The ICONIQ 2025 State of Go-to-Market Report

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 Benchmarket.

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 Metrics Brothers.

And we go together like Frick and Frack. That's an oldie, Ray.

Well, you know, sometimes I show my wisdom and experience, Dave, but, you know, we went from the eating in Spain to comparing ourselves to 2 Swiss comedic ice skaters, and I highlight comedic.

Well, I did not actually know they were comedic. The reference here is Frick and Frack, things going together like Frick and Frack, and unbeknownst to most, Frick and Frack were in fact Swiss ice skaters. I did not realize they were comedic ice skaters, Ray.

Yeah, well, you know, they were trying to put some humor into a beautiful sport, and we try to put humor into a beautiful thing— metrics. So, but you know, Cac, I might, or we might be skating on thin ice here, but I thought it was still an interesting metaphor for today's topic.

Okay, skating on thin ice, talking about go-to-market teams. We are today, to cut to the chase, going to talk about Another of ICONIQ's very, very good market surveys. This one is called the State of Go-to-Market in 2025 report dated in June. So, this is fresh off the presses and based upon a survey taken in April.

Well, Dave, I think we should move beyond the comedy portion of the show and let's hear a word from our sponsor before we dive in.

I didn't realize the comedy portion had started, Ray. Had it?

Only in my own mind, Dave.

In your own mind, the best place for it. Okay, sponsor word, here we go.

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. So today we're going to highlight some of the key things we saw in ICONIQ's The State of Go-to-Market in 2025 report. And just at a high level, why it looks like some go-to-market funnels are less effective While some, especially AI native, seem to be outperforming. So it's going to be interesting to look at some of the contrasts and comparisons later in this report.

So given this report is based on a survey, Ray, why don't you tell us a little bit about who they surveyed?

Yeah.

So, you know, this is the 3rd year I believe they've done this go-to-market. They surveyed 205 B2B SaaS go-to-market executives. What's interesting, it's a small point, when they showed who those go-to-market executives were, it was primarily chief revenue officers, head of sales, revenue operations people. But interestingly, it did not include marketing or customer success. So, it brought up the question in my mind, what is go-to-market? I thought go-to-market was marketing, sales, and customer success.

Yeah, 2 little quibbles here, Ray, and some are pretty little, but they say including A, B, and C, and in the included list, is not CMO or CCO. It doesn't mean they're not there. It just means they didn't mention them. And the even bigger quibble is, I did notice this when I went through, they said GTM execs at 205 B2B SaaS companies. So, there could be 500 respondents, right? They may be hitting 3 titles at one company. That was also unclear. So, all I know is that they had GTM execs at 205 companies. So, it's a little ambiguous there. And I don't think they're showing Ns on the slides, are they, for number of respondents? So—

I would Yeah, what's interesting, they don't show N at the bottom left like they did last year. But where there's a segmented view, they do often put the N under each segment bar.

Interesting. Okay. Well, backing out of the weeds there. Sorry, crowd. But, you know, this is how Ray and I look at these things. We read for the detail and hope it's footnoted somewhere. And Iconiq is pretty good about that in general. But I will lay one more critique on this, which is they sometimes segment by high growth versus low growth. And they have what I consider to be a fairly arbitrary characterization of the two. They basically say that if you're less than $25 million, which in this report really means $10 to $25 million because the cutoff is $10, you have to be 100% plus to be a high grower. If you are between $25 and $100, you need to be 50% plus to be a high grower. And if you're $100 million plus, you need to be 30% plus to be a high grower. And the thing I find weird about those kind of arbitrary growth thresholds is that on average, they take the 21% of the companies meet the test. But by segment, it's all over the map. For less than $25 million, only 11% are high growers by their definition, while in the next segment up from $25 to $100 million, 42% are. So I think they would have been better off to just say top 20% grower and have done that by segment, and that'll be much more consistent. Yeah.

