Beyond the Handoff: How AI Connects Sales, Implementation, and CS

[Speaker 1]
Hi, good afternoon everyone. My name is Harsha. I'm the Chief Customer Officer at Beacon.

Today I'm going to moderate this session. We have Avinash on your left and then we have Joan in the middle and Suhoor next to me. I'll let each of them introduce themselves and then we can initiate the session.

Can you guys hear me? Good afternoon, this is Avinash. I'm working with BuildTrust as site lead and vice president of customer experience.

My responsibilities are to make sure that our implementations are running right on time, customer success and support takes the right hand off, so that the experience for customers are kept consistent, scalable and we keep winning with them. So that's my job duty. In a nutshell, I have been in this industry for about 22 years, everything in post sales operations.

So I'm here to sort of try and share my experience with you guys and hopefully make it engaging and informative for every one of you here. Thank you Gainsight for this wonderful occasion and Beacon for inviting for this talk track. Thank you.

[Speaker 2]
Hello, my name is Joan Bellew. I'm principal solution director for Zealous. I've been in this industry more than 25 years, from all the way through everything that's ever happened through implementation to where we are now getting customers live.

And my role is really to look at solutions being offered to the customers, how we're using the product, is it repeatable, is it scalable and what can we do better to make it a better journey overall.

[Speaker 1]
Thanks, Joan. My name is Subo. I lead customer experience at Keka HR, everything post sales and Keka is a full suite HRMS, we service about 8000 customers globally and my team works with customers to ensure that they're happy every day.

Thank you. Thank you guys for the introductions. Today our topic is going to be beyond the handoff, how AI can connect sales implementation and CS.

As you can see in the slide, everybody have something that they're leaving and all of those pieces are, I think, yeah, you cannot see it very clearly, but essentially everybody have something that they are not handing off to the CS and accumulative effect of that is getting onto the CS. So we are going to discuss how exactly that handoff can be seamless in the AI era and each of the speakers are going to give their valuable inputs in terms of how they are doing it today and how they are viewing. It's going to transform over the next few years, right?

I'll start with the questions. Maybe if you can answer, Avinash, first. What is the most impactful AI use case today and what has changed internally in your current org, previous orgs?

Yeah. As we know, we are at the point of disruption. AI has been really impactful in changing overall customer experience from being reactive to moving towards predictable, proactive customer experience in a very scalable way where businesses can keep their costs in check, but everything else scales along in terms of human transformation and that is where AI has been pretty significant, right?

If you think about from customer experience perspective or for the companies that each one of us sell our products to, AI literally has to make it easier for our customers, smarter and efficient for our business internally, and then better for our people. That's when I think the combination works because all three sides of the world needs to move in tandem so that our customers get the best out of the investments made, right? Like AI can be costly if it's not really giving you productive outcomes that you invested in.

And we heard the gain side like the keynote session today, build versus buy, is a real problem to have on the table, right? Because everybody's like, hey, let's build because I've got AI tokens, I've got budgets for it, but then can we do it in a scalable way? And that's when I think we, I mean, I have experienced using Beacon as a product in my previous three organizations and landed at very, very scalable implementations and automation around support.

And my philosophy is, like I said, if it can be easier for our customers, better for our business in terms of efficiency and easier overall for employees to sort of adopt and change, that's when it becomes more relevant for me. And I think that's been the impact I have seen AI bring to our businesses overall. Yeah, so thanks, Avinash.

I'll talk about a specific use case that we have used Beacon for. We were challenged with faster onboarding. Our customers kept telling us and business sales kept telling us that if you do not increase or reduce the time to first value, we are losing customers' confidence on the product.

So we leveraged Beacon and transformed the entire customer onboarding journey. So data ingestion, I'm sure we have people from Workday also sitting here. In the HRMS industry, one of the biggest challenges is ensuring that the data from the customer is imported into the platform.

So how do you get the platform to quickly take the data, ingest the data? Customers give us data in various different formats. So how do we make the data consistent to our platform?

So we used AI from Beacon to get that data in. Similarly, customers' policies. In HRMS, customer policies are, again, varying.

