Beyond NPS - How to reimagine CX using operational data


Richard owen - one of the inventors or NPS came on the show to talk about Customer Experience and the re-invention of it using exiting operational data. Too long - CX has been viewed in the context of surveys (and only surveys) - Richard is proposing a new model - based on machine learning and predictive models, on operational data and calibrated by surveys.









Recording Transcript:


  • Alright, so we have another episode of broadcast today I'm joined with Richard Oh, and my good friend, I've known Richard for almost

  • 06:06Five, six years I've competed with them, I think. I think I competed with them, but I don't think he thinks I competed with them. And of course, my, my co host Romy Mahajan Richard. Welcome to the show.

  • richard owen 06:17Thanks. Good to be here.

  • Vivek Bhaskaran 06:19Awesome. So you invented NPS 15 years ago and now you're reinventing NPS tell us more about sex cognition. Yeah.

  • richard owen 06:27Thanks. Thanks for that. Well, first of all, I mean, I think that my my team invented MPs one hates to think of yourself as being sort of the

  • 06:35The sole source of this as much as anyone can invent anything I think NPS was by most people's

  • 06:44Estimation I think a tremendous success, right, and broadly adopted, I think you've got a lot of focus and attention.

  • 06:50On CX I think that when you look back now over you know what's approaching you know 20 years since since the concept was advanced

  • 06:59I think, at the same time, we have to be a bit sober and understanding how far companies have come in terms of actually improving their customer experience. And I think the results are

  • 07:08More disappointing than I think we'd like for a lot of companies measuring NPS and talking about customer experience has become a bit of a bromide. Do you say you do it.

  • 07:19You know, you might publish the numbers. It doesn't change. Worst case it might even be the wrong set of numbers and it might just be there to pay people a bonus

  • 07:29And so from our perspective there really has to be something more. And like a lot of companies were taking advantage of what I think is so next generation set of technologies that are being developed as part of

  • 07:41Machine learning technologies to really, I think, try and take the whole concept of net promoter score up to the next level. So that's, that's our focus

  • Vivek Bhaskaran 07:50And and from what I understand you've kind of like, you know, you're starting this company or not raised you didn't go back to subtle ventures and raise money years to kind of doing it yourself. This time around.

  • richard owen 08:04Assuming that

  • 08:05You know,

  • 08:08I think first of all is, you know, yourselves. I mean, there's, there's, there's such multiple choices entrepreneurs have right at some some level.

  • 08:16You can start a business with party capital and that used to be impossible 20 years ago you know entrepreneurs bootstrap businesses and then raise money.

  • 08:25And it's only recently, I would argue that that you had companies raising capital lots of capital on the PowerPoint and that can be quite appealing to an entrepreneur.

  • 08:36But you always have to remember, there's a bit of an asymmetry between what a professional investor wants and what the entrepreneur wants the professional investor is playing a portfolio game.

  • 08:47And the entrepreneurs playing a portfolio of one. Yeah. And so I think that we're going to learn that looking at the stats and the failure rates. We're going to learn that for a lot of companies taking capital too soon, taking too much capital.

  • 09:05Actually represents to higher risk for the entrepreneur. It's an asymmetry of risk. It's something that works pretty well for the VC, but not necessary well for the entrepreneur.

  • 09:16And we're in an era where there's been too much capital chasing too few deals. I would argue in the venture community.

  • 09:23And that usually means decreased returns for limited partners. So it'll be interesting to see, you know, five years from now or 10 years from now when we look back and say it was a golden age for venture backed companies or ACTUALLY, TERRIBLE TIME.

  • 09:37To be taking and an investing venture capital.

  • Vivek Bhaskaran 09:41And so this time around, you're kind of like obviously bootstrapping it you know kind of building it out obviously had a good exit with with

  • 09:49That metrics to Nice. So tell us a little bit, let's jump right into it. Tell us a little bit about sex cognition, what's gonna

  • 09:56What's your kind of rethink of the CX kind of vision. I think you have a bold vision. I mean, obviously, I know you. I read it. It's a bold vision, I can. I'm behind it does, it does tell us a little bit more

  • richard owen 10:08Well, I think the simple premise is this. There's the, the biggest challenge facing customer experience professionals, is that the data is extremely limited. And what we all want.