And then just 2 last things before we get into the actual content. A lot of their charts do show top quartile, and we'll call those out as we discuss them. But, you know, a lot of benchmarks, they only— they show median or 25th percentile, median, and 75th. So important to know what you're looking at. And then the last thing, and I love that they did this last year, but they didn't do it this year. They did not have a glossary or definitions. They do it in the footnote, but not as a dedicated glossary, Dave.

Okay. And I would say, what was your first point, Ray? There's no glossary in the first point. Oh, top quartile. Yeah, I agree with you that most SaaS benchmarks are medians. So when you look at these numbers, they all look pretty fast, right? They go like the growth rate is 135% for sub-$25 million companies. Well, that's not a median. That's top quartile. And to me, whenever we talk about Icahnic, we talk about what I call the Ivy League effect, which is Icahnic tends to be a growth stage momentum investor, i.e., investing in high-quality, fast-growth companies that have made it to be pretty big. And that means, I always say, you're kind of looking at Ivy League applicants to the extent they're surveying their own companies. So you got to be a little careful on that. And just compounding that is they're showing top quartile, not median. So these benchmarks may look rather intimidating to the audience when you see the numbers, But just make sure you know what you're looking at.

In fact, the best place to start, and please everyone go look for that iconic State of Go-to-Market 2025 report and feel free to follow along. We'll go to slide 9. And slide 9 is their, they call it GTM Health, the growth. And one of the interesting things on this slide that I found, number one, Dave, they segmented almost every other benchmark and metric by AI native versus non-AI native. They didn't do it here on growth on slide 9, which I was surprised at, or anywhere in the deck that I could find. And then second of all, their overarching commentary is year-over-year ARR growth has remained fairly flat, though there's a significant exception in that $25 million to $200 million. So just wanted to see what your thoughts were on this particular slide.

Yeah, there's 2 things to me, Ray, that when you say growth has gone from 141% to 135% and call that Flat. You could quibble with whether 6 points is actually flat or down, but the bigger point is, that's not growth being flat, that's change in growth being flat. So, they're looking at the second derivative here, not the first. If you're growing at 135%, you're growing at 135%. That's just not flat growth. It's more the second derivative, the acceleration, if you will, is flat. So, as often with Iconiq, I love their data, I quibble with their conclusions, I often find them sloppy, And in this particular case, they did a cardinal sin of surveying, which is the callout, Ray, actually highlights that AI-native companies are experiencing significantly faster growth than their non-native AI peers, but the data doesn't show that. And that to me is kind of a cardinal sin of making one of these reports.

Yeah, that was my big takeaway that they didn't do that AI versus non-AI. But let's see.

It's okay if they didn't do it, but then they call it out, right? That's my quibble. If you don't do it, you don't do it, but you don't put it in the callout and make people go, where's that in the data? The answer is it isn't.

Yeah. One of the things that you and I talked about, one of the things we like about ICONIQ's reports is they often focus on things that other benchmark reports don't.

Absolutely.

So the next few slides are exactly like that. And it's slide 11, and it's talking about funnel conversion rates. And interesting here that only the largest companies, those greater than $250 million, are reporting an increase kind of across the board on On funnel conversion rates. And things like sales qualified lead to close won were really down in most of the smaller companies, anywhere from a 21% to 26% conversion rate. And what's interesting, those conversion rates being down, Dave, that really aligns to some benchmarking research that we've done where that new CAC ratio, which measures how much sales and marketing expenses incurred to get a dollar of new AR from new customers. It actually went up about 14, 15% in the last 12 months from $1.76 to $2. So this tells me conversion efficiency is going down, new CAC ratio is going up. But I will tell you, if some of the prognostications about how using AI and go-to-market is correlated to higher performance, and based upon what they show in later slides about the increase an investment in AI tools and technology across GTM processes, I would expect this to go up next year. That would be my hope.

