18 Months, 7,000 Customers, 67% Support Resolution: Inside Intercom’s FDE Strategy ft. Diego Ballona (Intercom)

And we've seen a lot of success with the model because it makes the engagement of the customer feel like we're there together to solve the same problem together with incentives aligned.

Josh Schachter [Host]:
You're listening to Unchurned, brought to you by the Gainsight podcast network. Most SaaS teams think value shows up after the product ships. But what if value actually shows up when engineering ships with the customer? At Intercom, Diego Ballona leads something called Forward Deployed Engineering. FDE for short. It's not professional services, it's not support. It's engineers, product leaders, and data scientists sitting directly inside real customer problems early so outcomes happen faster. In this episode, we unpack why the FDE motion matters, how it accelerates time to value in the AI first world, and why it's changing how SaaS leaders should think about customer success, pricing, and even product roadmaps. If you're building AI, selling outcomes, or wondering how engineering and CS finally work as one team, this conversation might change how you scale value.

Josh Schachter [Host]:
I'm Josh Schachter. This is unchurned. Hey, everybody, and welcome to this episode of Unturned. I'm your host, Josh Schacter. I'm here this week with Diego Bologna.

Josh Schachter [Host]:
Diego is the senior director of engineering at Intercom. Now, Intercom's a big company. They've got several directors, senior directors, et cetera, of engineering. Diego is the one who leads. I think the most exciting part of the business in SaaS these days is he runs the FDE. I don't know. Do you guys call it the FDE motion at intercom, Diego?

Diego Ballona:
Yeah, we do. Yes.

Josh Schachter [Host]:
Yeah. Okay. Oh, you didn't want it. You didn't want to, like, trailblaze your own new. Because there's all kinds of different names and synonyms that people are coming up with, but you guys are right. Fde. So Diego runs the fde, which is a fin, right? I mean, effectively the fin part of the business. Welcome to the show, Diego.

Diego Ballona:
Thank you. Thanks, Josh. It's a pleasure to join you today. I'm excited to have this chat.

Josh Schachter [Host]:
So, you know, we talk all things SaaS and more specifically, customer success as part of this program, as part of being at Gainsight. And we're going to, you know, sprinkle that in Here, but we at gainsight, we're building atlas, which is our agentic play. Autonomous agentic, you know, thing of the future. We've been inspired by what you guys have done with Fine. And when I first reached out to you, honestly, it was just like out of pure curiosity to learn some of the war stories and how you guys have been able to accelerate, because at least from my purview from the outside, I don't have any special information. It seems like Intercom is really at the forefront with what you guys have been doing with your agentic business and have been very successful. So first of all, congratulations, and second of all, get ready to tell the entire CSS community how you've done it. Are you ready?

Diego Ballona:
Yeah, I'm ready. And like, there are multiple ways to do it. Like, we, we found our. Our way to do it. And I appreciate the kind words. Thanks, Josh.

Josh Schachter [Host]:
Appreciate it. Yeah, of course. All right, so should we go back to the origin story? Like, tell us a little bit. You know, first of all, just give us a little bit of set a little bit of context. I'm assuming everybody knows Intercom, but maybe there's 1% that don't. So give us for the 1% out there, you know, what is Intercom? You joined Intercom and you weren't immediately in this role. And tell us a little bit about the origin of your new role in this group.

Diego Ballona:
Sounds great. I joined Intercom actually to run a product engineering team managing FIN's knowledge and data features. So basically very closely related to all the rag side of the product, but also like tool use. We were starting to experiment with tool use in the product. And basically the more we built product over that year, the more we realized that actually building product was a part of it, but actually driving customers to success should be a big goal as well of our group. And we kind of walked into this FDE motion. We didn't set ourselves up to basically go and do an FTE motion. We figured that this was important because some of our customers were getting mixed results with fin.

