GTM 159: Scaling ZoomInfo to over $1B ARR, the Upmarket Playbook, and Launching CoPilot to over $200M ACV in 6 months | James Roth

The GTM Podcast is available on any major directory, including:

• Apple Podcasts
• Spotify
• YouTube

James Roth is the Chief Revenue Officer at ZoomInfo, where he oversees a global revenue engine supporting over 37,000 customers and more than $1 billion in ARR. With a background in scaling hypergrowth teams and building enterprise go-to-market machines, James has helped reposition ZoomInfo from a sales tool to the foundational data layer for modern GTM teams. Under his leadership, ZoomInfo launched Copilot, its flagship AI product, and executed a bold shift upmarket — transitioning from transactional selling to a data-first, enterprise-driven GTM strategy.

Discussed in This Episode

• How ZoomInfo transitioned from transactional selling to enterprise go-to-market
• The “Good Co, Bad Co” framework for segment-specific product retention
• Why ZoomInfo changed its ticker symbol to GTM
• Launching Copilot and driving $220M ACV in under 6 months
• How internal AI usage became the go-to-market motion
• Key shifts in comp design and team segmentation for scaling upmarket
• Using telemetry and real-time signal tracking to measure rep effectiveness
• Lessons in long-term execution vs. short-term growth pressure

Episode Highlights

00:00 — ZoomInfo’s evolution from a sales tool to a data-first platform
Watch: https://www.youtube.com/watch?v=szKcRK2bpPc&t=0

03:20 — Why data as a service is ZoomInfo’s fastest-growing business

Watch: https://www.youtube.com/watch?v=szKcRK2bpPc&t=200

06:43 — How to build and apply the “Good Co, Bad Co” segmentation framework

Watch: https://www.youtube.com/watch?v=szKcRK2bpPc&t=403

12:48 — The hard part about going upmarket? Aligning the full funnel org

Watch: https://www.youtube.com/watch?v=szKcRK2bpPc&t=768

16:32 — Changing your ticker to GTM: A bold positioning play

Watch: https://www.youtube.com/watch?v=szKcRK2bpPc&t=992

20:21 — Copilot’s $220M launch: The power of internal usage

Watch: https://www.youtube.com/watch?v=szKcRK2bpPc&t=1221

26:05 — Most AI tools show <5% revenue lift. Here’s why

Watch: https://www.youtube.com/watch?v=szKcRK2bpPc&t=1565

29:42 — How ZoomInfo is operationalizing AI across the org

Watch: https://www.youtube.com/watch?v=szKcRK2bpPc&t=1782

34:43 — The signals> activity shift in modern sales measurement

Watch: https://www.youtube.com/watch?v=szKcRK2bpPc&t=2083

38:32 — What James wants future CROs to steal: Think long-term, build patiently

Watch: https://www.youtube.com/watch?v=szKcRK2bpPc&t=2312

Key Takeaways

1. Shrink to grow.
   ZoomInfo deliberately reduced its SMB footprint to fuel upmarket growth, trading short-term revenue gains for long-term profitability and retention. This disciplined bet shifted their mix to 72% enterprise, unlocking higher net retention and stronger LTV.

2. PMF is not universal.
   The “Good Co, Bad Co” framework reveals which products win in which segments, using retention and growth data — not opinions. Doubling down on winners per segment beats trying to force-fit a product everywhere.

3. Internal usage is the ultimate proof.
   Copilot’s $220M ACV launch worked because 1,600 internal GTM pros used it first, giving real-world feedback and case studies. Customers saw exactly how ZoomInfo’s own team used the tool to win.

4. Signals beat raw activity.
   Instead of counting dials and emails, ZoomInfo tracks how often reps act on high-intent buying signals. This shifts the focus from “more activity” to “the right activity.”

5. AI’s ROI is still hidden.
   The biggest wins so far aren’t direct revenue boosts, but time savings — like cutting deal prep from 85 hours to 10. Those productivity gains let teams redeploy talent to higher-impact work.

6. Repositioning starts in the product.
   Changing the ticker to GTM was only powerful because ZoomInfo had already become the data layer powering other GTM tools. Branding moves land hardest when they reflect a real operational shift.

7. The first 90 days decide your legacy.
   James warns CROs against chasing quick marginal wins early on. Use that window to tackle the big, hard changes — or risk getting stuck optimizing at the edges.

8. Telemetry> gut feel.
   ZoomInfo monitors real-time internal usage to decide which features to double down on. If the market isn’t adopting something your own team loves, rethink it fast.

9. Build for workflows, not wow factor.
   AI features that slot into existing workflows — like Slack alerts or CRM push notifications — drive higher adoption. Standalone “cool tools” risk being ignored.

10. Segment discipline prevents wasted cycles.
    ZoomInfo avoids pushing irrelevant products (like SMB data) to enterprise accounts, even if it could mean more short-term revenue. That discipline protects rep credibility and focuses resources where they’ll hit.

Thanks to Our Sponsor – UserEvidence:

UserEvidence is the Customer Evidence Platform that helps you collect feedback, surface proof points, and turn customer wins into bite-sized assets your sales and marketing teams can actually use. Capture social proof at scale. Arm reps with credible stories. Close more deals with trust.