Yeah, so first, Ray and I look at these charts very differently. I think, in my opinion, Ray looks at them more the way Iconiq looks at them, and I tend to look at all of this one level further removed. Like, just for example, look at their MQL to SQL conversion rates. They'll show, for example, for sub-$25 million companies, by year, 28, 32, 29, and call that down. And they'll call it down because 32 is bigger than 29, but 29 is higher than 28. So, I look at that and see normal fluctuation, and I see that people convert about 1/3 of their MQLs convert into SQLs, right? So, I'm always looking for patterns and rules of thumb, and some people are looking more at the minutiae here. Like, for example, in that same row, MQL to SQL for big companies, 250+, they show 29, 21, 24, and declare that up. And I guess they could say that's up because 24 is bigger than 21. You could say it's down because 24 is less than 29. But to me, when I look at that row, I just see about 1 in 3 MQLs convert to SQLs. And I think that's very useful information. So I like this report because it gives you data that other people don't, segmented by size, where sometimes it is significant. A lot of the times, it's not, to be frank. Surprising how often it's not. But I don't like their conclusions because I just, I don't know how big their samples are. I don't know how scientifically valid it is. Unless I'm sure about all that stuff, statistically valid, I just look for the patterns. And this is really useful for making rules of thumb.

Interesting. Well, I hadn't— you're right. I hadn't pulled back and looked at it like that. I looked at the specifics. And when I look at things like the demo-to-close one, And the $100 million to $250 million, it went from 41% down to 30%. And I'm like, oh my gosh, what happened? But then when I zoom out, I'm like, oh, it went right back to where it was in 2023.

Yeah. The other one to look at is the $250+ in that row. So demo to win, which is a very important thing. A lot of companies measure— if there's only one rate most CROs know in their head, it's demo to close. Whether that's a good thing or a bad thing, we can debate offline. But in the $250+, it goes 41%, 42%, 39%, and then they declare that down. And those are really small moves. In Ray's example, the moves were big. They were 10 percentage point moves. Here it's varying by 1 or 2 points. But again, I think the value in this chart is most companies across a shocking range of size from $10 million to $250+ demo-to-close is around 40%. That's really valuable information.

Yeah.

Well, let's go to slide number 12. And slide number 12 tries to split this out by companies less than $100 million, greater than $100 million, and then they sub-segment it by AI-native companies versus non-AI companies. And the thing that jumped out at me on this slide, and this is me learning from you, Dave, was something that a lot of companies don't show, and that's the, the free trial or proof of concept to a paid version, right? Because one of the things I'm hearing anecdotally is that more and more native AI companies are being forced to do proof of concepts Or trials before the contract is signed. And what's interesting here is with companies greater than $100 million, and I'm wondering how in the hell many AI-native companies fit in that category because I don't see the end there, but 56% of free trials converted to paid while in non-AI-native, it was all the way down to 32%. So that jumped out at me. It's like, why are AI companies having a much higher conversion rate than What I will say is legacy SaaS companies.

Yeah, and there's some chicken-egg phenomenon here, Ray. I mean, one of the things I'd love to see is what percent of companies have that as a cycle step. Like, that would be really interesting. Like, my intuition is that AI-native companies do it more because they have to, because they're using some aggressive pricing model that's either outcome-based or results-based, and they're somehow saying, hey, you know, we can We can use our software to do the same work as 10 SDRs, right? So, you're going to save all this money and people say, prove it, right? So, what's the chicken and what's the egg? Is it the value prop they're pitching that says they have to prove it first, or is it they're better at doing proofs? And I don't know which it is. The thing I can conclude from this data is they do win more. They have higher conversion rates. AI companies have higher conversion rates than non-native AI companies by somewhere between 10 and 20 percentage points. in this test. So, and I believe it's true. And I think the reason is, Ray, because they have to, because their pricing model requires them to prove the benefit statements they're claiming.

I totally agree with that. Plus, it's a little less proven as far as how AI fits in a particular process. And often they have to tune the models to a particular company's environment.

Yep.

