Martech Conversations: Episode 6
Process of Technology Intelligence
Dorai Thodla: Okay, hi, Pravin. Hi, everyone. So this is kind of interesting one, what you said is exactly right at a very abstract level, right? The same thing can apply to market research and can apply to a business intelligence, or competitive intelligence and competitor intelligence, the level of details change in each one of those things. So let us talk about, I know that we informally agreed on we'll talk about the process, but then when I hear the term process, it puts me to sleep and there are a lot of people who swear by process, you have to have a process. Otherwise, yeah, agree, you need to have something that is repeatable kind of thing. The interesting thing is the amount of fluctuation between same two process. So I'll broadly put it into like three or four buckets, I say, first of all, you need to find the sources, right. And you know, in market research, secondary sources and then primary sources, sometimes they have interactions, like a secondary source will lead to some primary source and some primary source will lead to some secondary sources kind of thing, right? Or market data. So same thing is true, you start with a few set of sources, and we'll elaborate that a little bit. And then you start monitoring the sources and then you kind of validate the sources, and maybe rank them kind of mentally not very strict tracking, but you know that this source is closer to the space that you're talking about than some other source, for example, like a CTO or a CEO working on a startup, that is actually building an artificial intelligence application closer to the space, then the journalists writing about that application, through, you know, an interview process for example, they bring an entirely different angle, but both of them are sources, you're going to get data from both of them for doing it. So once you decide on sources, you start gathering the data from source. And there are a wide variety of sources. I think we talked about a diagram that couple of days ago, and all the web sources of information, right. And so we will not go back to the diagram, but we'll talk informally about what is a good source, then we monitor. Then constantly monitoring the source, and then you'll find out in mounting the sources, that some data changes, some of them generate data constantly, and some of them are fairly static. And there are few companies that have not changed website for 20 years. And are, you know, not done anything major. So, you've kind of figured out, okay, it's not necessary for me to just go and start looking at them fairly frequently, start reducing the frequency, somewhat like what the Google search algorithm does. Google has to visit your page again and again and again and again, you need to have fresh content again and again and again. So you know, you'll start saying okay, I don't have fresh content, Okay, I'm gonna come a little less frequently and a little less frequently, you know, that it's a known algorithm in the industry. So, you monitor the sources, and then you monitor the changes. And I will get less abstract about it when we take an example and run through the whole thing. And then once you gathered the data, and you've also gathered some amount of time series data in the sense that okay, this is what is happening, you know, at a regular intervals kind of thing, for example,
Dorai Thodla: not necessarily from the same source, but also how the information hops across different sources. Let us say that, you know, we talked earlier about the velocity of propagation, right? We talked about, you know, a TikTok or Clubhouse yesterday, kind of something. Sometimes because of the internet spreads very fast, but some information doesn't spread, it takes a long time to spread, it takes a while. So, what are the slow moving items? And what are the, like fast moving items, which grabs people, that kind of stuff? So, you track the frequency and the velocity, and the volume of content change, for example. So you look at them, and then you start drawing inferences, saying, okay, you know, you start drawing inferences in a wide variety of things. So this process is not like a straight flow, it's kind of recursive, in the sense that, as I'm monitoring, for example, let's say I start reading a blog post of somebody who came up as an expert, and then start writing about artificial intelligence in education. And then you start reading these posts, and then they start doing references to others. And then you start going to those references, and then finding out and finding out that some of these references were old, they had been blogging for 20 years earlier than this person. But they're blogging about an entirely different aspect of education, you know, learning theory, and this and that and that kind of things, styles of learning and mental models, and all that kind of stuff. So you start getting those kinds of information. So you discover some new sources, and you take and add them to the list of your sources. Now you start monitoring those sources too. So it's kind of an interesting process where you go through the steps, and then you come back to the previous step because of something that you gain. That is one way is the references. The other way is gaps. You find this and say, “Hey, there is this big missing gap that nobody's talking about? Why this gap?” And I'll tell you one interesting concept about the ‘missing nothing’ concept, which is favorite of mine, I don't know whether I’ve talked to you about it before.
Pravin Shekar: No, no, you haven't.