How do we get those policies quickly into the platform? And finally, all customers buy HRMS because of payroll. So how do you ensure that the payrolls are correctly configured?

Again, you leverage the power of AI, and those are the places where we've got the maximum impact. Thank you. Do you want to add anything?

No.

[Speaker 2]
Okay.

[Speaker 1]
Where do you think the handoff breaks today, Avinash? Like when you're transforming from, let's say, a pre-sale stage to sales and then to implementation, eventually when it comes to CS, I'm sure everybody has experienced that a lot of promises made at the top of the funnel, but all of those either turn into retention challenges or into support tickets at the end. So where do you think this entire journey is breaking off today?

Yeah, I think the handoff breakage points that I have experienced the most has been the one where the handoff happens to our customer success and support organization post-implementation and sales phase because it's not like there isn't information or someone's not captured it. It's just scattered through different emails, project documents, gone conversations, some scoping discussions that happen in the start of the whole process, and that doesn't sort of seamlessly flow to the customer success and support org. They just get to see what's black and white in the tool that's documented, right?

And that's where probably the attempt when the initial meeting happens is, hey, I'm your customer success manager. I'm responsible for these things. And then they figure out that they have to reconstruct the whole understanding of that business because they want to build that relationship forward.

And customers typically have to repeat themselves, which is where I think the trust sort of goes down right from the start. Otherwise, they've been sold the whole world during the sales phase. So I think that's where the most handoff breakage point I have experienced, and again, we are really lucky that AI as a technology today exists to sort of solve that in a very seamless way if we do everything right.

And then right from the day one, if customer success team and support teams know the context, know the background, know the history, all the conversation, they can be really, really impactful right from the first conversation because they weigh in on the problem statements that customers really had versus check again, like, hey, I'm so-and-so. What do you want me to sort of focus on? That shouldn't be the point, right?

So typically, we all have our org structures in the company. We all have designated escalation points, right people in the right places, but your customers don't need to experience the org chart. They need to experience the company, right?

So it has to be that seamless handoff that it is invisible to the customers. That's when I think the right handoff happens. And again, you know, how do we automate that portion to give that experience to the customers is where the loyalty sort of keeps going higher and higher versus a breakage point right when the relationship starts, right?

So that's been my experience of where the handoff breaks and how we handle it, like, make it very seamless. And Beacon's been, like, really, really impactful for us in terms of bringing that information up front, in terms of making that relational conversation really relevant from day one.

[Speaker 2]
I think as well, when we hand over to customer success, it should be a success. And if the first thing that we do with them is say, oh, let's look at transforming what you've just built. Well, I think that's rubbish.

I think we should be saying, let's use AI to speed up the whole implementation. Let's use the time then with the consultants to really look at their processes, re-engineer for them. Give them the work-life reinvention that we want to give them.

Actually give them that return on investment before we even have the success. Give it to the customer success team. Make that a good journey for them as well so they're starting from a good point.

[Speaker 1]
Yeah, I think retention starts when you're actually handing off the project to the customer success. Today, I guess, it's kind of a customer success job. With AI, I think we're going to change that and make sure that they're in the best place to take it forward and keep the retention going on.

In this journey, Subha, maybe you want to take this, how do you think AI will help us to carry this entire context? Let's say, when I'm thinking of hand-off, starting from pre-sales until customer success, how do you think we should be leveraging AI to make sure that none of that context which happened in the first pre-sales call to the last implementation call that happened is not missed out and, one, it is used to do the implementation itself, and then while handing off to the CS also, how do we make sure that it comes and what has been your experience so far?

So, I think across most of the sessions in this conference, I see that this being the common theme of how do we ensure consistency of context, correctness of context, and many sessions gives a lot of solution ideas and they're lovely sessions. What I'll do is I'll pick a specific use case of how, which is one of the areas where we see the hand-off was breaking. Product goes at a velocity sometimes which is either much slower than what customers expect or sometimes even much faster than what customers expect, and with the changes in product, the customer experience team also needs to keep up with what's changing in the product and then ensure that the same information gets translated to customers when they have questions on how to leverage the product or what concerns they have with their own specific configurations using the product, and we saw that as a gap. One of my support engineers would give A answer and another support engineer would give B answer to the customer, and we decided that, hey, we will leverage AI and we'll build this whole lake of information which has both help docs plus product roadmaps plus customer configuration data into a data lake.

