Building a Scalable Customer Success Program

Abbas Haider Ali


VP of Customer Success


Abbas Haider Ali
Abbas Haider Ali

Episode Summary

Today on the show, we have Abbas Haider Ali, VP of Customer Success at GitHub.

Before joining GitHub, Abbas held the position of Vice President of Customer Success at Twilio and Segment. Additionally, he is as an investor and advisor for various companies including Slack, Hasura, Adept, and numerous others. He also serves as a general partner for the GTM Operators Network.

In this episode, Abbas shares his insights on how to build a scalable and data-driven customer success program. Abbas emphasizes the importance of understanding the actual impact of customer success work, focusing on leading indicators rather than lagging ones.

We then discussed the differences in how smaller and larger organizations approach customer success and the need for a robust model of customer health and a data-driven approach to attribution work.

As usual, we're excited to hear what you think of this episode, and if you have any feedback, we would love to hear from you.

Mentioned Resources




High Touch CSM Models00:01:56
Operational Expenditure Pressures00:04:42
Data-Driven Customer Success00:07:52
Leading Indicators for Customer Success00:09:02
Scaling Customer Success00:13:30
Personalized Approach to Customer Success00:17:35
Value engineering in customer success00:21:40
Value hypothesis to reduce churn00:31:26
Importance of data-driven analysis00:33:50


00:00:00 Abbas: A lot of, in particular, customer success management programs tend to start with a very high touch model, where our customers need a lot of attention. We want to make sure they're successful. We want to hold their hands to the entire lifecycle, be present with them all the time. So you tend to wind up with these high touch CSM models as a starting point very commonly. And then, those customers tend to be happy. They're growing. They have attention from CSM. They have attention from account teams, from product. They're in the cabs. So, of course, they tend to grow very robustly because they have so much of a company's energy and product energy behind them.

00:00:36 VO: How do you build a habit-forming products?  And you saw this different– Don't just [unclear] for revenue in the door.

00:00:42 Andrew: This is CHURN.FM, the podcast for subscription economy pros. Each week, we hear how the world's fastest growing companies are tackling churn and using retention to fuel their growth.

00:00:55 VO: How do you build a habit forming products? We crossed over that magic threshold to negative churn. You need to invest in customer success. It always comes down to retention and engagement. Completely bootstrap. Profitable and growing. 

00:01:08 Andrew: Strategies, tactics, and ideas both together to help your business thrive in the subscription economy. I'm your host, Andrew Michael, and here's today's episode. Hey, Abbas, welcome to the show.

00:01:22 Abbas: Thanks for having me on.

00:01:23 Andrew: It's a pleasure to have you. For the listeners, Abbas is the VP of customer success at GitHub, where over 100 million developers shape the future of software together. Prior to GitHub, he was the VP of customer success at Twilio and Segment, and is also an investor and advisor in companies like Slack, Hasura, Adept, and many others, as well as a general partner of the GTM Operators Network. So my first question for you is, looking at your background, you've held roles in engineering, marketing, products and customer success. How have you managed your transition through these various roles?

00:01:56 Abbas: Yeah. So I think I've taken more of a renaissance approach to building my skill set as a leader. I think of myself as a people leader and business leader first, who happens to specialize in different domains over time. So with that framing, it makes it a little bit easier to be like, okay, how do you make a transition for product to engineering leadership as an example, I have a technical background, so I understand the technology part of it and I can make that switch. I've had to go to market roles, and so go to market roles. You can switch that into a product relationship from an outbound perspective. And then, different flavors of dealing with customers, whether it's acquisition on the sales and SE side, or on the customer success side in terms of retention and growth. So I found that framing to be helpful to make those transitions. But I will say, the key in every one of those transitions is it's easier to make it where you have credibility inside an organization than to do it brand new on the outside if you're doing it for the first time.