Learn more at: https://userevidence.com/gtmnow

Recommended Books

• Call Sign Chaos by Jim Mattis
• Ride of a Lifetime by Bob Iger
• What It Takes by Stephen A. Schwarzman
• Radical Candor by Kim Scott

James prefers biographies over “self-help” — favoring real-life leadership stories over theoretical advice.

Referenced

• UserEvidence: https://userevidence.com/gtmnow
• Databricks: https://databricks.com
• Snowflake: https://snowflake.com
• Gainsight: https://www.gainsight.com/
• Copilot (ZoomInfo): https://www.zoominfo.com/products/copilot
• ZoomInfo: https://www.zoominfo.com/

Guest Links

• LinkedIn:https://www.linkedin.com/in/james-roth-3a913b51/

Host Links

• LinkedIn: https://www.linkedin.com/in/sophiebuonassisi/
• X (Twitter): https://x.com/sophiebuona
• Newsletter: https://thegtmnewsletter.substack.com/
• Website: https://gtmnow.com

Where to Find GTMnow

• Website: https://gtmnow.com
• LinkedIn: https://www.linkedin.com/company/gtmnow/
• X (Twitter): https://x.com/GTMnow_
• YouTube: https://www.youtube.com/@GTM_now
• The GTM Podcast: https://gtmnow.com/tag/podcast/

Episode 159 Transcript

James Roth: 00:00

One of the biggest shifts is that we have a sales tool, a sales platform that all sellers can and should use. It’s really more about data. It’s about funneling our data into the respective systems. We had a machine that was very much set to significant growth

Sophie Buonassisi: 00:38

We’ll be right back. UserEvidence is the customer evidence platform for go-to-market teams that makes it easy to collect customer feedback throughout the customer lifecycle to rate the best proof points and share it with your go-to-market team to use. When asked about how Gong uses UserEvidence, Udi Ledegor, chief evangelist at Gong, who is also part of the GTM Fund community network said, UserEvidence helps Gong showcase the value our platform provides by generating high quality customer proof. With the ability to scale content creation, we can dedicate more time to enabling our sellers and building relationships with advocates. Trust is everything, and your customers are the best source of trusted feedback. Highly recommend you check out User Evidence, the customer evidence platform. You can do so at userevidence.com forward slash GTM now. That’s userevidence, all one word, .com forward slash GTMNOW. Now, on to the episode. This episode explores what it takes to scale a go-to-market engine past $100 billion in ARR and be at the helm of revenue at that level. James Roth is the Chief Revenue Officer at ZoomInfo. He shares how the company evolved its go-to-market motion to move up markets repositioned itself as the go-to-market platform and launched its AI product Copilot, which took off. You’ll learn how to transition from transactional sales to value-led enterprise selling, build a full funnel go-to-market engine, and operationalize AI across every customer-facing function. James also breaks down ZoomInfo’s good co, bad co segmentation model, how they think about product market fit by segment, and how comp design and internal usage drive adoption. This episode offers a rare look into what enterprise excellence looks like at the over a billion dollar scale. As the saying goes, the fastest way to scale is to study someone 10 steps ahead. All right, let’s get into it. James, welcome to the podcast.

James Roth: 02:41

Hi, Sophie. Good to be here. Thanks for having me.

Sophie Buonassisi: 02:44

It’s a pleasure. Super excited to have you on. And there has just been a ton of momentum around Zoom Info. I’m excited to pick your brain on it.

James Roth: 02:53

Yeah, absolutely. Happy to share anything I can.

Sophie Buonassisi: 16:43

It sounds like you incentivized the right behaviors. You structurally made the shifts to have the SMB function, firing on all cylinders, but also carved that out from the enterprise. And you went from $750 million to over a billion in revenue. Were there any subtle or non-obvious go-to-market shifts that you found moved the needle?

James Roth: 17:06

You know, I think from a product pricing and packaging standpoint, you know, there’s– One of the key things, and it’s arguably our best performing business, I think as you look into the upmarket, the level of sophistication, if you will, is very different from a small business. And it’s sophistication, maturity, but also just resourcing. And so one of the biggest shifts is that we have a sales tool, a sales platform, that all sellers can and should use. But as you start to move more upmarket, it’s really more about data. It’s about funneling our data into the respective systems that a large enterprise might have. And so I think shifting from a, I’m going to go try to sell every BU and every single group these sales licenses into these more mature, more tech-forward companies that are centralizing data. And they’ve got a massive Snowflake instance, and they’re trying to drive all of these different territory analysis, TAM, and segmentation. One of the things we found is don’t fight that. really move to where the ball is going. And so, you know, with, you know, you think of a Databricks or a Snowflake or this, you know, this push, if you will, for enterprises to centralize data, you know, we have that as a tailwind for us, especially in an AI world where Larger companies are less looking for something outside of the box from an AI tool perspective. They’re really trying to build it internally. And I think for a period of time, our sales teams were trying to just go sell what we’ve always sold, which is a sales license. And shifting that in what we call our data as a service business, which is basically a direct feed of all of our go-to-market data from a graphics signals into these respective systems. It’s been our best net retention and best growth business. And so I think from moving upmarket, it’s one thing to say, okay, I’m going to go hire a bunch of salespeople from Salesforce and Oracle, and I’m just going to put them in there, and then upmarket’s going to happen. I think it’s understanding the product market fit by segment, and I think it’s understanding what the customers are really asking for upmarket, and looking at, we’ve got this every quarter, we look at what we call good code, bad code, like different products into different segments. and really understanding what the data is telling us, aka what the customer is telling us in terms of the retention of each product in each segment, in each vertical, and saying, okay, this is clearly having our best impact in the enterprise space, so let’s double down from a resources perspective here, and it’s exactly what we’ve done, and that data as a service business, again, 150% net retention, it’s growing significantly, 40% year over year, and it continues to do so, and I think rather than trying to fight that shift or fight that market dynamic that’s saying these folks are looking for less sales tools and they really want to build their own, that’s okay with us because we can be the underlying data foundation for it. So again, that’s one example of just having really the customer dictate or a segment dictate where we should really focus our resources.