  • 10:22Is we want data on every customer every second, we'd like to understand the health of every one of our customers, we'd like to understand if something terrible happens to one of our largest accounts yesterday.

  • 10:33Why, what happened. Can we act on it. And we've been anchored in this universe which you guys very well understand which is

  • 10:42We, the only source of data we can use almost exclusively has been survey data and it's certainly the business model that we used at SAP metrics.

  • 10:50That the problem that model is that the the fundamental boat anchor for that is always the participation of

  • 10:58Customers in the process. And at the very best really really well executed survey programs, you know, get data from customers, maybe twice a year.

  • 11:08And they get maybe data for 20 30% of their customers and I'm talking about business to business B to C would probably be a lot lower than that.

  • 11:15And so saying that you get twice a year snapshots of a customer and only 30% of your customers to boot.

  • 11:22isn't exactly, sort of, you know, 2020 levels of data that you'd expect in almost any other business. So we worked really for a couple of years to try and find a solution to the problem. And it turns out that

  • 11:37The solution was staring us right in the face. Number one surveys are really, really useful.

  • 11:42But they're really, really useful for calibrating machine learning engines that are really, really useful for predicting where customers are

  • 11:48So instead of saying my survey is the only source of data, my survey becomes a great way to teach a machine to predict what customers are actually going to be thinking

  • Vivek Bhaskaran 12:03And the question. Yeah.

  • 12:04Yeah, good question. Yeah.

  • 12:05So the are you looking at behavioral data to make the prediction operational data, what kind of data to give you

  • richard owen 12:13Because the end of the day, you need a data source that is responsive, right, because if you're going to predict offered in these to be responsive.

  • 12:20So the responsive data sets, essentially is the operational data set. So it's really the data exhaust for companies, you know, log, good job on that. Let's make it really simple every time.

  • 12:33In most companies a customer incur some sort of experience. They leave a footprint digitally in that company right they something to curse you ship them a product. Well, you knew Sherman product they take delivery of it. Oh, you know it arrived.

  • 12:47It turns out that, as we know, customer journeys are made up of, you know, thousands of these events, but a relatively small number of them turn out to be

  • 12:56massively influential and whether customers are promoters are detractors. So within that entire universe of data.

  • 13:04The challenge is to figure out the ones that were the matter to figure out how to calibrate them and to figure out how to teach them machine.

  • 13:11To say, which might be entirely obvious look if you ship the product late the likelihood the customers are promoted just dropped right and so very simple level you're teaching machines to solve those problems in a bit more of a sophisticated fashion.

  • Romi Mahajan 13:26I have a quick question. You know, Richard. You've talked with NPS and

  • 13:30Measurement of that being sort of a bromide a thing that people do and sort of pretend that they go through the performative act, but it doesn't really necessarily mean anything. I do have a larger question that might seem a little heretical, but

  • 13:44When you think about Experience, customer experience. And you think about measurement right when I experienced something. I'm not trying to be experienced something that should be necessarily measured

  • 13:56So if you have to reduce an info eight emotional thing into data are we talking at cross purposes here is experience. Actually, should it not be measured right or am I crazy

  • richard owen 14:10I look, I think that as a practical matter, the way corporations get their arms around their operations and take action is through measurement

  • 14:20And whether we like it or not, whether it's the best way of doing things, which I think is a legitimate debate.

  • 14:26It's the way in which the command levers of companies get executed. I mean, I think that if you're if you're the CEO of a large complex organization, how do you get

  • 14:371020 30,000 employees to move in a direction. Right. We can do strategy decks.

  • 14:42But the fact is we have sort of the perhaps the equivalence. The military commander control systems and we have that through measurement

  • 14:51And we have that through goal setting, we have that through incentive management that ties and I think that if we ignore that as the instrumental factor behind execution. I think we are

  • 15:02Unable to sort of control companies now your race. Another really interesting point, which is that

  • 15:07We shouldn't presume that measurements is the entire limits of our imagination here, right. I don't think it is.

  • 15:17Possible to make it the dominant and single purpose behind the way we think about customer experience for a couple of reasons. First of all, it can't be inclusive enough of all the elements that make up, what's a good experience to be able to give you a great result second problem is

  • 15:37There is a unquestionable creative elements to how companies formulate experiences. I think this is true of any technology problem though.

  • 15:48What do computers do really well. What does let's let's be buzzy about this. What does artificial intelligence do really well for us as human intelligence.