00:02:49 Andrew: I found it very interesting just looking at the background, because at the service level, you'll think like, wow, these are quite big jumps and moving. And it wasn't like there was linear, like there was progression, I think, throughout your career as well, which is really interesting. Normally, you would expect maybe to take a step back moving into a different industry. But the way you framed it, basically, as a people leader and then from there, being able to move into different areas, I think is really, really impressive.

00:03:12 Abbas: And the other secret is the network. The network is very helpful in terms of you build credibility through that, and moving from one to the other where you don't have to prove that… you don't prove yourself from scratch. You can say, hey, you already know me. And being a leader in this area, we're just doing a little bit of a context switch in terms of solving the current needs.

00:03:29 Andrew: Exactly. And today, obviously, you're a leader in CS at GitHub, but you also have a really good vantage point, I think, being an investor and advisor in so many different startups in the industry. And today, I thought that could be interesting, a little bit, to chat about some of the CS trends that you're seeing in the market today. And maybe we could just get started there. What are some of the interesting trends you're noticing in the market in this current environment? Obviously, things are shifting and moving so fast today.

00:03:55 Abbas: So the current trends you see on the outside, on the investing side, I see it all the time. Companies are facing a lot of operational expenditure pressures. So if you're a seller of B2B products, all your customers, in the same book. They're sitting in their boardrooms and their C-suite saying, how do we cut back on our expenses so we can buckle up for the current rough ride that we're in now, and be sustainable in the long term? So there's a ton of operational pressure in terms of selling. And so what that means is, that plays through the entire industry. You'll see sales cycles lengthening, especially across new business. And what that does is it puts additional pressure on existing customers both to retain them, but actually, to drive growth through them as well. So the growth at all costs for new customer acquisition faces a ton of pressure. So in order for companies to continue to grow, they have to do a better job of growing within their own organizations with existing customers, and doing it with an eye on profitability and efficiency in driving, increased gross margin versus the historical trends of, money is free. You can spend whatever you need to grow. That's just not the case anymore.

00:05:00 Andrew: It's not absolutely, I think, recently as well... So it was a post by Jason Lemkin from SaaStr. I think, if I'm correct, we had something around the median time to close deals now, and how it's almost increased like 30% over the last year for sales teams. So much longer sales cycles need to go in and really needing not to, as you say, buckle down and trying to focus on profitability, trying to maximize customers that you do already have and have that shift. It's interesting as well that you say, from a startup perspective and from a large organization, having that vantage point is something special. What other things are you noticing amongst these teams, and maybe what would you say, difference between earlier stage startups versus a larger organization that you work with?

00:05:43 Abbas: So I would say, there's a couple of things that are common. There is a lot more demand on every dollar we spend, especially on sales and marketing. And in this context, our group, Customer Success into, not just COGS, but also part of the sales and marketing spend, because you're talking about acquisition revenue growth. In that context, there’s a lot more, ask on how we're going to spend every dollar, what's the efficiency, and what's the actual measurable return that it produces. And that's true across all scales. But how it plays out is a little bit different based on the size of the organization. So for smaller companies, I'd put that at Series C or smaller. You'll see a lot of non-differentiated CS teams, a lot of roles that are blending between services and CSMs, and support and premium things. And that's okay. You kind of operate in that aggregate mode as you get into the larger organizations. The efficiency and attribution work you do is much more specific to the function. So you want to be able to articulate, what is the impact of professional services? What is the impact of CSMs? What is the impact of our premium offerings if you have any paid customer success offerings? What's the impact of any strategic advisory work we do,or any freebies we give to customers to help them be successful? That impact analysis becomes much more discrete at the larger scales, and you wind up having a lot more data, of course, to do that work too.

00:07:02 Andrew: It's interesting. And definitely as you get larger as well, you have a lot more of these things happening as well throughout the organization. So keeping tabs and managing and tracking, understanding really what the costs, they go into all of these services. It must not be an easy [for you], I think, as well to get done.