Sophie Buonassisi :20:29

So in good code, bad code, it sounds like every quarter you sit down and you look at the data, thinking if a company wants to replicate this and take inspiration from this good code, bad code framework, what are kind of the step by steps to doing that?

James Roth: 20:43

Yeah, I mean, it’s very simple. I mean, we have over the years made several acquisitions. We’ve got a significant amount of products, if you will. And really what we do is we look at segment by segment, subsegment by subsegment, and we look at each of those products and we look at their retention, their net add, their sort of upsell or their new business. We look at each of the respective ways a customer could buy, renew, grow, or shrink with each respective product. And so what we see is that in the down market, again, a less resourced, you might have a RevOps team of one or two, you might have a sales ops person but you don’t have like a large data science team, you might see that we don’t sell any data as a service into down market. And you might see in the up market, you know, you have less kind of sales tools or certain products and you see, you know, basically looking at each of those saying, okay, from our down market perspective, these are the two to three best products that are being confirmed by net retention, being confirmed by growth, et cetera. I think a lot of times bigger companies are, you have the ability to fall into the trap of anecdotal feedback, or you have somebody that loves this product because they were the ones that led the acquisition, or somebody from that legacy product ends up in your leadership. And so you might get differing opinions in terms of where to focus and what to focus on. And I think every company, irrespective of size, deals with limited resources. And as I think about even our account management team, it’s roughly 400 account managers covering 37,000 customers. the ability to distract is a really, really big detriment, if you will. And so I’ll use an example. We just came to market with a specific data asset. SMB data is very challenging. There’s not nearly as many firmographic attributes or there’s not as much, there’s just not as much data on a 10 person company. And it was something that we’ve been trying to solve and we went out and we built this, what we call first mover advantage data set. And basically, Secretary of State of every state saying, give me every new business formed in the last week. And so we can then go to our customers and say, if you sell to small business, this is a goldmine of data because if you’re selling to small business, you really don’t have any way to understand. They might buy an office or they might rent office space. They might be in a garage. You just don’t know. But if they file and they have an actual business, you have this drove of data that can tell you these are the small businesses small businesses. And a good example of this is that we probably two years ago would have rolled this out to every salesperson saying, hey, go sell this as a data asset. And if you look at companies like Zscaler, Zscaler has 7,000 enterprise customers. They have zero interest in SMB data. That may change over time, but for the time being, if that account manager were to go to Zscaler and say, you should buy this data, it would be a waste of time, it would be a waste of energy, and frankly, you’d probably lose a little bit of credibility because they have no interest in that. Our ability to say, okay, this segment, This subsegment, this vertical, this company cares about this data. And so rather than taking it to 37,000 customers, we’re able to say these 2,000 are the ones via first party data that have talked about our SMB data in the past or on an earnings call, they’re saying they’re moving down market. So getting very targeted with each respective product into the right segment is really the foundation of this good co, bad co. So again, I think as I’ve spent many years at this, and you think about a room where you have a lot of loud voices, and you’ve got a lot of anecdotal feedback that might not always be driven by the right motive, and you’ve got this particular product that maybe new business sells a ton of, but its retention is very low, having that to say, Why fight this? Is there something that we’re missing in the product? Or should we kind of decommission resources and potentially put this product elsewhere and then look towards a partnership or something else like that? And I would say over the last two years, that has been a huge initiative for us. I think a lot of companies in our space raced to get to this last product you’re going to need. And so you don’t need these folks anymore. You don’t need these folks anymore. And everybody built out or acquired some of these bolt-ons that could say, OK, we’re the consolidator, if you will. And I think what the market really dictated is that they wanted best in class across each of these respective things. And so we’ve taken several products, whether we built them internally or we acquired them externally, and said, rather than continuing to try to move this rock up the hill, let’s go explore partnerships that get us to a better customer outcome where we can actually get this best of both, if you will. So again, it’s long-winded in terms of the good co, bad co, but that is something that we are hyper-focused on just to make sure that every resource that we have and everything we’re going to invest in, we know that it is the right move and the right bet before you end up one year significant investment to, you know, again, keep pushing the rock up the hill.

Sophie Buonassisi: 26:07

Well, I appreciate the long-winded answer, as I’m sure everybody else does, too, because it really is, it sounds like that data-driven approach where you’re getting really, really tactical and relevant with the data. And as you scaled, you go upmarket, what was the hardest part?