Diego Ballona:
Some of them had really good results, some of them got results that were not excellent. And we wanted to understand why and what we learned. Yeah, all we learned is like, there are multiple components to the equation, one of them being education, which we heavily invested self manageability, which is incredibly important. And also like this white glove motion, which is the FD motion that we have.

Josh Schachter [Host]:
Well, give me the timeline around this, like around how long ago? Because you guys been doing this for a little while. This was, you know, maybe if this was three months ago, you would have just been like, okay, we need an FDE team. Right. So you kind of stumbled into it. But that's because it was not three months ago. When was, what's the timeline of all this?

Diego Ballona:
18 months ago, roughly. We started doing this. Yeah. At the time there were a lot of companies like doing this like at a scaled way, not companies the size of like Intercom for sure. You know, like with all the customers that we have and so on and so forth. And it was very quick realization as well. Like it was obvious, it was glaring that this was needed. And I think it's a byproduct of AI really.

Diego Ballona:
It's new, right? It's very new. Customers need to be educated, they need to be taught how to do this. And there's a lot of complexity also, especially in customer service. Like there's just so much siloed knowledge when it comes to your knowledge around your business, the guidelines. Right. Like the policies and so on and so forth. And we very quickly realized this was necessary really.

Josh Schachter [Host]:
If you can share around how many customers does Intercom and Fin specifically serve and then what segment of that are you guys now focused on?

Diego Ballona:
Yeah, starting with fin. We have nearly 7,000 customers using Fin. And the average outcome that our customers get is 67% resolution rate. So out of all interactions that you're.

Josh Schachter [Host]:
An engineer man and you're, are you trying to audition for sales here? You're going straight to the outcomes, to the value based proposition. I love it. 67% resolution rate. All right, we got that up front. Diego. I love it.

Diego Ballona:
I think it's all this time spent like working with customers made me realize that value is really important. That's right. Details. But yeah, so I think we have a very sizable like AI business and Intercom, of course, like we've been in this industry for 14 years, like with our help Desk products, we're very well known for that and we've seen assumption growth really in our AI business over the last three years.

Josh Schachter [Host]:
So, okay, so you've scaled it to 7,000 customers. Fin, by the way, is a part of Intercom, although it's, it's. To say it's a part of Intercom is almost like belying like you go on your landing page and it's like the right side is Intercom, the left side is Fin. I mean it's, it's almost its own business now. Right. It's become a big beast to its credit. So. But, but Finnis part Intercom.

Josh Schachter [Host]:
It's the effectively like the autonomous agentic support vehicle. Right. Product of intercom. Yeah.

Diego Ballona:
REI agent. Yeah, exactly.

Josh Schachter [Host]:
You've got 7,000 customers that you're servicing right now. Are they the full gamut of SMB, Mid market, enterprise, strategic or where does that lie?

Diego Ballona:
Yeah, yeah, we offer the agent to all segments and we have customers in enterprise, we have customers in mid market, we have SMBs, and also across the globe, really, like in Europe, North America, Asia Pacific, Latin America and so on and so forth.

Josh Schachter [Host]:
And yeah, and when we're talking about fd, excuse me, this for deployed engineering motion, you're not going to companies of 25 people. It's the more strategic accounts. When you started it, who. I don't know if you can name names, but. But who or how many folks design partners did you start working with?

Diego Ballona:
Yeah, we started actually with a very small number of customers and it was an experiment. Like we wanted to drive some customers to get better outcomes within. And we chose five customers. They were a mix of all these segments that we spoke of and customers like located in different regions as well. And we wanted to validate the hypothesis like is there value in partnering more deeply with the customers and which value there is? Of course there is customer value, but we knew there was more than that in it and it was very quick. Josh. Within the first month we learned so much more that we could be doing in our products to make it more self servable, more self manageable and so on and so forth that we realized that our fortified engineering motion doesn't exist to serve as a consulting arm of the business. It's much more to drive customers to success.