00:07:18 Abbas: And I think one of the hardest things about this is, Customer Success in general, I found across all scales, has tended not to be as data-driven as other parts of organizations. So you won't see as much rigor in most CS organizations typically than you would, for example, for revenue, where you have a super detailed focus on conversion rates and pipelines. Or even on marketing, where it all, you wanna be very, very data driven. That tends not to be as true for CS, but in the current climate, that's definitely something that's changing very rapidly. There's a lot of demand on building more data-driven and measurable and attributable results from CS, and that is a challenge for CS leaders to step up into that work.

00:08:02 Andrew: I think, definitely, the shift is... I think maybe also I live in a bubble with this podcast as well and speaking to a lot of CS folks. And yet you tend to hear a lot of talk about metrics, and understanding and how it's shifted over the years. But then when you step outside of that, you realize, okay, Customer Success is still not really clearly defined in most organizations. Most people don't have good direction when it comes to it. And I think, as you say, definitely now the way people are feeling more and more of the impact, and more really need to show the ROI of the work that we're doing. Customer Success has become a lot more robust in their understanding of the metrics that they're tracking and how they influence. And I think a lot of leaders I speak to on the show, they really try to argue the fact that, everything you need to bring it back to a dollar value to say, how can we prove that we're actually bringing value to the company? And some of this is subjective, but otherwise there are clearly good ways in which you can do this. What are some of the ways you see good leaders actually showing this value throughout the organization that CS is delivering?

00:09:02 Abbas: Look, I think they're very common measures. You'll see CS organizations report up, are going to be things like, retention is going to be NDR in terms of talking about the expansion of the revenue base. And that's when things start to get a little squishy beyond that. You'll see things like efficiency of ARR per CSM and number of accounts. But those things are all very lagging indicators of what's taking place. The types of questions that I look to understand for all the teams that I support is, what is the actual attributable impact of the work that's being done? Not in the lagging sense, but what are the leading indicators? We're going to use the judge at something as effective. I'll give you a couple of examples. We say, QBRs are super important to do with our very large customers. What is the difference between a customer outcome when they do a QBR every quarter versus every six months versus once a year? What's the impact that that has? If CS includes professional services, you have ten different services packages you offer. What is the impact that those have in terms of customer outcomes? 

00:10:04 Abbas: So generally speaking, what I look to do is understand what are the leading indicators we can have in the CS business. One that I talk about very frequently is having a pretty robust model of customer health. And then, you have a very data-driven approach to do the attribution work. Say, customer health has this attribution with retention and our expandability. And then, the work that we do in each CS function is directly attributed to that as well. So you wind up having to build a multi-touch attribution model essentially for the CS activities to the underlying customer outcomes. It's incredibly powerful when you do that because it tells you what are even the points of diminishing returns. So if you say, it takes all these things in our bag of tricks as, CS organization, do you have to hit a customer with all of them for them to be successful or is half enough? Because if half's enough, we kind of set the bar there saying, hey, we don't want to actually spend so much time with our customers that we actually are leaving some customers behind, because we don't have the coverage for them. So that type of data attribution work is very new in the CS world to go back and actually have that multi-touch orientation, not towards lagging revenue indicators, but leading ones that tell you you're heading in the right direction for the future.

00:11:15 Andrew: I love that. It's sort of adopting the marketing framework for multi-touch attribution, and bringing it into customer success, really trying to understand what are the different activities that we're doing and how those are influencing the overall retention and expansion opportunities. It's really like a true advancement in terms of where we've come over the last few years with Customer Success. And it really starts to productize a little bit more. And it also makes it more familiar for others in the organization to understand because it's something that marketing's been doing for decades now.