We built a data lake. We call it the brain on Snowflake, and on top of Snowflake, we stitched together AI workflows using beacon and some of the other AI capabilities, so now every customer success or customer experience person in the team has a copilot, and when they see a ticket coming in, they can answer the question correctly and also with speed. The next evolution of that, we were talking about it, is that how do we take this to our customers, so we want to now just not solve what we call as how-to queries for our customers, but also how do we troubleshoot, so we want to take troubleshooting to our customers so that the handoff between customers and support agents is reduced, and more importantly, there's more ROI for the board.

Of course, I think everybody here are looking towards that. Yeah, Joan, you want to add anything? No.

Yeah, I mean, just quickly, right, like from show of hands, like all the ones here who care about deflection as a metric, right, like do you track it when you've implemented AI and you want to track deflection? You do? So, I mean, for me, deflection is an internal metric that we can be proud about saying 30%, 40%, 50%, and all of that, but actual metric for customers is have we eliminated the amount of effort that they used to do pre-AI to post-AI?

Is the effort really reduced? And that's the deflection actuals for the calculation, right? There are different ways to measure it, but deflection isn't just about a ticket that was coming in and then an automated answer sort of went out.

People measure deflection differently based on the tools they use, but the real deflection is have we reduced the complexity, the kind of effort that customers used to put yesterday versus today, right? That's, I think, the moving metric, and that's where AI has been really, really impactful. Along with the CSAT not dropping.

[Speaker 2]
Absolutely.

[Speaker 1]
Yeah, I mean, you don't want to hinder the experience by automating things that they didn't ever want to, right? So, it has to be a right balance, and that's when customers get the best of both worlds, I think. Yeah.

I think while we sing a lot of praises for the entire orchestration of AI, where do you think it breaks today and who do you think should be owner and where should you think the human should be a part of this entire workflow in your experience? Where do you think these workflows are breaking today and how do you think we should handle them? Yeah, I mean, for me, it's pretty simple.

If technology breaks or process breaks, it doesn't matter. Humans have to show up, show the ownership, grab it by the neck and solve the problem because for the ones that have invested in your product or service, it doesn't matter who is solving the problem end of the day as long as it's consistently timely solved, right? So, when technology breaks, someone has to show up and take the ownership and solve the problem before explaining why there was a delay, there was a glitch, there was a wrong answer.

That's all your root cause analysis that everybody can, you know, noodle on after you have sort of moved them from the blockade they had, right? So, for me, I think that's the most important aspect, whether AI transforms, evolves, disrupts, there has to be a show of ownership and accountability. Someone at the leadership or company has to really own that problem for the customers.

That's when I think it all works out or balances really well.

[Speaker 2]
I think the use of AI really needs to be human-led. I think you do your workshops up front. You know, we've worked with Beacon and we've done the workshops so we get all the validation rules, the points of when does it ask for help and things like that.

But exactly as you said, Abby, you need to be present with the customers so that they don't feel they're just there on their own and they don't really have anybody to talk to. You need to share that journey with them and you need to help make decisions so that the next time around, whether it's Beacon or any other AI, it knows what to do and has a sensible answer and something that's repeatable. So, I think it's very important that that journey of the customer continues to have the human aspect and not whatever tools we use.

[Speaker 1]
I have a different perspective. The perspective is that there are certain situations where you don't want AI to come in. I'll give you an example.

In India, we run a monthly pay cycle. Every customer of ours runs a monthly pay cycle for payroll. And they typically run it between the 25th to the 31st of the month, depending upon what their pay cycles are.

And most of the time, the payroll administrator is under tremendous pressure to ensure that the numbers are right, the payout goes on the right time, the bank transfer file goes out to the bank at the right time so that the employees get paid. And the HR is under a lot of pressure to ensure that the payouts are happening, especially when you have blue-collar workforce which are unionized. And if the payouts don't go at the right time, there's somebody at the CHRO's door banging that the payment hasn't come in.