Hey, let's go to slide 14 because this is something that we've talked about on a couple of previous episodes, and that's cost per opportunity, or cost per opti. And what was interesting here is, you know, they showed it by 4 different segments, which was great. And it really highlighted that smaller companies in this particular sample are really having a hard time with being efficient and generating opportunities, where it increased 82% year over year. So for those companies under $25 million, it went from $5.5K to $10K on a cost per opportunity. Now, they didn't break that down by ACV, which is what I'd like to see. Nor did they break this down, I don't believe, by AI versus non-AI companies. So it's missing some of that consistency of segmentation. But it was interesting to see that every category was up pretty dramatically except for larger companies over 250.

So once again, I'm going to look at this through a slightly different lens. First, here's my first reaction. These things are expensive. It costs $10K to get an Opti. Well, you're closing, they said a few slides earlier, 25% of those. So, it's costing you $40K to get a 1 opti, right? That's what this means. Maybe $50K at the high end. $32K to $50K if you just multiply these figures by 4. The next thing is I'm not sure if these are stage 1 or stage 2 optis. I believe they're stage 2 optis. So, you may have a higher close rate. That's what I would measure. And I often talk about cost per opti being $4K or $5K. These are roughly double what I say is my rule of thumb. And I think the answer is because they're taking sales accepted opportunities i.e., stage 2 optees. And I talk about stage 1 optees. Not sure about that. But a simple kind of reality check on your business model is to look at how much it's costing you to get an optee, either stage 1 or stage 2, and then look at your stage 1 to close rate or your stage 2 to close rate and say, wait a minute. It's costing me this much to win a deal. And how big is my deal size? If your deal size is $20K and it's costing you $50K to get one, it's going to be a problem. So, if there's $250K, then it's just fine. So, those are my observations here, Ray. As usual, I was more interested in kind of the absolute dollars and what didn't change than what did. But certainly some great data.

Yeah. Okay.

We got to keep going because it's so hard to pick out all the, you know, most relevant slides and do it here in our 20 to 25 minutes. But go to slide 20. And slide 25 is another one that honestly, I don't think I've seen this before, Dave. And that is late renewals and the percentage of customers with late renewals. And they segmented it by both ARR and by company segment, you know, enterprise versus mid-market. And what was interesting was the material improvement across all company size. And this really aligns with what I've seen on 2 fronts. Number one, a significant increase by customer success organizations of trying to understand the early indicators that are most predictive of churn. So there's been really a focus on that, including using more product analytics. And the other thing I've seen in other benchmark research is gross revenue retention is fairly stable. And that's been based upon maybe even over-indexing of CSMs, customer success managers, focusing on preventing churn. So I was really encouraged by this chart. How about you?

So, so look, I think this alone might be worth the price of admission because I've certainly seen late renewals in my day, but I've not seen anyone else benchmarking late renewals. So, so I think just Having this data is enormously valuable, first thing. Second thing, depending on how— look, they don't define late renewal. They don't define the impact of late renewals on NRR and GRR. My basic belief is companies that track late renewals will count them in NRR as if they renewed and then call them late, because the alternative is you call them churn and then have to call them new, right? So, I think I would not expect changes in this chart to result in changes in NRR or GRR, because I think it's already kind of adjusted into those numbers. My belief here is that when times are good, companies are sloppy operationally, and therefore, late renewals will be higher. And that when times are bad, people tighten their ship, if that's the right expression, tightening your ship. They run a tighter ship. There we go. People run a tighter ship, and therefore, I see late renewals down. That's what I think is driving this because times got tough and people said, hey, wait a minute. We can't be running around with all these late renewals. Even though it probably isn't impacting NRR or GRR because the way we count those metrics, it's a bad habit and we need to get out of it.

Yeah, totally agreed. By the way, I never understood the adjective tight going with the ship, but that's a whole other episode, I think.

Yeah, yeah, what's a tight ship? We'll have to look that up.