00:11:46 Abbas: That's right. And it also exposes any gaps you may have in the system. They may be... We operate with a lot of assumptions. I feel in CS, we're like, this is the thing we should do because it's what I did at my last company or what I saw work. And then, someone came on the podcast and said, this is an amazing recipe for success. But it's different for every business. It depends on the buyer, the value proposition for your product, the adoption pathway, the growth pathway. And so this provides a very robust, data-driven mechanism to answer the question of, why are we investing in CS? What benefit does it produce for the business and our customers? It's a very objective way of doing that analysis.

00:12:23 Andrew: And to the point about, podcast, if anybody just hears advice and tries to follow it blindly in the organization, I think that's a very, very bad move. In general, I like to prescribe [inaudible] practices, because I think it's very easy to assume that something that's worked somewhere else is going to work within your business, within your environment. But really, every startup is unique in its own little ways. And you might strike it lucky once when you hear something, but then other times you get burned for not really trying to understand. Having something, like you say, this [multi-touch vision] model really allows you to say, okay, let's experiment with this, and then see, does this actually move the needle in a meaningful way. That's awesome. And I think you mentioned something as well that was really interesting. The notion of like, how many touches do these customers really need to have? And are we overdoing things versus understating things [and understand] how we invest our time and money in customer success. And I think this comes up a lot as well with the concept of scaled and digital customer success trends and strategies. What are some of the things you're seeing that are interesting in the market now, that people really need to be cautious of how they're spending and how they're investing in order to be able to manage these scaled and digital CS efforts?

00:13:30 Abbas: I think, historically, my sense of what I've observed, because a lot of portfolio companies and companies that have been at directly is, a lot of, in particular, Customer Success management programs tend to start with a very high touch model, where our customers need a lot of attention. We want to make sure they're successful. We want to hold their hands to the entire lifecycle, be present with them all the time. So you tend to wind up with these high touch CSM models as a starting point very commonly. And then, those customers tend to be happy. They're growing. They have attention from CSMs. They have attention from account teams, from products. They're in the cabs. So, of course, they tend to grow very robustly because they have so much of a company's energy and product energy behind them. And then of course, the question is, as you scale, we have a lot of customers, how are we going to scale this model to everybody? And it just doesn't work. It's too much. It's going to be way too inefficient to actually go out and do that. 

00:14:24 Abbas: So a typical question there is, well, let's scale this out somehow. And so you tend to go from high touch to slightly less high touch, where you have ratios that are 15 to 1 or 10 to 1 or even fewer accounts per CSM, and you sort of crank up the dial slowly. You get to 20 to 1, 25, 30 to 1, and you're like, well, that's it. That's all we can do. There's no more time to be had. And I think that's why this data-driven approach is really important, because if you have the data, you get a sense of what are the meaningful engagements that you have with a customer that you can then attribute to the outcomes you want. And you can actually start to really filter down the team, say, okay, let's build a scale team that only does the most valuable engagement with our customers. But the problem with that is, they don't have relationships with all those customers. If you get to even, let's say, 100 to 1 ratio, what's the realistic nature of maintaining a hundred warm customer relationships as a CSM? That would be almost a nightmare job, trying to keep all that and being able to operate it is extremely difficult. Which is why I think digital CSM programs are actually the foundational piece you need, because they should be taking the same high value engagements and driving them digitally. So you find a building, a very powerful tech touch engine that fully automates as much as you can, but also drives demand for the scale teams.

00:15:48 Abbas: So they're not out there seeking engagement. Engagement, being driven by the tech touch team, and that forms an incredibly powerful look in my opinion. It's like, you have the high touch thing where you get to try things out, see that they work, they're effective. You have all the data to support that. You push that to the tech touch team to drive that engagement, and support that engagement through a low touch scale CSM program. And now you have the best of all things. And having done this, the impact is very measurable. You see the impact on retention rates. You see the impact of net retention rates. And it's not half a point or one point, it's multiple points, including... I've seen impact to double digits on both those measures as a way to really put your arms around the customers. And it can be a very, very powerful motion window.