James Roth: 26:26

You know, I think building the machine, if again, back to account loads, you know, our CSMs in 22 when I first started were all on activity metrics, again, which is very much a down market metric for a CSM. And so I think setting the foundation right to the earlier point, it’s one thing to say upmarket. It’s one thing to hire some expensive upmarket salespeople. It’s another thing. to have a truly upmarket engine in that full customer journey. And that’s, you know, new business was not segmented. New business was a one size fits all. new business, close anything that comes in, whether it’s inbound or outbound, and shifting that new business machine towards a segmented model where you had enterprise new business reps closing enterprise deals that then went to enterprise onboarding implementation delivery and then went to enterprise CSMs and enterprise account managers. It sounds rudimentary, but we had to do all of that so that we knew that the acumen of the new business rep and the sort of products and the solutions that they were selling to land then went to the right folks to onboard and implement that then went to the right folks to support and ultimately manage the account. And so I would say the first six months was very much just kind of table stakes in terms of we have to set it up this way. And then I think the hardest thing to your point earlier was on the talent side. is saying, okay, you can’t just go out and hire everybody externally because it’s going to take them three, six, nine months to ramp. You’re going to have this gap and you’re going to have a bunch of, you know, quasi disgruntled people that have been at ZoomInfo for a long time that got kind of passed over for this. And so I think one of the hardest parts was just getting the right teams or the right individuals into the right places. And this is all, by the way, while we’re a public company, so you’re kind of changing the tires at 80 miles an hour. Yeah. We had an example where one of our best reps over, call it 10 years, moved them into enterprise because they were one of the best reps. Six months later, they said, I don’t love the enterprise. I really liked the action of closing transactionally. And so moving that person back into a down market role, you know, there’s just That takes a lot. It takes a lot of manager conversations. It takes a lot of leadership conversations. And then I think more importantly, training the folks and building out a framework for true upmarket, having a sales methodology, having a rigorous forecasting system, like none of those things existed. And so having to build those out, it just takes time. And then you have that believability of someone who’s been here for 10 years and seen extraordinary success that says, I’ve never had to use a sales methodology before. I’ve never had to forecast like this. I’ve never had to, you know, dealing with that kind of noise, if you will, was probably one of the more challenging parts outside of just kind of setting the foundation for a real upmarket business. And then I think educating the customers, again, on these are the products that you’re accustomed to. These are the products that we’re moving towards in an enterprise standpoint. You know, that just comes with the territory, I think. And I think having the right teams and the right Emotions certainly help that. But yeah, I would say just the shift in mindset, the shift in kind of culture, what we rewarded, what we were celebrating, that was probably the hardest part now looking back on it.

Sophie Buonassisi: 29:52

Makes complete sense. It sounds like just lining up everybody so that everybody’s firing on the same wavelength would be extremely challenging. Yes. And you didn’t just take the company upmarket, you actually repositioned the entire company. When did you know that it was time to kind of put stake in the ground with changing your ticker symbol to, and just for everybody’s context, ZoomInfo changed its ticker symbol to GTM?

James Roth: 30:19

Yeah, you know, I think what’s interesting is For the better part of 10 to 15 years, one of the core value propositions was fix your underlying CRM data. CRM data, if you’ve ever been in sales, if you’ve ever been a sales leader, I mean, there are meme pages about this in terms of just like, hey, go update Salesforce, update Salesforce, update CRM. And I think for a long period of time, everyone knew that it was a problem. And I think these companies that are, by the way, amazing, there’s no discredit to them, but they’ve become kind of more backend systems, finance, IT, they’re kind of the box, if you will. And I think when you think about go-to-market, there are companies that, frankly, have gotten so big and so good at what they do, one of the key areas in terms of like real go-to-market data or a go-to-market interface within those, it really didn’t exist. And so we saw the opportunity just given, again, being known for sales and marketing and being kind of the de facto in that space, 37,000 customers, call it 70% of the Fortune 500, using us as this central go-to-market data foundation or go-to-market intelligence foundation. We saw this opportunity, the ticker was available, and I think becoming synonymous with all of the different go-to-market tools, our point of view is, You know, if you look at forecasting, there are phenomenal companies that focus on forecasting. Those need great go-to-market intelligence. If you’re running forecasting over just CRM data, there’s a whole world of first- and third-party data that you don’t have. You know, the example being, if you’re trying to sell to, you know, me at ZoomInfo, and you don’t have the recent earnings call data, and you don’t have the fact that we just promoted a new CFO and promoted a new CTO, like, those don’t typically exist in CRM. Because that rep, the likelihood that they’re inputting that into the information, you’re getting a fraction of the picture in your forecast. And so forecasting is one. If you think of sales sequencing tools like Outreach and Sales Loft and great, phenomenal tools, if they’re not getting all of this go-to-market intelligence across first and third-party data, you’re just running these spray-and-pray sequences. And so it’s kind of like this hub-and-spoke thought of each of your respective go-to-market tools tools or functions, if you will, require great intelligence. And we really wanted to be at kind of the epicenter of that, not saying that we’re going to go build a forecasting tool or we’re going to build a sequencing tool. We want to partner with the best of those and make sure that they’re getting all of their conversation, all of their email data, all of that rich first party married with the best third party data to then go drive, whether it’s a forecasting input, whether it’s a sequencing, whether it’s a workflow, whether it’s signal to action, making sure all of those are based on great foundational intelligence. We saw an opportunity to kind of exemplify that, if you will, in changing the ticker. Again, changing the ticker doesn’t really change anything except for the ticker, but I think really centering ourselves as that foundational go-to-market intelligence that fuels all of your different go-to-market outcomes, inputs, outputs, et cetera, was really the mission.