Okay, let's move to slide 22 because it was another one that I really liked, and I don't even see enough benchmark reports that look at the distribution of go-to-market headcount FTEs by team. Now, once again, at least they added marketing this time, but they did not include customer success. But a couple of the key points I saw was that— and they segmented AI native versus non-AI native. The big difference I saw was the amount of go-to-market headcount being allocated to post-sales, where, you know, in AI native companies, it's around 31% of all GTM FTEs. Dedicated to post-sales. And in non-native AI companies, it's 22% to 23%. And boy, this aligns with an article you highlighted just a few days ago, which was Andreessen Horowitz talking about, I think, the customer-forward or GTM-forward engineer. So this is really interesting how much money AI companies are investing in post-sales. What say you?

Yeah, the term you're reaching for there is forward-deployed engineer. It's a Palantirism. And I think it's kind of in vogue right now. And they're pulling— I almost feel like they're contorting themselves to talk about it because it just ultimately the data says that AI customers need more post-sales than non-AI native companies. And I believe that to be true. I don't know. I mean, technically speaking, a forward-deployed engineer is not a post-sales resource. It is an engineering resource that is placed out with customers. So, I think they're a little bit contorted on that.

Yeah.

But I like this data. I like the fact they're splitting AI versus AI native versus non-AI native. The other thing I'd note about this chart and some other charts in this report is they appear at ICONIQ to have a time machine because they are showing 2026 data. And I presume it's planned, but it's a little weird. And there's another place where they're showing 2025 data. Even here, they're showing 2025 data.

2025.

As far as I know, Ray, 2025 isn't over yet. And they're doing that a lot in this report. So, for example, if you just look at slide 19, they're showing 2025. Maybe it's the year of their survey. I don't know. But they haven't done a 2026 survey. So, I'm a little confused about their time dimension. But happy to have the slide.

Yeah, time dimension is important because look at marketing, Dave. They've got in 2026, marketing is going to be getting 4% more of the GTM headcount than they get this year.

Probably invented to keep the CMOs from quitting. There you go. Hey, we're going to increase the marketing budget in 2026.

Okay, let's talk about slide 36. So, and I really love this one too, and it's AI's impact on sales productivity. And once again, I didn't see— and Dave, maybe you did— but for slide 36, I didn't find a detailed definition of what a high adopter versus a medium or low adopter of AI and GTM is. Did I miss that?

Uh, I don't know, Ray. I don't recall, but, um, I'll go Ctrl+F the document for you while you continue. But the reason— I don't recall seeing it. No.

The reason I mentioned this, slide 36, it breaks out funnel conversion rates and sales efficiency rates between high AI adopters and medium to low AI adopters. And some of the things that jumped out at me was High AI adopters had a 20% decrease in sell cycle. But once again, my mind goes to, okay, but what ACV is that, right? Because if more of the high adopters are lower ACVs, then that would make sense. So some of these need to be viewed across multiple company profile attributes.

And by the way, Ray, the only definitions are the ones offered on the slide itself. I just Ctrl+F'd the PDF. So you do get a definition. It's just not very precise. It says high is AI is fully embedded in GTM processes, and medium/low is AI is not fully embedded into GTM processes. So, they give you a definition, although not a great one.

And the thing, the other thing, thank you for looking at that, by the way. And the other thing that jumped out at me was, you know, being a former sales leader, you know, they say that the percentage of ramped AEs that achieve quote on this slide is 5% higher for high adopters of AI versus the medium and low adopters. And I know if I was a head of sales or CRO, I'd be like, exactly how do I leverage AI to increase my quota performance and quota productivity? So those are some of the things that jumped out at me, Dave.

Yeah, I'm glad they do the split. As usual with this stuff, I'm worried about definitions. You know, that 56% to 61% of ramped AEs that achieve quota, I always ask, is that annual or quarterly? Is it the most recent quarter or is it the year? That's a pretty high percentage for people making it on the year. So it's not that high, I guess. I say 80 at 80 is good on the year. So I guess it's not that high. But I don't know. I just hate when they don't tell you the time period on that. What does ramp mean? Ramped within the year? If it is an annual figure, does it mean they ramped within the year or they started the year ramped, in which case it's awfully backward-looking metrics. So this is a case where I wish they had more and/or better definitions. But I love the fact that they do this. I think it's interesting to split it AI versus non-AI in terms of how much it's used for internal automation. And I don't see, Ray, to be honest, you know, the big— if this is it, if this is all there is for using AI tools, it's not too impressive. So I'm hoping to see a lot more.