00:16:33 Andrew: I think it's very interesting also combining it in, with the knowledge that you have in the multi-touch model, because I think... In the show I hear a lot as well when we talk about how have you structured your CS teams. And typically, you hear, we have high touch for our enterprise clients. We have a sort of mid-touch for our mid-market clients, and then we have this low touch or tech touch for our SMB clients to try and service them all. But I think that's taking a very one-size-fits-all approach model where you might have your enterprise clients that are just going to come in and get on with things. They’re always going to stick around, they like perfect fit, and then spending that extra time. Maybe not the most efficient use of the time within the organization. I think layering on the data side of things, like having that really good knowledge and understanding, knowing what's working, knowing at which levels are you going too far or not, and then having the ability to scale these different breaks. It feels like we're getting to a point now where we can start to be providing a more personalized approach to Customer Success without boxing customers into generalized segments.

00:17:35 Abbas: So actually made a great point that, if you want to get back to... Doesn't slip away from my prior commentary. It's very tempting to think about these as being distinct layers. You have high touch, you've got low touch, you've got digital and never shall they cross pathways. That's actually... I don't think about that that way at all. In fact, I think it's really powerful to have the tech touch programs working in the high touch CSM portfolios. If, imagine your CSM [who have] like 10, 20 accounts, but you're supporting customers where you have hundreds of users, thousands of users potentially, there is no way that as a CSM, you're going to be able to cover the engagements with all those. So what is your mechanism to do that? It’s the tech touch programs. So tech touch programs that work should actually be driving activity in the higher… in the high touch portfolios as well, that you drive that engagement. So you have, not just human touch and tech touch, you kind of have cyborg touch. You have both human and tech working together to drive that scale across the entire [unclear].

00:18:34 Andrew: Very nice. Definitely having them work together is a much more powerful mechanism as well for the organization. The other thing, then, so we're talking a little bit about the technologies and the different structures. Let's talk a little bit about the actual Customer Success managers. What are some of the skills that you're seeing are absolutely vital and a must have in Customer Success organizations?

00:18:57 Abbas: So look, I think when you talk about CS organizations in general... I do think of CS organizations, it's pretty broad. So there is CSM. You may have Customer Success architects or even more deeply technical. You have services, support, premium offerings, the whole set, partners, ecosystem, all that can fall in the CS umbrella. Specifically for CSM, so I think about it as the skills vary based on the type of CSM your product needs. So one thing I do see as a typical failing point or friction point within building out and scaling a CSM organization in particular, tends to be around not understanding the type of CSM motion you need. So my mental model, I have four different flavors of CSMs that can exist. You have a CSM who feels [A-like], so they may actually paper deals, like renewals. So that's an example of an [A-like] CSM model. You have a solution architect, like the CSM model, and they tend to have some elements of professional services into their skill set. So they may help manage your projects. So they're managing a very detailed, in-depth onboarding lifecycle. You may have support like CSMs, where they provide almost like a premium level of support. So they pull together root cause analysis or multi-ticket type activities into a single workstream. And then you have SE or Solutions Engineering like CSMs. They tend to spike higher on things like discovering orientation for use cases.

00:20:22 Abbas: And blends of these, but I do see CSM organizations that aren't clear on what flavor they need, what flavor they are for their business. That causes its own scaling and consistency challenges. But if I were to pick one skill that to me is a superpower for CSM teams, that really would be a general skill set around value engineering. And the reason I say that is very much anchored around the discovery pieces. But one of the most powerful things CSM can be really, really good at is being very buttoned up, articulating the value of a product that the customer consumes. Not in the sense of, they use these features. Here's how many users are adopted, but articulating value in the language of the customer's business. So articulating how they help generate revenue, save money, or mitigate risk at the business level. And that is not a very common skill in the CSM portfolio. But for the CSMs that have it, when you think about, like why are certain CSMs just full on, superstars with their customers, the odds are, that skill set floats amongst them.