  • 15:55Artificial Intelligence does a really good job of managing complex sets of data to give humans ideas and to give them direction humans do a fantastic job of taking

  • 16:07Very difficult problems that poorly defined and require tremendous creative input and building on that.

  • 16:14So I look at this as, how could you provide data sets that enable humans and inspire humans to make creative decisions.

  • 16:23And create breakthroughs. And I think that's actually where a lot of AI is going to go. I think AI is going to be less about replacing human decision making and more about augmenting it

  • Romi Mahajan 16:33Yeah, so I'm with you at quantum care and we think about the old language man plus machine, but now we think about it as human plus machine, of course, in a much more language when I'm fully their art and a science and it's all sort of jumbled together so beautifully thank. Thank you.

  • Vivek Bhaskaran 16:52Well,

  • richard owen 16:52Look what we're talking about week AI here, right. So, so, to some degree, we're not here making an argument for strong guy anyway that's not that's not and let's face it, probably 90% of what people are talking about in terms of investable AI today's we absolutely, absolutely.

  • Vivek Bhaskaran 17:11And and Richard I want to switch gears a little bit, you've talked a lot about, like, you know, you know how CX and you had a deep understanding of the CX business over the last 20 years realistically

  • 17:23And and from an organizational perspective. Are you do you think like big organizations or even mid market organizations are ready to kind of say

  • 17:32We're gonna, we're going to stop doing what we were doing for 14 plus years 18 plus years or

  • 17:37Any number pick and pick a number, realistically, and say, like, we are going to really dig into our own data sets kind of tried to kind of map it to machine models and then calibrated up. What's your take on kind of on the on their kind of ability and kind of to rethink themselves.

  • richard owen 17:54No, I think, I think there's two, there's two things for this two aspects to that first of all let's let's acknowledge that we are in a unique moment of time when

  • 18:04A lot more things are possible. The little possible six months ago. And you know, I was on a panel with a group of financial executives, two weeks ago from big banks in North America.

  • 18:17very conservative organizations and they were expressing close to disbelief at the degree to which their organizations.

  • 18:25had opened up the checkbox to invest in digital transformation. I mean, these were career veteran saying we've never seen anything like it in you know 20 years at insert big bank name here.

  • 18:36And so, all bets are off. And I think that's the first thing. Unique moments in time. Having said that, if histories and guide.

  • 18:47We're going to see a curve. We're going to see a group of companies, lead the way be transformative. Some of them will succeed and achieve great results. Others will maybe fail.

  • 19:00Will see a conservative group of companies that will not

  • 19:05That may even just go to the wall before they change either because they can't or they want and that's played out with every technology transformation or frankly back to the Industrial Revolution.

  • 19:16Right. So I think we can't generalize and say, Will business or change to answer your question directly

  • 19:21I feel will be leaders that will be laggards I think we are probably going to see more experimentation and more leaders, because no leadership team consider around and say the world hasn't changed. You can't

  • Vivek Bhaskaran 19:33Yeah, that's a good one. I mean, you know, obviously, you started. I mean, I obviously other private knowledge to you launched the company literally on May March, the first week of March, I guess.

  • 19:45Literally, like, you know, I'm still in New

  • 19:47Business literally the first week of March, right, when we were things were going into lockdown mode or in the beginning of this kind of pandemic or gonna bring of the realization of the pandemic, quite frankly,

  • 19:57But that didn't stop you right i mean i think this is a great you make a great point around like you know transformation and changes in the air, even in mid to large companies, which was not there. Six months ago. And so you can tap on that.

  • richard owen 20:10Don't have Rami walking

  • Romi Mahajan 20:11Sorry about your go to market because

  • 20:14I would assume. And I, you know, I've known you since her SAP methods days and obviously an amazing company and so on and so forth.

  • 20:20You work only with top executives. I mean, because you know the change has to start there. In other words, they'll work you're doing. Isn't it much more around strategic cultural approach process approach to CX more than anything. So are you only working really CEOs or am I wrong here.

  • richard owen 20:38So, so I think, you know, it might have been the presumption of ours that, you know, it would be something that would work more closely with CEOs around what I find still true about the CX industry.