Sophie Buonassisi: 33:47

Definitely. I mean, while it is three letters, I think the impact is massive. Overall, it felt like a wave around go-to-market and around your positioning, like you said. So I think it’s a good lesson for everyone, too, around unconventional ways of repositioning yourself or painting your own story.

James Roth: 34:08

Yes.

Sophie Buonassisi: 34:10

And you launched Copilot. Congratulations. Yes. And in less than six months, you achieved over $100 million of ACV sold. That’s incredible. Staggering go-to-market achievement, proof point for the product’s value. What was that go-to-market motion like? Because that is an incredible timeline for that revenue. Yeah.

James Roth: 34:31

And we, again, it’s all public. We just publicly announced that we’re over 220 now.

Sophie Buonassisi: 34:38

Congratulations.

James Roth: 34:40

Thank you. You know, I think… a big part of it back to the earlier point on why I decided to come. Anytime you have a new product launch, if it’s something that you can deploy to the teams and they can see it, touch it, feel it, believe in it, and more importantly, use it, on a regular basis to see, okay, I’ve used SalesOS my entire career, which was our legacy platform. Now I have access to Copilot, which takes my territory and it prioritizes based on signals and based on firmographics and based on best fit, and then basically tells me what to do. Rolling it out to our team first, it’s one of the unique advantages that we have. because if you think about a beta or if you think about early access, we got to go give it to 1,600 go-to-market professionals, get their feedback, who’s using it, who’s using the heck out of it, and then more importantly, as we optimized and made the platform to their liking, you think about an org of that size, you’re gonna get a little bit of everything in terms of feedback. We felt incredibly confident with the utilization internally, with the feedback that we were getting internally, And then you basically flip a switch, and from a go-to-market standpoint, it’s go to your customers and just show them how you’re using it. That is a unique opportunity for us. where if we were releasing, you know, again, a cybersecurity product or, you know, it’s hard to get that. I’m a salesperson. I’m a marketer. I use this. This is how I use it. And then basically a demo is a day in my life. You know, even at my level, I use Copilot. You know, we’ve got a tool called Account AI, which basically summarizes all of the first party data and all the third party data. And so I can see, like, I used to have to go to my reps every time I’d go to Dreamforce or go to Money2020 I’d have to say I’m meeting with these 30 customers. Write me a brief for each of them. Now, I just go into Account AI and it gives me the summary of the conversations, summary of all of their, you know, public filings, a summary of who we’re talking to, the good, the bad, the ugly, and then there’s an AI, you know, queryable, I can build an account plan, I can build a prep doc, I can ask it questions, what’s Sophie going to be most mad about, what’s she going to be very happy about. And so, you know, for the first six months, it really was as simple as get in front of your customers, get in front of your prospects and show them how Copilot has changed the way you do business. And so from that perspective, it was kind of a dream scenario where you really, it was that simple. If you as a salesperson are using this product, which they all were, and then you can go show what it’s like to, you know, basically take old sales OS, which is pull out of Zoom Info. I want to contact Sophie. I have to pull that profile out of Zoom Info. to now a push motion, which was AI driven to say, this is why you should talk to Sophie. This is what her business is doing. This is what they just said on an earnings call. And they just raised $50 million in funding. Here’s a pre-built message based on all of these signals. Anyone in sales, anyone in marketing would say, wow, this is really impactful. I think aside from just the ACV number, which of course we’re proud of. I think one of the things that we’re very proud of is seeing that net retention follow. It’s one thing to release a new product. It’s another thing to see significant upticks in the net retention or the utilization or the health score of that product. And again, it was our bet, which is if we force people to have to pull information out of ZoomInfo in a world of conflicting different tools that they’re using on a regular basis versus something that can run in the background, either in CRM or alerts you in Slack or alerts you in Teams or alerts you via email to say, here’s all of this data that I’m now summarizing based on your product. I’m summarizing based on your territory and I’m shipping it to you. I would say that we are far more excited about than just the ACV number is seeing not only the growth, you know, kind of the growth in footprint. SalesOS was a lot of times an SDR deployment where it was like, hey, it’s a prospecting tool. It’s going to help you prospect. And we’re seeing growth and co-pilot into account management functions, into sales leadership, into CSMs, because getting that full 360 picture across all that first and third party data, that’s where we’re seeing more growth in the actual footprint if if you will, of the overall go-to-market team, and then the retention of that even coming up after we’ve hit our first year of anniversaries.