00:21:29 Andrew: Can you talk about that a little bit more as well? What does the concept of value engineering mean? And a few examples, like how you've seen good CSMs conduct good value engineering with their customers.

00:21:40 Abbas: So this has its own podcast, by the way, talking about my perspective on value engineering. But fundamentally, in the CSM context, what I think about it is, when someone buys a product from you as a company, they're looking to solve certain business problems. And generally in the pre-sales process, somewhere along the way, they have told you that. They didn't say, hey, I wanted to use it for these features, or I like these reports you produced. They have said, hey, if I use your product, I'm assuming I'm going to unlock growth and generate revenue of this magnitude, or it's going to make me more efficient and allow me to scale better or allow me to cut back costs. And here's how I'm going to measure that. Or, there's some risk it's going to help me mitigate. So essentially, those are the only three reasons fundamentally that anyone actually buys any technology. And on the pre-sales side, if the team has not captured why that customer is looking to capture that, you really don't understand intent. So that's the first phase. That on the selling side, you have to be really buttoned up about capturing the intent behind why a customer bought your technology. And then the interlock course, CSMs in particular, is to track that intent along the realization journey, is to know that these are the reasons that my customer bought it for, not for unlocking features, using parts of my product, but actually getting to those business outcomes.

00:23:00 Abbas: And the better a CSM is articulating the realization journey from that intent that was stated early on, the stronger that account is going to be in terms of retention. And also if you apply those same skills to identify what are the next pieces of intent that the customer has, you're essentially doing value discovery for the next phase of growth. And then you're able to pass that back to the selling organization. Say, we had intent, we have achieved realization of this amount. And by the way, here's the next wave of intent that I've detected, and that is the fundamental interconnect. You'll see representatives, when we talk about leads back to sales or CSQLs or expansion opportunity identification, however you label it, it's all the same thing. You're identifying business outcome intent that is for the future, that you, then, pass off to the selling organization to go out and capture as the next wave of growth.

00:23:54 Andrew: I love that. And I think one of the things that struck me when you were talking now as well is the second, latter part of, it's like, following up in a year and then saying, what is next? Now, what are you trying to achieve? Because this is commonly another reason for churn, either through graduation churn or for the needs not being met a year later into the product. Because when people come to you on day one, they have certain needs and certain goals that they're trying to achieve. But a year passes, they've become a lot more sophisticated. They've started to solve those challenges, and now new challenges are presenting themselves. And when it's a great opportunity to explore and find expansion revenue by actually doing the work of asking and trying to understand, and not just relying on usage or features as a marker, indicator of, are they going to renew with this or not?" And another thing as well then, like you mentioned, was the relationship then between the sales and Customer Success. Because I think when you first started talking about that side of things, in my mind I was like, surely this is something that every salesperson asks the customer when they're trying to close the deal because you want to try and frame it in a way that you're going to close the deal. How have you seen these CS and sales teams work effectively to transition this knowledge, so that they come in and you don't get another call where they ask, what are you trying to achieve? And I was like, didn't I just talk to someone in sales about this?

00:25:07 Abbas: Yes. My least favorite conversation with a customer, when  after having worked with them for many, many months potentially, someone says, so what are you looking to accomplish with our product? I can only imagine the frustration, the customer side, when they're like, I cannot believe that I'm talking about the same thing with these people, with these guys again. So I think, in terms of the interlock, every organization I've seen has had this story. This is like a continuous thing in every company. No one is ever fully satisfied with it. And that's like the pre- to post-sales handoff. Whatever you want to call it, that basically is always a friction point. I would say the organizations that do it best, in my experience, have really good sales methodology to start, because what it means is that they're consistent about how they want to sell to a customer and are really good at capturing knowledge about customer intent. And ideally, it's in the form of... Not narrative. The narrative is important, but also some element of structured data.