Dorai Thodla: So once you start finding gaps, why isn't anybody talking about the ethics in AI? Why is everybody taking whatever the recommendation being given as gospel truth? I mean, why are they taking these predictions and just acting on them? Or are they really taking all these things? If they're not, then what else are they doing? What are they doing with these predictions? And so I remember this almost like 20 years ago, when I demonstrated my product first, saying very excited and saying,” Oh, no, this big. We monitor web pages, I monitor 300 web pages, and every day I get like 1 alert, 30 - 40 changes. I look at it I know what is happening in the industry.” I said, “Great”, I said, “After that what?” And I said “Okay, you got all this done. You’ve finished all that, so after that what are you doing with this kind of thing?” I said, “What after that, what do you need to start thinking about what are you going to do?” And then when you say, after that what gives you a new set of requirements go back. And say, “Hey, let me go find a resources”
Pravin Shekar: I would relate it too. We used a social media company for a couple of conferences for an association that both of us are also a part of. And the report came that there are 3 million eyeballs etcetera, etcetera, etcetera. Well, how many came in to buy the ticket? How many even visited the website? There was no correlation, there's just a whole lot of information that probably affects the vanity but doesn't touch the top line or bottom line. So this is exactly how I would look at that there's so much information that we get stuck in it. And taking a bit from the HG insights blog that you shared. That information has to get converted into actionable intelligence. And if that is not there, then what's the point? And back to you here.
Dorai Thodla: So not all of it gets converted, right? Sometimes you don't even find the kind of thing you're looking for in the information. But in the connections between these informations like for example, you suddenly say that company X, bought from company Y or company X made this company Y, a partner and then you start looking how did company X and company Y ever meet? They are in different industries, they are this thing sales guys, what is going on? Then you suddenly find out they share a board member, you know, there’s some dude sitting on the board of both these companies? And who might have just suggested, “Hey, you know, have you looked at this company that I am on the board of? Maybe, you know, there is some synergy maybe we should.” So the connections are many, and there are semantic connections in the sense that they're not like, “Okay, I linked your website, you linked by website” kind of stuff, not at that level, but much more than saying that, hey, you know, there are these gaps in my product. And you know, Jeffrey Moore, I have to keep bringing you back is the whole product concept. And how do you go from your product to whole product kind of stuff, and then you start finding all the complimentary products around it, that gives customer the experience they want, which is what they're looking for. And you know, selling them just an isolated piece of doing this kind of stuff. And that's the reason like why electric vehicles will not come and become popular unless you have the infrastructure to charge them all over place and, you know those kinds of things are what do you invest in.
Pravin Shekar: I can relate to it, because I invested in an electric scooter, the Ather. I love it, except there's always this fear of “hey what will happen if I run out of charge?” Because there are fewer points, of course, they are working on it. But it still isn't a movement from what you say.
Dorai Thodla: Yeah. So there's two things that I said I talked to you about is the gap. So the concpt of missing nothing is what do you think happened before the concept of zero was invented?
Pravin Shekar: Nothing.
Dorai Thodla: Nothing. But everybody was missing something. Imagine writing all the numbers you can write in Roman numerals. So it's inadequate, it was good up to a certain point. But beyond that, you can’t. So somebody came and said, “Hey, there is something between negative and positive, there has to be some little thing called zero.”, you know, and I think the concept is zero is much more than that. But you know, zero, and the moment he said zero, it fell, everything fell into place. And then the positional notation, I mean, everything became you know, so that is one missing nothing, and other missing nothing and always get confused between this Uranus discovery of a planet, right, Pluto. And nobody saw it. But somebody felt the gravitational pull. And some dude came and said, “You know, this is happening consistently in this orbit. So there has to be something in this space, let's go look for it.” And then looked for it, and then found it. I mean I’m getting my planets mixed up. But you know, that's a missing nothing. It's a missing nothing.
Pravin Shekar: I have a segue to add here, Dorai. When my daughter was younger, and they were curious, she came and asked, “Is there another world? I don't believe there is another world” etc. And I said, she was familiar with using iPad and computer so, “I don't want to file what I do? I delete it. Where does it go? It goes to the bin. Okay, I removed from the bin. Where does it go? I don't know.” So maybe there is, we don't know. But like the data, you never know the missing nothing concept. Lovely.