So when we get payroll-related concerns from our customers, especially the last week of the month or the first week of the month, we don't push those tickets through AI. We get them to a human, and a human talks to the payroll administrator to ensure that the customer is happy, the problem is solved, and the employees are paid on time. So sometimes we don't get AI into the loop.

But do you think that at some point that those kind of deterministic things are also handled by AI?

[Speaker 2]
I think some of them will be, yeah. I think, I know we've got a concept with our product which is the intelligent assistant, and it is very much to get those questions answered without somebody else being involved. But when it comes down to something that's really tricky like someone's pay is wrong or whatever, if that happens, then I think you need the human then.

But I think a lot of things can be set up using AI so that it can answer the basics and keep people involved enough with their own pay or WFM or anything like that, but they know where to go if they need a quick answer for something.

[Speaker 1]
An interesting question. In like 18 months from now, do you think the handoffs still exist or all of these pre-sales, sales, implementation, everything is being done by agents, and on the other end also you have a lot of agents dealing with this, or do you think there is still need of a CS team and implementation team and all of this? What are your views and what are your predictions?

I hope there is.

[Speaker 2]
We all need our jobs.

[Speaker 1]
I love mine, for sure. But I think what we do today to what we will do in the next six, 12, 18 months may be more complex, may be more different, while AI can take over a lot of redundant thinking jobs for us. But I think evolution has been one of the standing truth for the whole human race, and we will keep evolving.

Jobs will keep evolving. I think there will be very optimistic view of these jobs doing something more advanced and something more adventurous, if I may, in the future than today, and that's all because AI is going to probably uncomplicate some of the things that we deal with today. So it's the playbooks, it's the workflows.

How easy can we document what happens today in a simple storyline that AI can understand because contextual engineering is so relevant. It's a garbage in, garbage out situation in the AI right now. So we need to get it right before making it smarter.

That's what I tell my teams, that let's make it easier for our customers before making it smart. They need to get used to a certain pattern and workflows, and then you keep adding smartness to it. That's when they feel the ingestion of real investments that we're making.

Don't overcomplicate it right from the get-go. That's been my philosophy, but I think it will be relevant, but we will end up doing something else in 18 months from now.

[Speaker 2]
I think it will be relevant as well. I don't know necessarily. I know what the future is going to look like.

I know in Zealous, our question we ask ourselves is, what's next? And we keep asking that, what's next, what's next? Once we get somewhere, we want to do the next thing.

I think what it will do, I think from consultancy and implementation point of view, project management, is it will free people up to really genuinely do the transformations. And we'll use the tooling to do the more mundane, drudge sort of work if we can, but also use its intelligence to suggest things to us that we maybe can then look at and say, well, you know what, let's do something with that. Let's make that repeatable and scalable.

Let's actually use that to give us some time back to do other things. Use it as a tool, not just to get data onto the system, but to actually suggest other things to us and to help with that transformation journey for the customer. So genuinely, they get that return on investment.

It's not just get to the end of the project, wipe your brow, thank God that's over, and move on. What are we going to do next with it? What can it suggest?

So yeah, I think there's a role for it. I don't think it will replace the humans, though. Sure.

[Speaker 1]
I think my perspective is that value, when communicated correctly, needs a human in the loop. There's this famous book called To Sell is Human by an author called Daniel Pink. If you haven't read it, please read it.

He talks about even parenting being sales. So you will need the human in the loop to build better relationships. You will need human in the loop to do better transformations.

So I feel that AI is giving us to work on steroids. So essentially, we are better off as humans with AI with us. So that's the future, is what I think.

Okay. From a metric standpoint, Avinash, how are you tracking the productivity that you're getting from AI, maybe in the implementation journey or otherwise also? What kind of metrics are you tracking internally?

How are they looking right now? I think it starts with different metrics and different milestones. Number one, like I said, your customer experience points cannot move down.

It has to keep going up. That's our measure, and that's why we take an approach of pre-discussing these changes with our customers and we go to our cab and check what do they feel about these changes that we are thinking about? How do they read it?