Diego Ballona:
And as a byproduct of that, build a better product, build better tooling and serve every single customer. Right. We want every single customer to be able to create perfect customer experiences. Really. And that's possible with AI now, right? It was not possible in the past and it's possible with AI now.

Josh Schachter [Host]:
So if I'm a CS leader listening to this podcast and this is piquing my interest and I don't currently have NFT emotion in my organization and it could potentially apply and maybe you can tell us, tell our listeners, you know, which organizations and setups you think this applies the best too. So let's talk about that. But how did you get the mandate and the ability to forge ahead with this? Was this something that your CEO slept on and it came top down and said, hey, we need to do this? Was this something that you and Janan Pang, you know, VP of cs, came together and had to make A, you know, build a case for. So first of all, yeah, like what were the motions you took to assemble, get the right to move forward with this FTE and then listening for folks listening out there.

Diego Ballona:
Yeah.

Josh Schachter [Host]:
Who is this most applicable for? Who would you recommend it to?

Diego Ballona:
Yeah. So Janan is our VP of Customer Success Solutions and also services and he was also new to Intercom like I was when we were figuring this out, right. And there was a leaders off site. It's like an event that does every year. We were reaching to similar conclusions from different angles and we got together and we were like, hey, there's value in this, we should explore it further. The first mandate of doing something with the five customers was the scene that we made. We're like, let's try this out and see. And with the findings that we had, it was very clear, it was jarring.

Diego Ballona:
We were like, hey, we can provide so much more value to our customers and also to Intercom. Right. As a consequence of like us charging for outcomes, not for conversations or seats and so on, that it wasn't a very hard business case. Like we had support from our SVP of engineering, Jordan Neal Des, one of our co founders, was very, very supportive of it as well. And it's like it all happens very, very quickly. After those first findings, we had a very clear business place. And I guess my main advice for folks listening Josh is don't also complete that. This is a solution for every single business, right.

Diego Ballona:
Like you see a lot of companies experimenting with an FD motion and they're not even quite sure what is the outcome that that motion serves. They think it's part of like selling or success and they get involved in models where you have fdes, managing agents for the customer on an ongoing basis and that doesn't work financially and it has severe, severe. The passive issues at Intercom is that in the early stages of your customer journey with Intercom, it makes sense to have an FD motion because we accelerate your time to value, we get to see outcomes faster, we return your investment faster. But the real value of fin in our AI agent is basically for you to self manage it, for you to be able to actually build your own procedures, be able to drive insights towards our business without depending on the vendor really. This is not like consulting or professional services, like rebranded, like it's a different motion.

Josh Schachter [Host]:
Diego, I want to talk about the team structure that you're working with here. So did you have to change anything in the assembly of your engineering team and the other parts of the go to Market that came in as you established this forward deployed engineering motion. How does the structure work today?

Diego Ballona:
Yeah, so we actually tried a few different models actually Josh. We tried different ratios and compositions of different people on the team really. And we got to the realization that you need three roles and sets of skills. You need a product manager. That person basically is accountable for multiple accounts. They are on point to actually treat the deployment and the evaluation of the AI agent as a product really. So they come to the customer, they tease apart the requirements, the problems to be solved, the outcomes that the customer wants to drive in collaboration with an AE or a csm. Really.

Diego Ballona:
They also work closely with engineers, which is the second role set of skills. Engineers on rfd Emotion. A large part of their work is like configuration. It's not like shipping custom features, it's like fine tuning prompts, running experiments, iterating on performance of the agent and giving insights into customers, enabling the customer to be able to self manage their own agent as well.

Josh Schachter [Host]:
Is it a different type of engineer then that is best fit for this role?

Diego Ballona:
Absolutely, yeah. It's like an intersection of product engineering. Same bar we have for product engineering we have with our fdes, but there's a lot of sales and go to market skills that that type of engineer needs to have. Right, like how to engage with the customer and how to communicate value. We were joking earlier about driving with the conversation with value. That's something that we expect our engineers to do because we sometimes engineers, we fall in love with a solution, not with a problem. And that's a very important skill that you need to have on FTEs. And then the third role is basically data science as well.