00:26:06 Abbas: So my personal view on what is important for structured data, is a hierarchy of value use cases and outcomes. So if you think about for any given product, if I gave you a blank sheet of paper, could you write down the set of use cases and problems that product solves, and how a customer would measure the outcomes of that in terms of engineering hours saved, or reduction in… spend on marketing to acquire a customer, or reduction in sales cycle. Whatever the ultimate outcome of your product is for a customer, can you capture it in that language and build a general framework? And if you can, and by the way, I do believe that pretty much any product can actually do that. It's hard work to do, but important to actually do that. That's a whole other conversation as well. But once you have that in structured form, ideally the sellers, and that would be the combination of AEs and SEs for most products, but articulate what are the use cases the customers try to solve, what are the measures that matter to them. And that, in structured form, becomes a handoff to all the post-sales teams, so they know where a customer is trying to go in that same context and can provide the next set of structured data back as well. So that tends to be the most effective handoff, but it does require strong selling methodology and frameworks that also the post sales teams are trained on. So the CSMs, in particular, should be very much steeped in the same language. So when someone says, what is the account plan? There is no CSM account plan and sales account plan. It is the same artifact because that becomes a connective [tissue].

00:27:40 Andrew: It's very interesting as well, the way you mentioned structured data as well coming from these conversations are just... Obviously, my mind, thinking of large language models, thinking advancements that we've seen now from OpenAI and others, that some of this work almost, now, becomes fully automated to a certain degree where you can, then have these really good handoffs and just have conversations at the end of the day. Are you seeing any startups leveraging these technologies in interesting ways in Customer Success?

00:28:09 Abbas: So that is a very interesting conversation on LMs, because if you think about, like what is every selling organizational... Any selling organization, pretty much terrible at consistently, it's providing structured and captured data for this organization. I've yet to encounter a sales team. It's like, our sales force is 100%, but we got all the things. There's just no such thing. So I do think there's a massive opportunity here to leverage these large scale LMs, actually analyze things like sales calls or conversations and emails, and extract from them the structured data dimension. And then, that becomes the same thing that you utilize on the CS side as well for those conversations. So I do think that is an interesting development area, but I don't believe I've seen any solutions that really solve it yet. The closest is going to be things like the gongs of the world that capture that content, but it's not being put into a value framework. You get transcripts and highlights, but it's not fully baked. In fact, I was just talking literally this morning to a friend who's like, hey, I have this cool startup idea. He's a go to market leader, who I've worked with for a little while, and he's like, I'm thinking about doing, like, a ChatGPT thing for selling teams. And so we were literally talking about this topic for probably about two hours.

00:29:23 Andrew: It is interesting, because I've also explored it a little bit actually in depth with [inaudible] Avrio. And also then I was thinking, okay, I do podcasts every week, how can I get structured data out of these conversations so that I can automate some of the content that I share? And it's pretty amazing how good of a job it does. So what I've just tried to do is, like, give me the top five highlights in JSON format with the speaker name, timestamp, and the comment that was left, something. And pretty much every time, it gives me some good results that I can use, and it saves literally hours. And I could see the same thing just with the right prompting being applied towards these conversations that are happening in sales and success. Just knowing how you can prompt engineer correctly to be able to get the output that you're looking for.

00:30:07 Abbas: Look, I think it's definitely a lot easier to edit the content than it is to start from scratch. Looking at a blank… Fill out all the details about the customer story, you're like, oh, man, this can take me forever. If I start out with something that's like, even 60% complete and good, that's a huge step forward in my view. The only concern I have with some of these things is, of course, the problem you see in all these LMs, is the hallucination problem. Which is like, the answer that's not in there at all, but is delivered with great confidence. Now, since this is reviewed by people, hopefully you're able to kind of filter those things out. But that is certainly one consideration, someone would be like, yeah, this customer bought it for X use case, and we tell the customer that they're like, I didn't say anything like that. And it turned out that the model was just super confident about it.