Sophie Buonassisi: 39:25

Incredible. And not only did you publish, or not only did you launch Copilot, you actually published the State of Go-to-Market Intelligence 2025 edition report. And I was pretty shocked, actually, from a stat in there. I think it said something around people are spending millions of dollars on AI, yet people are seeing under five percent revenue lift what’s where’s the disconnect happening there between revenue lift and ai utilization

James Roth: 39:53

you know i think we’re still so early into ai if you will and i think folks are clearly being pushed to adopt ai and it’s tremendous if you’ve ever used any form of it, it does dramatically increase your productivity. I think From an ROI perspective, internally, we put it into three buckets. You’ve got reduction of number of people that you need based on the AI being able to do certain things. And you’ve seen very large companies announcing large kind of efficiency-driven, we’re using AI for this. We used to have 100 people here. Now we only need 10. So you have that. And then you have the actual productivity gain, especially in things like go-to-market, where one example that we use internally Because of all of this data that we have, we took a case study on a large deal that we had one of our best enterprise reps do. And they had, call it, 35 hours worth of meetings in a given half, half a year. And then they had 85 hours of prep, deck creation. And that’s all, again, it’s all data that’s driven by your time spent in Google Sheets, time spent in Google Docs, et cetera. And then the actual process presentations that they put together across a buying committee of 20 people. And the ability to shrink that 85 hours down to call it 10 hours when you have AI that can build the decks, they can do all of that prep work and do all the summary for you. There’s an unlock there that I think is still yet to be articulated in a clear ROI. You know, it’s very different to say, hey, we’re going to buy this thing and it’s going to drive up win rates by 10%. I think there’s a layer under that within go-to-market where you’ve got, okay, the ability to say, we’ve served up this many signals, this many signals have not been actioned. And so there’s kind of a hidden ROI to say, if you can just drive that team to go action against these signals, again, that’s very specific to us. In the overall state, I think we’re still in this early days, AI is cool, these use cases are cool, but we haven’t really figured out how to unlock either the incremental productivity or the, for lack of a better term, shrinkage of certain teams that are doing this. There’s a lot of noise on AI SDR. Is there going to be an SDR function? And I think that there have been several cases where largely unproven in terms of a true AI SDR. Some of that’s regulatory. Some of that’s just the ability to do outbound into certain people with an AI bot. But again, I think all of these are probably still early runway. And as we start to see more and more of them come out and drive that incremental productivity, take that 85 hours down of that rep that’s building decks and doing prep work. But then more importantly, I think the biggest unlock is to be able to say, this is everything happening in your territory. These are the signals that you should go action, and then a maniacal focus on the folks that either action that or don’t. I think that’s probably the next phase of unlock for us specifically, but I think in general in the market.

Sophie Buonassisi: 43:16

And how are you currently operationalizing AI? That sounds like the next move that you’re moving towards, and obviously you’re probably using your own product, but are there any external or overall systematic ways that you’re leveraging AI right now?

James Roth: 43:29

Yeah, I think we are, like many companies, our size. There’s a build and there’s a buy. And some of the lower-hanging fruit use cases, like support, You know, I think if anyone is looking at where they can make an impact the fastest, support, you know, deflection rates, that AI is incredibly strong. And there’s a variety of companies that do it very well. But, you know, support is where we started low-hanging fruit. You have like tickets and intake. You know, basically what we do is we look at each one of those systems, if you will, that requires a lot of manual effort, but it’s more just kind of hands-on keyboard work non-strategic effort if you will and then how can we go augment that and then i think one of the biggest areas and we built this internally again not for sale but just for our own internal we built a company-wide chatbot that basically allows you to query everything so we’ve got our snowflake instance we’ve got crm we’ve got every bit of anything happening within zoom info every every employee now has access to this chat bot and so i can say show me the top 10 sales reps from h1 and write a note from the cro to them congratulating them you know that’s something that probably would have taken me a couple hours over a weekend prior Now I can have it in an instant and it is better than most anything that anyone’s putting together. And you think about the deck creation, you think about the prep work. Again, that is a product that we go to market with. But those are the, I think, low hanging fruit use cases, our ability to, again, not necessarily cut heads in support, but if you can take, call it 30 tier one support people out of the mundane password reset, the easy, if you will, things that AI can do, and then repurpose that capacity to go then hire Tier 2 and Tier 3 for a much more complex support instance. I think it’s a big reshift in where do we want to make bets, and then where can we subsidize those bets in some of these lower-requiring, resource-heavy areas. areas like support. So that’s really the mission we’ve been on. And I think there’s a handful of things, if you think about, you know, again, in the account management function or the CSM function, the ability to take, you know, tools like Gainsight, tools like CRM and say, okay, these things traditionally weren’t really talking well together. How can we put this AI wrapper over it so that at any given time, an account manager can say, what should I talk to Sophie about? You know, where does she want to grow? Where does she, you know, I think that’s where we’re most focused, both self-serve internal productivity, but also as we can optimize Copilot and our products to serve those particular use cases and go to market. Back to Copilot, it’s a great way for us to test out where is the unlock, where are our teams having the most success. And so we’ve got this internal competition, if you will, between like this chatbot build and account AI and all of these different areas to see like, okay, who’s using what the most? And then how can we productize that? Because if we’re struggling with it, I know a lot of other go-to markets are as well.

Sophie Buonassisi: 46:42

It sounds like a similar system to when you went on market, you’re actually looking at all the different products or systems and then measuring the utilization and doubling down on what is being used and is effective.