Diego Ballona:
And that's something that I haven't seen many companies do. But we chose this path at Intercom because we think there's a lot of value in having an analytical mind working with a customer and understanding what they care about. No customer ever comes to us and they just want to move the metrics that we care about. Right. Which is resolution, rate.

Josh Schachter [Host]:
Diego, I want you to go deep into that with the data science because that's really interesting. I want to go deep into the product and other stuff as well. The product management, but data science, how specifically? As specific as you can get. How do they engage in each of these customer relationships?

Diego Ballona:
Yeah, they work very closely with the customer understanding metrics. And the DS team, they own the outcome of the customer, they own the measurement, they own the iteration, they own the direction of optimization as well.

Josh Schachter [Host]:
So let's say what types of things are they Optimizing and tinkering with.

Diego Ballona:
Yeah, let's say Customer for example, comes in and they have a large set of topics that they get contacts in and they give us a sample of conversations. And through that sample conversations we can define, the DS team, can define what are the largest opportunities, where is the most time spent in customer service and where is more value really for the customer? Because it's very easy to achieve great containment and great deflection. It's not easy to achieve great value, sustainable value. And our data science team, they're on point to basically assess the quantitative side, as I said, like the metrics and so on, but also the qual side, the quality. Right. And how to measure quality. Like we built a bunch of very sophisticated tooling and some of this is making its way into the product as well, which is how to use LLMs as a judge for which conversations have high quality or low quality. And one of the things that we surfaced recently in the products is our CX score feature, which is basically an AI CSAT that scores 100% of the conversations with CSAT which originally you could only score a very small portion of contacts because it's survey based and so on and so forth.

Diego Ballona:
That all started with tooling and manual analysis from the FDE motion. And I think that's a great example of something that it started. White Glove. Right. It started with like a DS running manual analysis and it made its way into the product and we have many other features that follow this path as well. So just to summarize, three roles. PM engineer, data scientists, and they work together in teams. A team usually has 1pm, a couple of engineers, three to five, two data scientists and they work in a book of accounts.

Diego Ballona:
They're working at all times with a set of customers and they own the customer outcomes. They work very closely with the rest of the GCM team as well with like customer success with sales and so on and so forth.

Josh Schachter [Host]:
I feel like there's one group that you're leaving out of this orchestra which is the customer themselves. How are they and who from their team is working with you?

Diego Ballona:
Absolutely. And it's, it's a very, it's a very important part because sometimes customers don't have that person. Josh. And that's when. That's something that we learned over like the last almost two years that if they don't, you need to set expectations up front.

Josh Schachter [Host]:
By the way, sorry, sometimes they don't have that person or 98% of the time they don't have that person.

Diego Ballona:
It's yeah, it's been increasingly getting easier because I think people are adopting AI more and more. Right. Like I think two years ago it was like 98% to your point. I think these days, like people understand that without a great knowledge base, there's no way you'll get great outcomes with AI. Right. Like without someone that understands the business policies and like someone that's somehow working on support automation, you won't get to those outcomes. Right. So it was very rare to have someone on the customer side that would partner with us.

Diego Ballona:
And it's never one role. It's usually like a group of people as well. It's a team as well on the customer side working with us. But over time, it's getting more and more common, really. Customers have conversation designers, they have people working on content full time knowledge managers. And it's getting more and more common, which is really exciting. It is, it is a, it's an unavoidable direction, really. These roles will become more and more prevalent over.

Diego Ballona:
Over time.

Josh Schachter [Host]:
Yeah. Shifting gears, one of the advantages of your team being directly embedded with customers is you're hearing their voice of the customer firsthand. What they want, what they need, what's working, what's not working. How has the product roadmap for Fin gotten stronger as a result of that engagement model?