Are there something else that matters to them? And then make it more in-house. But I think from other metrics standpoint, what time does it save for people?

How can I reinvest that time to more transmit the work like Joanne just quoted? Or get them to doing something more challenging that they learn and make their careers to the next stage. That's how we are looking at from reinvestment of that time perspective, but definitely customer experience is at the center of every change or transformation that we want to do because without that moving any percentage points up, it's meaningless to go through any automation that will come in and probably show greater numbers.

But if it's not felt from our customer base, it's worthless for us to invest in technology or in any process change. So that's the number one metric I care about. Thank you.

So AI, I would say just not AI, but automation. AI being getting automation to be fast-tracked is a line item on our annual operating plan. So it's here and now.

But the interesting piece is that the metric that we should, one is obviously productivity is an important metric obviously, but I think quality is an even more important metric because end of the day you don't want to drop your CSATs, you don't want to drop your NPS. And another important aspect of the numbers is that what's your cost of AI? I mean we keep seeing the output results, but I do have an AI forward deployed engineer in my team and one of the key metrics that I have given him is what's the cost of tokens versus what's the cost of effort that we are deducing?

We need to constantly look at that, right? Because end of the day, effort is one, but what's the effort? What are you paying your employee?

What's the cost per employee on that particular activity versus the cost of tokens? You need to also compare those metrics continuously. How is that looking right now?

In some cases it's not good, but in many cases it is significantly improving. But the important part is, and in the previous session, the experimentation that you have to do and getting your experiments right before you scale is important. Only then you will start seeing the right return of investment.

Understood. Joan, do you want to add anything?

[Speaker 2]
I think from what we're doing with AI at the minute, it's the speed of getting the customer's data on so they can actually see what they're working with. And that's a bit harder to measure, although we can measure and have measured how quickly we get that data onto the system and how much time that's saved us. But it's harder to measure the impact on the customer when they can see their own data and actually start to make sense of it, as opposed to waiting until parallel run or whatever phase that you're at, or EIT or something like that.

So we are measuring it as well, and we're definitely seeing benefits of that.

[Speaker 1]
Benefits of mostly productivity is what you're referring to?

[Speaker 2]
Absolutely, productivity, but also engagement with the customer. It's more time to engage as opposed to, that's done, quickly move into the next phase, quickly move to the next phase, right, we have to have a go live. There's time there to explore things with them and help them transform.

Because once they see that data on the system, that's their own data, it's their own data, and they can start to think about things. And that's when you get them really involved and engaged. So that's a bit harder to measure, but it is there, it is present.

[Speaker 1]
She makes a very valid point. The time that you spend on adoption, the first adoption, is extremely important. It's just not getting the customer through the door on using the product, but hand-holding them on specific use cases and getting them to actually leverage those use cases is important.

What Johan just said is that we are now getting more time to do those activities rather than just moving from one customer to another customer to another customer. So yes, absolutely. And that eventually helps you in the retention.

Oh yeah, exactly.

[Speaker 2]
Overall adoption.

[Speaker 1]
In fact, retention points, we are able to attribute retention to AI initiatives now. Like GRR has moved three percentage points more because of AI initiatives. Interesting.

Okay, I think that is it. From my end, I open the floor up for questions. If you can quickly submit the questions, we'll go to that.

Yeah, we have it here. The first question is, what happens when something promised during sales cannot actually be delivered during implementation? On this, I will give it to the gentleman from Birthday.

This was my question to him yesterday. And his answer was, slap the sales guys. Slap them on the head.

[Speaker 2]
I think we all know the answer to that one is that somebody picks up the buck for that. And if it's the implementation team, or worse, if it goes as far as the sales team, that's a hard fail. That's a really hard one to come back on.

If we free up time for the implementation, we can at least try and recover some of that. It shouldn't happen. We all know that, but it does.

Even a slight of how people say things can influence what a customer thinks they hear. Not even what's written down on the page. I've been a customer, so I know.

11 days into contract, oh, by the way, we don't have a Republic of Ireland payroll, but we'd like you to work with us. I was like, what? Seriously?