Well, then that's the perfect transition to slide 40, which is the last slide we're going to cover in today's episode. And that is, what are the AI spend plans?

Yeah.

And this was amazing to me if I read it right. So correct me if I am wrong, but it basically talks about the average percentage increase in GTM spend on AI tools technology is 70 to 80% across all ARR scales. And to me, that is material investment. And you better have some pretty good proof points, I think, of how AI is impacting GTM efficiency and effectiveness. If you're going to be spending that much more money. That just jumped out at me.

Yeah, I mean, again, I don't want to be Negative Nelly here, but 70% or 80% more than what, right? And it's more than what I was already spending. And was I spending a lot or a little? I'd prefer if they benchmark this to the headcount budget, like compared to your headcount budget for sales, how much are you spending on technology today and how much will you spend tomorrow? And will your headcount budget be going down or up, right? So, If they gave us a little bit more to work with, I could get more excited about interpreting this. I mean, this unsurprisingly says people are planning to spend a lot more on AI. Great. But how much were they spending? And when it comes down to it, how much, either as a percent of assigned quota or percent of headcount or percent of sales, as a percent of something, what are we looking at? Because here it's a percent of what I spent last year, and that's going to be all over the map.

Yeah, I would say, you know, the thing that jumps out at me, Dave, is there was so much sales tech and marketing tech purchases over the last 5 to 10 years that just never produced the ROI and never got the adoption that you would think because companies weren't really good at the deploying, enablement, and adoption of those technologies. And if you're increasing AI tools and technology by 60 to 80%, what kind of sales enablement or other AI adoption in usage resources do you need to invest in? And they don't mention that at all.

Yeah, I hear you there. And sometimes, look, this stuff is just keeping up with the Joneses. So, the business benefit may be if you invest, you get to stay competitive, and if you don't invest, you fall behind. And I think that's going to be the case a lot. I think getting competitive advantage from AI tools is going to be hard. I think everyone's going to use them, and I think it's just going to kind of increase the stakes if you want to be in enterprise sales. And I think over time, it should increase ARR per seller. But I think in the short run, it's literally just going to be your salespeople won't be able to keep up with deals and volumes and opportunities if you're not using AI tools to boost productivity.

Yeah, exactly. In fact, I don't want to pitch anything that we're doing, but hopefully Benchmark will be coming out in the next couple months with how much investment is being made in AI technology and tools. by SDR, AE, and marketing.

So— Normalize it to something, Ray. While you're building the instrument, go normalize it to either total quota signed or total headcount budget or percent of sales or something.

Great advice. Great advice. Okay, let's summarize here. I think there's some great insights in this report, some new insights. Like we've said before, they cover things that we don't typically see covered in benchmarks. I really like this, how much companies are going to invest in AI. And then I want to see how that correlates to performance. One of the things I walked away is a little scared that they still basically suggest that AI-native companies are performing at a higher rate than non-AI companies. And the one chart that really blew me away was customer acquisition efficiency. It was really only improving a lot, right? If you look at the granularity and believe it, in larger companies. And at the same time, it told me GTM productivity does look like it's available. It's just what tools, processes, and focus you put on it. And then last thing is, it really seems like proof of concepts and trials are being used a lot by native AI companies, and that's going to change organization structure, investment, and measurements.

Okay, well, that's a good summary, Ray, so I won't attempt to re-up on it. I'll just say, Great episode, and thanks for doing it.

Okay, everybody, that was the latest episode here of SaaS Talk with the Metrics Brothers, and we covered the iconic State of the GTM 2025. Thank you for listening.

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 at Ray Reich. Dave Kellogg is based in Silicon Valley and available on X at Kellogg. Schenectady, which is French for unspellable, is not our actual production location. Thanks for listening.

Thank you.