00:30:51 Andrew: Confident, yeah. But I think all of these, in due time, are going to be solved. I see you talking about time. We're running up on our... and so I wanna make sure I have time to ask you a couple of questions that I ask every guest. First question is a hypothetical scenario. You join your company, [Channel Attention] is not doing it great at this company. Sierra comes to you and says you're in charge. You have three months to turn things around. The trick is, you're not going to tell me I'm gonna go speak to customers, and figure out their pain points. So, look at the data and start there. You're just going to take a tactic, a single tactic that you can implement fast, hoping that it works with this company and run with it blindly. What would it be?

00:31:26 Abbas: Yeah, my default tactic for all this is, the biggest reason customers churn is because they don't appreciate the value the product provides. So if there's one tactic I would run consistently, is sharing our value hypothesis, because we don't have it in this case based on our churn problem, with our economic buyers. I would run that play all day long and that would tell me immediately where actually our problem is at.

00:31:48 Andrew: Nice, short and to the point. What's one thing that you know today about [Channel Attention] that you wish you knew when you got started with your career?

00:31:55 Abbas: Oh, that's a great question. So if I think back to early on, one of the things that was definitely something that I wish I had started with sooner was not being as assumptive about why I think a customer is happy with their product and should keep using it, but taking very much a customer lens. And I think this is still a problem coming from a product background, you know, technical. You tend to be very introspective, like, I think these are the magical metrics that if a customer does it, they should love our product. And that was not always connected into what the customer's measurement of why we should keep using our product is. And I wish I'd taken that lens more earlier on in my career as well.

00:32:35 Andrew: Very nice. Is there any final thought you want to leave the listeners with today before we wrap up? Maybe how can they keep up to speed with the work that you're working on?

00:32:44 Abbas: Yes. So look, in general, I do like to share the wins, and things that we're doing in every role that I've been at. So I'll continue to publish those and share those, whether it's on LinkedIn or on our company blogs, and things like that as well in terms of showcasing the amazing wins we've had as an organization and publishing them as actual recipes they can utilize in context of our product. So there's several ways to keep up with things. If I just leave one parting thought here is, in the current climate, there is a lot of pressure on all parts of the business, rightfully so, to be more efficient, to generate, and contribute more to margin for every business so they can maintain and grow their valuations if you're private, or continue to be profitable and gain credibility in the public markets as well. And so the one thing, if that's provided as guidance, would be super focused on doing the attribution and impact analysis work of all the CS teams. There is so much stuff that those teams do. And I think the more data driven you can be in articulating what that actually does for customers in the business, that should be on the top three of every CS leader's list.

00:33:50 Andrew: I love that. And I think you've given quite a few different ideas today for the listeners to take away and see what they could apply in their businesses. So thank you so much for joining us today. It's been an absolute pleasure hosting you and wish you best of luck now going forward into the new year.

00:34:04 Abbas: Thank you, Andrew, for the invitation.

00:34:05 Andrew: Cheers.

00:34:07 Andrew: And that's a wrap for the show today with me, Andrew Michael. I really hope you enjoyed it and you're able to pull out something valuable for your business. To keep up to date with CHURN.FM and be notified about new episodes, blog posts, and more, subscribe to our mailing list by visiting Also, don't forget to subscribe to our show on iTunes, Google Play, or wherever you listen to your podcasts. If you have any feedback, good or bad, I would love to hear from you, and you can provide your blunt, direct feedback by sending it to Lastly, but most importantly, if you enjoyed this episode, please share it and leave a review as it really helps get the word out and grow the community. Thanks again for listening. See you again next week.


Abbas Haider Ali
Abbas Haider Ali

The show

My name is Andrew Michael and I started CHURN.FM, as I was tired of hearing stories about some magical silver bullet that solved churn for company X.

In this podcast, you will hear from founders and subscription economy pros working in product, marketing, customer success, support, and operations roles across different stages of company growth, who are taking a systematic approach to increase retention and engagement within their organizations.


Listen To Next