James Roth: 46:55

Exactly. Exactly. And I think seeing the utilization real time, like having that telemetry, both with our internal reps and external customers, it does allow you to, because again, I think a lot of companies will have product and engineering and they’ll think something is amazing. They’ll build this tool that really, really proud of. but then the market might not utilize it, but they’re all in on it. And so I think the ability to have that kind of internal AB testing to say, you might be really excited about it, but the teams aren’t using it. And so, especially in, again, the account management base, we saw account AI, you know, we’re talking 50% increase on utilization. It’s very, very easy to go in and say, okay, we’re really focused over here, but look at this thing. it is performing better than anything we’ve ever brought to market. We need to go iterate and invest here. And then on the go-to-market side from an outbound and a marketing standpoint and then a sales standpoint, we know that this is the thing. And so let’s go double our efforts in getting this thing in front of everyone else. In this world of AI, there’s 15 other tools that we sell. And then everyone else has 15 other tools. And so knowing what is driving the adoption, driving the utilization, what customers are really using and wanting, the ability to then go make that the main thing, I think is really important.

Sophie Buonassisi: 48:16

Sounds like you take a very data-driven, almost signal ingesting approach to a lot of things. How do you translate that to your teams? How are you measuring your teams? Is it activity, signal? What does that actually look like for you?

James Roth: 48:29

Yeah, it’s a great question. And I think we are heavily data-driven, and I think we have to be. In terms of the metrics that we used to track versus the metrics that we track now, there weren’t a lot of great options traditionally. You had a call detail report. You saw how many calls SDRs were making, how many emails they were sending, and it was primarily a volume game. And I think one of the best… kind of transitions, if you will, is the ability to track really what matters more. And so, you know, talked about the signals actions. That is probably the number one thing that we track. If you think of being a salesperson and you think of having to make a hundred phone calls in a day, blind phone calls are back in my day, the door knocks and you have no idea if you’re knocking on the right door, the wrong door, if it’s a government building, if you’re going to get thrown out, you have no idea. And now we can basically say in your territory, in your total addressable market, in your particular 50 named accounts, you have five that are exuding significant signals. They just hired a new CRO, they just said on earnings that they did XYZ, one, two, three, they just raised funding. Why on earth would you not action that? And so in the old days, you would have had those 50 accounts. You would have had no idea what was going on inside of them. And it would have just been a volume game. I’m going to call each of them until somebody picks up and tells me what’s happening. Now we can take it to a point where it’s, here’s exactly what’s happening. Here are the five out of your 50 that are exuding the highest density of signal. Looking at the signal to action and the action taken and what that action is, I think some of the tracking that we have today did not exist five years ago. And so the ability to say, these things happened, these were the outreaches that you did, here are the conversations that you had, the ability to to track those things and then AI summarize. I mean, that’s one of my favorite things is like, show me the five best AMs. And you’ve got conversion rates, you’ve got all of those kind of legacy metrics. But now you can take it such a step further to say, okay, you actioned the signal, good. The actual outreach was good, but here’s your first call. And again, conversational intelligence, AI summary, good. you really missed the boat here. And so you can track into a level of detail that just wasn’t feasible five years ago. And I think looking at overall volume metrics, it’s an indicator, don’t get me wrong, we still look at it, but it’s a fraction of the pie in the overall, like what we’re tracking. We have a tool that basically tracks all kind of systems that would touch your day to day. And I still think seeing someone whose activity level drops significantly is always an indicator. So you have that in the background. But I think what is far more important to coaching, developing, training, ramping, are that next layer of what are the activities you’re taking? Which of those are effective? Are you following those particular signals? And are you following them effectively? We just didn’t have the ability years ago to do that.

Sophie Buonassisi: 51:44

Makes sense. And James, when you joined ZoomInfo in January of 2022, the company was already at a quite impressive ARR, about over 700 million ARR. But under your leadership, like we’ve broken down, ZoomInfo crossed the $1 billion revenue mark within the year, which is nearly a 47% jump. When future CROs study you, what’s one leadership advice or principle or system that you hope that they steal or take inspiration from?