Diego Ballona:
Yeah, that's, that's a great, that's a great question, Josh. And actually materially changed how we build product. I think not just the FD motion, but our investments in Fin as our main product really over the years, we always took a lot of pride in ourselves at Intercom of being customer obsessed. It's one of our values, one of our company values. But I think RFD Emotion has taught us that there is so much more to be more of that. And the reason for that is when you get engineers sitting with customers directly, they feel the pain firsthand. They're not just hearing through aggregated research, they're not just looking at the metrics and so on and so forth, which are very valuable, by the way. They're very valuable tools.

Diego Ballona:
But there's a different speed of iteration and ownership that comes from having your R and D team be part of the customer team. They understand the problem much more quickly. They can connect to all the complexity of our product, of your product, the problems that a customer has much more quickly. And they can empathize much better. They understand how bad it is. And look, we built so much better product as consequence of investing in the FD Emotion. Much better tooling internally for our GCN team. And I think it has helped us.

Diego Ballona:
I think this is one of the greatest values that RFD Motion has brought to Intercom. It's not just working with a subset of customers where it makes sense to have this motion, but actually has helped us build a better product for all of our customers and to make possible to build much better customer experiences for their customers as well. So I think that's one of the really, really important things that RFD Motion has taught us.

Josh Schachter [Host]:
Well, I'll bring up an example that you told us when we were prepping here was Slack. Slack channel support, right? So tell, tell the story about, you know that.

Diego Ballona:
Yeah, yeah, this is an. A very early story actually, which is quite interesting. One of our customers, they had a very large share of their support volume happening on Slack. We did not have fin for Slack at the time. And like, reiterating what I said, engineers sat down with the head of support of that company and they were like, hey, it would be so great if we had something like this because see all the questions and how the experience looks like. And engineer came back to us and he was like, hey, I think I can put this together in two weeks, three weeks or so if I really focus. And I think we should do it. This will never make it to the product roadmap because we have wider priorities.

Diego Ballona:
Right. We're serving a very large customer base, but I think I can put together a version very, very quickly for this customer and made a decision to go and build it and make this customer much more successful out of consequence of that. And it turns out that we built a product, it was great, customer was very happy. But on top of that, we actually decided to invest in that because we validated demand. Once we have a V0 one of the products out there working for one customer, getting the next customer is easier to use it then getting the next customer is easier. At some point you get the product to general availability, right? Like, you iterate over it and you end up with a product that's very polished. And we have an interesting division of labor also, which is the early prototypes when we're working with specific customers. That's the FDE motion.

Diego Ballona:
But at some point it makes sense for a core product team to own the product because it becomes a big part of our infrastructure, of our platform and it comes with more investment as well. So other teams take on the product and actually productize it for all customers, really. And this is what happened. In a time span of three months, we went from V01 to actually a GA product for Finover Slack and we have many customers now using it, which is a great success story and we would never have prioritized this if it wasn't for this special set of circumstances. So I think one key takeaway for the audience also Josh, would be this is one great way to build products as well. You go and solve one customer problem and you go and solve the next customer problem. And this is how you build products when you're in a startup. Right.

Josh Schachter [Host]:
It's classic design partnership too. Right. I mean that's, that's kind of the name of the game. Yeah, yeah. So your, your orchestration again with engineering and go to market. Very cross functional, collaborative internally. What are the clear ownership boundaries between you as the senior director of engineering and let's say Junan who runs customer success and solutions?

Diego Ballona:
Yeah, we think of ourselves as one team, Josh. We don't try to ship the org chart. A lot of my team in background are R and D folks, and I'm both. I have a feat in GTM and a feat in R and D and so on and so forth. And I think overall the feeling of working together between the FDES and CSMs and professional services and all the other functions, like it's like we're one team. We're here for the customer now division of labor. The way we split the work and accountabilities responsibilities is CSMs own the success of the customer in the account itself. So they're accountable for the plans and making sure the customer successful in the long run.