Okay, so I'd say it is a fail, but it's not a complete fail if we give people time to recover from it. But really, we want to get it right first time, I think, for everybody. And it's what you said.

I think you said it earlier. Everybody has to have that same story. We must all know what can be delivered.

[Speaker 1]
Do you think AI can change this false promises, maybe, to some extent? Yeah, I mean, it can hold people accountable and show where the breakage points are to make those adjustments and changes. But when something that's sales sold cannot be true, I think that sort of also shows the spirit of the organization and the leadership.

Are they willing to show up and make those amends and make it right and accept and change the course? Whether that means that, hey, we may not be the right product for the expectations you have. Versus here are things we can truly deliver.

Let's work on this journey and then take it forward, right? Like, it probably tests the spirit of the leadership and the whole organization. Hence, I think, it's the kind of show of that human in loop.

I have an idea for you, Harsha, is that, and I've been contemplating about this, that during the sales cycle, the biggest challenge that the sales people face is they need to retire their quota. And they have a certain amount of time by which they need to retire their quota. And they have long lead times for sales closure, right?

Their problem is also real. We can complain about them, but without them, we don't have our jobs. Now, the point is, can AI re-imagine the customer's ability to evaluate the product at a significant high velocity?

Because today, the biggest challenge that I see is that these customers need to spend their time, significant amount of time to evaluate the nuances of the product during the buy cycle. Can we shorten that life cycle but yet improve the quality so that the customers get the right visibility? And then maybe there will be no false commitments because customers would have self-discovered the capabilities of the platform.

I think, yeah, few platforms are, I guess, working towards it. But I think like Avinash mentioned, once they're accountable, I think then the practices will eventually change. That's the only way.

There is no other way to deliver on whatever promises that they have made. I'll move on to the next question. How do you keep AI from carrying forward the outdated or incorrect information from earlier in the customer journey?

I think in my view, it's a constant effort of flagging every wrong information and giving it right context. It's an iterative improvement because like we know, AI learns from every interaction and those interactions, if we have placed right people at the right moments of truth and they keep correcting and flagging because, you know, I think we have already evolved on the AI input-output system, but I think governance models now are really, really important on how do you govern the information being right or relevant and have those flags in place to keep correcting because that's when accuracy improves.

I think it's as simple, but we need to really write it down and put the human in loop for those judgments to happen. Yeah. Do you want to add anything?

I mean, nothing much. I mean, I think what Avinash said, and then how do you solve this technically? I mean, you ensure that all your channels are tracked.

You get all your recordings done through different, like you have Teams, you have Gong, you have Aloha, whatever, just ensure that you get the context and keep ensuring that the context is refreshed from time to time. I mean, I don't think there is a simple answer to that. You just have to execute it.

Yeah, I think one thing in at least our practical experience that we have seen around this is constantly generating a fit-and-gap kind of outcome so that we keep knowing if there is any incorrect information that is either promised or a request or something, and that's not possible, right? So if you're able to generate those fit-and-gap documents or steps so that at every stage we know exactly in terms of what is not fitting into and what is fitting into, so that way it is, because garbage in, garbage out, so if I'm eliminating certain things or clarifying certain things early in the stages, then I think the overall outcome would be much better. At least that has been our experience while working with multiple clients on the implementation journey.

We'll move on to the next. How do you define the moment a customer has received the first real value rather than just completed implementation, and who owns this milestone?

[Speaker 2]
That's a tricky one.

[Speaker 1]
No, I don't think it's tricky at all. No, I think that's an easy one. Black and white.

[Speaker 2]
I think if you're sitting with a customer and say, that used to be such a horrible thing and we used to spend hours on that and that's not given us any value, you hear a customer say something like that or that's been a process that's involved 15 people and now we can do it with three. Anything like that is a win. I think that's really easy with a customer.

A really simple one is where a customer's lived with a hierarchy that has been there for year dock, it doesn't give them what they need in reporting, and you put their data onto the system really quickly and they can see their own hierarchy, HR hierarchy, and you can actually allow them to move it around and move it to what they want it to be, and then suddenly their reports become easier. That's a win. Straight away, that's a big win for them.