James Roth: 52:18

Yeah, I mean, I don’t know how many will study me. I think I would love to go back to our market cap back then. You know, I think all joking aside, you know, one of the key things to like the what’s hard in that shift up market, you know, we’ve been in this sort of shrinking our down market which we know does not retain as well and all that while growing our up market you tend to on paper look like you can run in place so yes a billion two in revenue we spit off close to 500 million dollars in free cash like we’re very proud of those things but you know from a growth perspective and this is to the what i would hope people take away You have what we know is right, which is we want to be a business that’s 75% upmarket and 25% downmarket. The willingness to put your helmet on and go for a year, basically say downmarket’s going to shrink while upmarket grows. And so from an overall growth perspective, it’s going to be flat-ish. It’s really hard to do that because you get bored, you get investors, you get shareholders, and while everybody knows it’s the right thing to do, at a public company every quarter you have to go tell them basically what you’re doing. And so, you know, we made that decision, and I think the commitment and the willingness to build for a future great company versus it would be really easy to just say, go turn on the SMB spigot. We’ll go add $20 million of SMB business, which is relatively easy in the grand scheme of things, and it will make our growth in the quarter look really good, and everybody will give us high fives. I think, you know, that… that is the the easy way to do it and i think for what we are building here we’ve gone from basically 50 50 up market down market to now we’re 72 28 72 up market it is really hard it’s not a ton of fun and i think that commitment or that optimism of saying when we get to a business that is 75 up market the net retention is well north of 100 and the growth is in the you know high double digits And the SMB business, albeit retained 60%, 65%, that’s okay because it’s a much healthier portion of the SMB and it’s a smaller overall number. I think if anybody’s listening to this, you are always faced in this role with like, oh my God, I got one quarter. And so I’m gonna go do the things that in this particular quarter are gonna make the board happy or the investors happy. And I think balancing that, And it’s very similar in the shift up market. If you can say, this is where we want to be in a year, this is where we want to be in two years. And several folks from my background, and they’ve ended up in the CRO role, and this is always the advice that I give them, is like day one in the seat, you’re going to want to go in and make these little incremental tweaks, and you’re going to say, hey, look, this was over here, and we increased it marginally, like I’m doing great. And they don’t take the time to say, these are the big, hairy issues. And it’s going to be a long slog to get there. But this is what we’re going to do over this quarter, next quarter, this quarter. And I think getting really prescriptive on this is what we’re going to do with down market. This is what we’re going to do with up market. You know, you’re not just going to see up market pipeline overnight. It’s going to take time. And so I think plotting that out versus taking the kind of quick wins quarter to quarters. Harvard Business Review did a study where CROs like the number one lowest tenure, fastest fired, all that fun stuff, which certainly plays through, I think, everybody’s head. And so I think the tendency or the advice that I would give is when you first get somewhere or you’re in this role, whether you’re promoted or you’re external, is don’t fall into that trap of just saying, okay, this is the run rate. and I’m going to go do a little thing here and I’m going to do a little thing here and I’m going to come in with a little bit better of a result and I got to hope the macro changes and helps me out so that the board gives me a high five in the first quarter. If you don’t go after the big things, you will be in six to nine months in a hot seat because you will have missed your opportunity to articulate what those big things are quickly and then the commitment to doing them through what are challenging times, and we have certainly seen our share here, when you take a business whose net ad was primarily SMB for the great years, and then you shrink that SMB significantly, and now you gotta go do it in where it’s really hard in the upmarket, you always have that in the back of your head, and we joke internally, like, should we just go turn SMB back on? because we could go show a massive growth because upmarket is growing. And if we went back into the, like those types of things, the commitment to the company that you want to be or that turnaround that you want to execute has to outweigh kind of the insatiable need for short-term results. And that’s really easily said, really hard to do.

Sophie Buonassisi: 57:27

Discipline and commitment, long-term game. Sounds like you’ve got the long-term vision, you’re playing the long-term game, which is speaking our language, James.

James Roth: 57:35

Well, we’re trying to. We’re trying to.

Sophie Buonassisi: 57:38

That’s incredible. Are there any books you’d recommend for anyone listening that have really inspired you throughout your career?

James Roth: 57:45

Oh, that’s a hard one. You know, I’m always a big fan of the… I’m not a big fan of the self-help-y, kind of catchy titles. I love business biographies. And not just business, basically anyone who’s done anything great, I love to read their books. So I just finished Steve Schwartzman’s. Jim Mattis is a general. Call Sign Chaos was great. Bob Iger, Ride of a Lifetime, he was the Disney CEO. I love those because there are these amazing stories and there’s tons of tidbits, and I try to get as many of them as I can across industries, across different vocations, and just love to kind of map back what awesome people do versus somebody who you’ve never heard of who writes a book with a catchy title that’s sort of self-help guru. I’d much rather read about… you know, Bob Iger or some of these folks that have done it and done it in like an awesome fashion. So those are a couple. You know, Kim Scott is another favorite, especially from a leadership standpoint. Kim was at Google under Sheryl Sandberg and then she went to Apple. You know, Radical Candor. Those are probably the ones that I recommend is just any sort of great success story, biographical, this is how they did it. And then trying to map back as many of those things that are kind of relatable to what I’m doing or just great people in general. That’s where I spend most of my time in reading those. And, you know, they’re fun to read because these people have amazing stories. And I just think they’re chock full of great information that are relevant.

Sophie Buonassisi: 59:31

They’ve been there. They’ve done that. Maybe we’ll read a book from you someday. And where can people find you if they want to follow along?

James Roth:59:40

Yeah. I mean, I’m on, I try to be as active on LinkedIn as one can with a busy schedule. Um, and so I try to get to the messages that come in. Um, you know, zoom info. We, we are out there quite often. I mean, people can reach me via email. People can reach me. If you use zoom info, you can reach me via email, via text. Um, I’m probably better on text. I’ll regret saying that, but much better on text, just given how crowded LinkedIn box and email boxes are. But, um, You know, I try, especially knowing that our users are primarily salespeople and I’m a salesperson at heart, I try to answer as many inbound requests as I can. But yeah, my LinkedIn is James Roth. It’s probably the best place to find me. And then, you know, I’m pretty accessible.

Sophie Buonassisi: 1:00:28

Amazing. Well, who knows? You might get a couple texts after this, but either way, we’ll drop your LinkedIn in the show notes for everyone. Probably skip the phone number in there. But thank you, James. This has been a phenomenal conversation. Can’t wait to share it with everyone.

James Roth: 1:00:41

Yeah, thank you, Sophie. Thanks for having me on.

Sophie Buonassisi: 1:00:44

Absolutely. Thank you to our listeners. Thanks for joining us, and we will see you next week.