Diego Ballona:
Right. Like in our FD's, they're more, the engagements are more ephemeral. Like we're usually engaged with the customer for a duration of time. Back to my point earlier, we're there to accelerate the time to value a CSM and a solutions architect. They would be able to get a customer to success pretty much in the same ways. Right. Over time. But it would take so long, it would take multiple months of effort.

Diego Ballona:
And the FDA team can shortcut again product roadmap priorities. They can shortcut technical gaps and they can accelerate the customer's value really. So the account is owned by the success team, the account is owned by the GTM team and FDEs. They come in the Ford deploy team. It comes in with a mission. We have a bound engagement that we're there for and this makes like their roles and responsibilities very clear between, between our teams basically.

Josh Schachter [Host]:
So time to value is one of the key metrics for you internally as a company. Success metrics. And there's reason for that too. I mean we all want to create value for our customers. In the case of Fin, you're also charging on positive outcomes, right? Outcome based. And so, so the quicker you get them using the platform, being successful with it, you know, cha ching, cha ching, cha ching, right. Those dollars start coming in faster. So that's the nice thing about this as well.

Josh Schachter [Host]:
And I, I suppose that's probably where another reason we're going in this direction in this new agentic world as, as pricing becomes more outcome based.

Diego Ballona:
Yes, exactly. We, we, we realized very early, we realized very early the outcomes based pricing was like the, the way to go for fin, basically because it aligns incentives. Right Josh? Exactly what you said. Our success is a customer success. So like everything that we do internally from comp plans to performance reviews to all the incentives that we built around all of our go to market teams and the FTE team as well, they're around resolutions and customer outcomes. Even when you think of ARR for fin, it's related to how much work we're getting done for the customers and it beautifully aligns our interests. We're never interested in selling the customer interactions or more usage like we're, we're trying to take work off, off of their plate and we've seen a lot of like success with the model because it makes, it makes it, it makes the engagement of the customer feel like we're, we're there together to solve the same problem together with incentives aligned with a clear goal that we, we can share.

Josh Schachter [Host]:
Basically I want to end the episode with a key takeaway that you want to leave for other engineering customer success. Go to market SaaS, leaders that are listening. It can be about the virtues of what you've built, 67% resolution rate. It can be a lesson that you learned along the way, maybe from something that didn't work. Whatever you want. Leave us with your golden nugget for our listeners.

Diego Ballona:
Yep, yep. I would say, I would say main learning that I had personally and it's less an advice for everyone and more a personal learning and maybe it's useful for the audience as well is to be open minded. You know, we had to iterate so many times to find a model that works and that we felt like it felt natural, that it made it made us realize that there's no like right or wrong answers in this model. Of course there are things that work and don't work and it's very clear when they don't. But being open minded, including to fde, dfde motion, not being the right thing for your business, but you should be open to that. You should validate that, right? Like, there's a lot of, like, there's a lot of noise around, a lot of buzz around the name and the motion. And every AI company is doing a version of this. And reality is maybe you don't need it.

Diego Ballona:
Maybe you don't need it. We learned that it's useful. We learned that we build better products out of it. We learned that we get insight from a customer that we otherwise would not have had. But it might not be true of you, you know, and staying open minded, I think has been one of the greatest, like, joys of working with this team. And I think if you, if you decide to try this out, you should too. I think that's the main advice that I have, basically.

Josh Schachter [Host]:
I love it. Okay, well, I'm hoping that in 12 months we can have you back on the show and you can talk about all the growth that you've incurred in the past in 2026 into 2027. If it's, if, if, if what you guys did in the last year is any indication of the upcoming year, then I am incredibly excited for you. Diego Bologna, senior director of engineering, owner of the FDE Motion at Fin at Intercom. Thank you so much for joining. Unchurned.

Diego Ballona:
Thank you. Thank you, Joshua. It was a pleasure. And thanks everyone.