They already see a return on that investment and this is usually the people who are doing the doing, not the chief operating officers and things like that, but even so, that's a feedback up the line to people to say, we're starting to see a good return. If you can get them bought into that, they'll put more effort in again to do more and think beyond just the basics. So I think you can see that quite early in the journey if you really put the effort in with them.

[Speaker 1]
I should own it. I think the way I agree with the operational elements, so I'm just quoted, but for me, the real value for that whole transaction is by investing in your product or service, if customers can focus on their day jobs 100% and not worry about you guys learning your product or maintaining your product or whatever that is, that's where the big win is because that's why they first place chose you or believed in your product or service, so that you guys can take care of them and then they focus on their rest of the responsibilities or duties. So that's my balance between how do I judge a real win for the customer when we have sold a service or a product.

In my view, the customer champion who's championing the implementation of the product, that person owns the first value. If they can't do the change management within their organization, then as a provider, you can only do so much, but that person needs to own and if you don't have a champion on the team, you won't succeed. I think it's kind of set up for failure kind of a situation.

Yeah. Okay, we'll move on to the next question. Where is the customer journey do you think CS should be introduced?

[Speaker 2]
I'm going to go first.

[Speaker 1]
Okay, go ahead.

[Speaker 2]
We introduced it at Sales Handover, right up front, because they need to know why is this customer bought our product? What do they want to succeed with? What are we ultimately trying to get for them?

And what else do we need to know about them? So we bring them in right up front.

[Speaker 1]
Is this after certain learnings or this has been the process at Sales for quite some time?

[Speaker 2]
I'd love to say it was forever, but it has been an evolving thing. And years ago, and I do mean years ago, we used to all sit in a room in Bristol every single person who might be involved in that project would be there for a whole day. And we did a full handover for a whole day.

Reality moves in, you can't afford that anymore. And it evolved one way and then it changed to another and we've really pushed the customer success management, customer success teams. And so we said, you know, they need to come in at the very beginning.

So it has evolved over the years, but certainly the last five years maybe, we are very keen to get them involved up front. And they are a genuine great help, you know, because they will drive that story as well for you.

[Speaker 1]
Are they being very active in this journey or passive?

[Speaker 2]
Very active. Very active? Yes, they are very active.

[Speaker 1]
Yeah, fully aligned with John. We'll move on to the next.

[Speaker 2]
And we did it from day one, just kidding.

[Speaker 1]
What do your process look like after the customer renewal? Do you have like a similar handoff from account manager to CSM and you do an initial sales to CS process? I think, I mean, part of that has been still traditional.

Some of the earlier answers sort of addressed to that. But I think that is why the whole like, you know, the keynote today made a difference to me because the way customer success as the whole world is sort of transferring that knowledge, the intimate knowledge that only a few people hold and how that now becomes organic to most of the organization. That's where I think these processes take care of themselves because if we are doing the things we discussed in the right, most easiest way, these then become the, you know, non-negotiable change management for the company, right?

Because it's moved organically to the smoothness that you want. So these just become tick in the box because you're no longer worried about the whole focus on the churn, retention and all that because your percentage points on the other side are so up that your customers are engaged in the right way and the right fashion. So I think these are processes that will evolve.

But some of the other words like we already discussed.

[Speaker 2]
Yeah. I know with us, actually, when a customer in use, and what they're buying as well, so not just renewing their license but maybe buying extra services or whatever, it's the CSM led because they've been working with that customer all along since Hypercare really.

[Speaker 1]
Yeah.

[Speaker 2]
Long before that but really at Hypercare we handed over. And they're the ones that know what do they want to do next? What are they trying to achieve?

And so they lead that for us. They will come back and do another form of sales turnover to say what are the next steps that they're trying to get? And that's where they have that value because they're dealing with those customers and their monthly catch-up calls and, you know, the NPS scores and things like that.

They're dealing with all of that and the problems, whatever they may be. So they drive that. Okay.