  • 20:50Is that your co sponsorship is essential, but engagement is often more limited so you know what tends to happen when this works well is

  • 21:00You know, the CEO finds a proxy within the leadership team that is somebody who has a burning, but I think frankly a passion for it. I don't think too many of these people are sort of just

  • 21:12arm's length. They have a passion belief for CX. But I think more importantly they have that clear role where they are a business person.

  • 21:21Yeah, and CX is a weapon to accomplish a business objective, you know, one of, one of the things I was reflecting on the other day with

  • 21:28With a group of CX lead as was really three types of CX professionals emotionally or, more accurately, perhaps, everyone's a hybrid of the three combination

  • 21:38One attributes is almost the religious crowd. I got into CX because I'm on a moral crusade. This is something we need to go fix

  • 21:47The second group got into it because they love serving

  • 21:51I'm not joking. I mean, this is why they're in it. Right. The survey authors are and they define their existence by, you know, I think it's really fun to write surveys and collect survey data.

  • 22:02And the third group, I would characterize as basically all about business at the end of the day you know CX is just another tool, but I'm all about you know EBIT da growth cross sell upsell retention reduction.

  • 22:17And you know, you could argue the perfect leader for anything like this is a mix of the three. The issue is

  • 22:24Our challenge in the CX community too often is, we've got too much over weighted towards the religious side we've got people who are in it for the moral reasons and not the business reasons and the CEOs to your point, I mean it for the business reasons.

  • Vivek Bhaskaran 22:39And the steel is and I mean I've had great conversations with chief operating officer is not necessarily always the CEOs. But second, like you said, like the a proxy or some guy.

  • 22:48Like, like, you know, VPN sometimes during the VP of finance is like, dude, I we and I think you've you've you've kind of described it perfectly right, the right to the right person is who use a CX as a weapon to do something else. Really, which is a bit

  • richard owen 23:01Money talks.

  • 23:02I mean I'm constantly amazed these conversations you have where people frame this constantly as again a moral issue.

  • 23:10Right, we have to love our customers because we have to love our customers, right. I appreciate that sentiment.

  • 23:15On the other hand, that's not the boardroom. And frankly, it shouldn't be the boardroom. Right. We do have corporations legally with a fiduciary responsibility to generate shareholder returns that owned by shareholders.

  • 23:28So we have to speak the language of business and we need to, I think, candidly grow up a little bit in the CX community and start realizing, if we can't show financial impact.

  • 23:39Why should anyone invest that time and energy in us right they can write a page in the annual report saying we love our customers to the center of our organization, blah, blah, blah.

  • 23:49But if we want to put resources behind transformation of CX we better start showing that, that equals significant delta to the bottom line. And that requires a different approach, I'm afraid.

  • Romi Mahajan 24:03I know Vegas doing some work Aaron and Rick and I believe you are as well.

  • 24:06Round, if you take the metaphor of your if you're a large company or employees or your customers and you need a CX mentality or methodologies to make your employees happy. Is that an end in itself, or does that also have to be tied to financial

  • richard owen 24:21I, I do believe it's not an end in itself, I do believe at the end of the day, look, I think it's interesting. There was this hot debates sort of dropped out the news with coven but if you remember, a year ago.

  • 24:33You had the likes of jamie diamond out there talking about stakeholder capitalism. Right. And, you know, should capitalism be all about, you know, returns to shareholders and let's go, you know, Milton Friedman cover here is, you know, stop turning into graves. It's gone.

  • 24:51What I would say about it is this, I do not. I think it is to choose with a cliche a false choice to think about

  • 24:59getting engaged employees delighting customers and making profits. I don't understand why we think of that as a trade off. In fact,

  • 25:07I'd make a different argument every successful high performing profitable business is characterized

  • 25:14By engaged employees by engage customers and delighted customers and high profits. And if you think about it, if you said one of those three wasn't true.

  • 25:25Well, how is that business doing so well if it's not profitable the shareholders get sick of it anyway. You have a whole bunch of happy employees, but they'll be unemployed employees, pretty soon.

  • 25:35If the customers aren't engaged. Well, you won't have a growing business and how profits, if the if the employees are not engaged.

  • 25:44Then you cannot generate either happy customers or more profits. So I think it's a three legged stool that we have to stop thinking of as being choices and see as being common characteristics of success.

  • 25:57Or three now. Now, I think one thing we do get fooled by is this notion of employee satisfaction and we have a little cautious here.

