. But these are, these are details.

And that's why we do SaaS metrics with the Metrics Brothers, to be able to identify those fine, um, hairlines.

They matter. You'll sound better in a board meeting if you say run rate rather than annualized run rate, in my opinion. But okay.

Hey, Dave Kellogg, thank you for discussing implied ARR with me.

Thanks a lot, Ray. Take care.

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. Ray, information, opinions, and recommendations presented are, according to our spouses, probably wrong and 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 @kakellogg. Kellogg. Schedectity, which is French for unspellable, is not our actual production location. Thanks for listing.