[Speaker 1]
How do you decide which part of the handoff AI should own versus where a person still needs help to step in? In my head, it's an analogy, right? If you know what's the time your nanny needs to spend with the kid versus the parent, that's exactly the balance between how much AI should handle versus human needs to get involved because emotional and the judgment calls needs to be really well-guarded, well-governed, so that you keep iteratively improving that experience.

And that's how your baby would sort of grow up to be a very sane individual. And that's where this whole process for me balances out. I was more interested in the next one.

[Speaker 2]
Yeah, the next question is exclusively for you.

[Speaker 1]
Very, very interested in the next one. So I think very, very valid question. Lovely question.

The starting point in my view is understanding the customer's product adoption maturity roadmap, defining the matrices that will allow you to understand where they're in the journey. In simple words, we call it as product adoption matrix, but it's got both depth and width. So get those numbers and quantifying that into a product adoption score and then able to see your customers where they're in their tenure and what scores they should achieve so that they don't churn.

If you can do some regression analysis, that is the starting point. And then feeding that information into your customer experience executive so that they can see that and then you can drive some actions around where those scores are. I think that's when you start to build the lake and then you can add all of these as add-ons.

But getting the product usage matrices in the data lake is for me the way you get the maximum bang for the buck. Yeah, I think that gives you more insights for you to take the decisions. And mapping that to the tenure of the customer.

Like you can get a score of 80, but a score of 80, assume 100 is the scale, you would not get 80 on the month 3. Month 3, if you're at 40, you're good. So you need to define those milestones as well.

Yeah. Okay. We'll move on to the next.

Where do you see AI actually becoming the bottleneck or where humans are relying on how much of AI and they shouldn't be? I think if your businesses cannot scale without scaling the cost of operations, that's the bottleneck, I think. Because all the investment sort of goes on toss if it's not making sense from efficiency of the business perspective.

That's what I probably would like. Yeah. Okay.

We'll move on. We have a minute. Yeah.

When you don't have more budget to spend, how can we leverage the AI we have to fill the gaps? In our case, we are a Google Workspace company with use of Gemini Pro, okay? Yeah, I think talk to the board.

Get more budget, because again, money or budget is not the real answer. It's how much transformation or outcomes are you aiming for and what does it require? If you can relay that story in a way that people understand that it will take money now, but it saves in the future for scale, I think people won't be hesitant in sort of planning for that.

It may be a quarter difference, but then if they see the value in investment, everybody will invest. I would say that for sure. Till then, I think Gemini models are improving.

Maybe you can use it till the board approves. I think you need to have an AI gatekeeper in your team. Somebody who's constantly looking at what's the spend, because that's extremely important.

If you're able to justify your spends and monitor them, I think you can get more money if you're able to justify your spends. At least that has been our experience so far. We'll take the last question.

What is your view on a sales to success handover for an expansion on an existing logo when it is not a renewal, but a new product or LOB was sold? How would this differ for a new logo handover?

[Speaker 2]
I don't know that it's hugely different from a brand new customer. We still need to know what are the goals, what are they trying to achieve, what can we help them with, transform, is it scalable, repeatable, all the things that we've all talked about. We still need to find that out and then make use of what's available to us to help them achieve it.

[Speaker 1]
My point is that we have better control. You want to take one more? You talked about when not to use AI, how do you efficiently route the right things to the right place without important customers getting caught in AI.

I think you're the one. Parallel example, it is directed at you, I guess. Go ahead.

I'll just tell you what we tried. We're trying. We don't have a right answer for it.

What we're trying to do is to keep a log of every time we get these tickets where we could not get AI to answer, we are starting to keep a log. It's a tedious process, but that log is helping us to constantly analyze where those gaps are, see what improvements we need to make not just to our help dogs, but also extremely important is what signals are we getting to improve product experience? Because those signals are increasingly important.

It's not about solving the customer's queries. How can you prevent those queries to come? And then give that feedback back to product so that they can improve the product experiences.

I think that's where the maximum value is. I think that's the eventual outcome. That's the value that you're trying to drive.

Okay, then. Thank you very much, guys.

[Speaker 2]
Thank you so much.