  • 26:04Right. You know, I'd be very satisfied if people pay me to do nothing. Right. Um, you know, I could really enjoy my job if I barely had any stress and didn't have to work at all and had great benefits you know

  • 26:17Engaged employees are somewhat different in the sense that they are aligned with the company mission. The mission for customers the mission for profits and they contribute in that way. So,

  • 26:29I'm I'm big on employees, but I'm concerned that too many HR departments, start with a very simplistic premise, which is how do we make sure employees are happy as opposed to how do we make sure that they are engaged and contributing

  • Vivek Bhaskaran 26:45I think yeah and you follow the money trail. I mean, I can see that were like, you know, if you are the head of HR, you don't give a shit about customers you are care about employees and there's a kind of a

  • 26:53A focus on employee experience focus on employee engagement and then the CX guys come in, like, we don't care about the other stuff. And then the CEO and the CEO. It's like, well,

  • 27:02Neither of you guys don't give a shit about. We got it, we got to have a board meeting in three days, but we have to do.

  • 27:07We have to do the financial part really. So I think it's a good way of describing like all three competence contribute to the CX guy ecosystem and

  • 27:14Instead of like everybody

  • 27:16You know, covering their basis and saying that this is my, this is the only game right that's what ends up happening is, like, this is the only game in town. So there because obviously they're trying to sell a product or a service or

  • 27:26That are assigned to sell their vision of what it should be. And I've talked to even HR leaders like, Oh, well, all you need to do is to get all your employees happy. Everything will be fine. Well,

  • 27:34You know I asked any CEO and. But yeah, I mean, they may say that. But realistically, they need to turn it over to the water. That's a good way of the three legged stool is a good way of describing it.

  • Romi Mahajan 27:42issue is whether we can agree or disagree. What these these these you know axioms is an attribution. How do you attribute some financial outcome.

  • 27:54To customer experience or customer satisfaction or customer engagement. Right. It's that attribution cycle, the proof piece, which I think is the Bugaboo here.

  • richard owen 28:06Well, and look, you're a great marketer, so you understand the irony there right marketing exactly the same problem.

  • 28:14You know, I think that it's easy to sort of draw the facts sort of post and to you where we say

  • 28:20You know what companies are really successful. Let's take a look at them and go wow they have characteristics like great customer loyalty. Right.

  • 28:28Um, and I feel like you know in general in business. That's as close as we typically get we we look at successes.

  • 28:38And we try and look post down to why they are successful and we draw conclusions about their characteristics and if you know you guys have been involved in really successful businesses.

  • 28:49And so you know when you're really successful. There's a whole combination of factors that go into that success.

  • Vivek Bhaskaran 28:55Absolutely, absolutely. Man, I mean,

  • 28:58I get Dalio like random my shit happens

  • richard owen 29:01And trying to sort of pass them out and say, well, it was, you know, let me say that it was all about this.

  • 29:08Just, just isn't the reality. So I think that, as you know, I'm I'm I'm big English Premier League fan and as you as you know, Liverpool won the Premier League for the first time in 30 years

  • 29:20I grew up in Liverpool sisters, a big deal, but to win in a sport like that you've got, you know, 11 players on the field 10 in the goalie, you know, saying, you know, well, was it all down to the striker

  • 29:34You know, he know he didn't have nine other field players and a great striker that that didn't that didn't do it. You, you essentially have

  • 29:43Bench you have team strength. And I think in a corporation. Think of those players.

  • 29:48As functional and operational assets they have and and if they all come together if there's enough strength companies perform incredibly well now I think customer experience.

  • 29:58You know, I think the analogy falls over, because I think customer experience is probably one of five or six things that companies really need to do well to be successful, but it's not so

  • Romi Mahajan 30:09I would agree, the only example in business. I've ever seen in terms of direct attribution is Vivek success after meeting me

  • richard owen 30:19That is the case that is the case study that we all tend to think of

  • Vivek Bhaskaran 30:22Yeah, you go on that note, that's funny. Note. But on that note, Richard. Thank you so much for joining us. This has been a great conversation will thank you so much again.

  • richard owen 30:32Always a pleasure. Thanks, guys.

  • Romi Mahajan 30:34Your company grow and prosper.

  • richard owen 30:36Thank you. Yep.

  • Vivek Bhaskaran 30:37Thank you guys. Bye.


91 views

Wait..Why is the "subscribe" in Spanish? 

Find out why..