Dorai Thodla: Yeah. Okay. So when you delete from the bin, it becomes energy. Matter gets converted into energy. Yeah, yeah. Right. So I think these are, these are interesting gaps. I think these gaps, give you some ideas. And the gaps come about when you start using a product, but let's get back to the flow business. The discovery is, so how do you find and this goes back to our influencers talk earlier. So if I want to, you know, do higher end, you know, like education, and everybody knows, education is going to change in some way. Maybe it's a very slow pace, you know, colleges and schools will still be there, because there is a need for them. Because people are going crazy sitting at home. So there will be need for these buildings, that will be the social spaces kind of thing. But their function may change for example. Because think about a teacher, having 20-40 students sitting in a class, half of them sleeping, or maybe 75% of them sleeping, and two or three people playing games, and only some three people hardly trying to listen in front row, can be very disheartening for somebody who's delivering a lecture.
Dorai Thodla: On Zoom, it’s “I don't really care.” Because there is a static picture that somehow makes me feel as if there is a student present there only and they are also muted. So only, sometimes you need to say, “Hey Pravin, Pravin.”, two three times that person wakes up and says, “Yeah, yeah I'm there.”
Pravin Shekar: My mic was on mute.
Dorai Thodla: So, maybe, you know, they've figured out everything you can get from the internet. So teacher is more like a guide, you know, somebody who can answer all the questions for self-learners and that kind of stuff. So, maybe it will shift, it will change kind of thing. But why do we get clues for this change? So you look at all the issues of remote learning, look at all engagement metrics, and there's a whole bunch of learning analytics, and then you start getting all these kinds of things, and then you start. So why do you gather the data? So firstly, you start with some influencers and say what they say about the future of education. So that's always a nice Google search to do anything you want to know the future of x. The future of x is y is the popular thing is that what is going to be the future of education. And then there are a bunch of futurists who are sitting and thinking about, you know, what will happen if, you know, if you move to Mars, for example, you can't export all these teachers, what happens to those kids, their parents are to be teachers kind of thing. But that's too far away. But let us look at a near future. When you start thinking about this, you come up with these ideas that are applicable to even today's problems, right. You know, what is the biggest problem in remote education right now? Other than the infrastructure and all that sort of stuff? The inequity, right? I mean, the current system of education requires you to have a device that is connected to the Internet, and you are supposed to have the certain amount of speech. And I had classes where two of the students would not send any exercises saying, “Sir, we don't have anything, we are listening to you through mobile devices. I can't type the code and send it from there because no, I can't test it here.” So what do you do with them, and they don't have laptops at home. So that gives a new meaning to the infrastructure as to what students need to have, if you shift to this mode of education norm. So there is this problem of haves and have nots? So you start identifying all these kind and then you also start identifying, this system is not perfect, because it has got all these gaps and these gaps, or opportunities, or gaps, or, you know, our barriers to entry are, you know, challenges for you to go into the next step. So the monitoring is, in fact, there is one diagram in which I saw is gather data, monitor, monitor, monitor, monitor, monitor, that's the next 10 steps of the process. Oh, you know, and that is true with the traffic, you know, like, satellite, you know, basically GPS monitoring devices, why are there clusters of traffic here, and all that kind of stuff, and all. So that's one thing. I think the large amounts of data that you get is, there are many ways in which the data can be handled. One is that data can be segmented, you know, all data is not segmented. So each model you create looks at one slice of the data, for example. So if I take in a smart city, I look at the data from all the corners, I am definitely going to be inundated, right? But if I look at, hey, the traffic data, the consumption, energy consumption data, and they're kind of dependent but the dependency is not as crucial, as you know, like, what we only know that a couple of traffic lights go out, and this place is going to be a mess in terms of traffic.
Pravin Shekar: I get it but clearly Dorai, and I'm just trying to equate it to business here again, from HG Insights that from all the technology information that is available, how marketing and sales teams are doing territory planning? How they are figuring out whom to contact? One is of course, the market sizing that is there, however good or bad, they use the data to go ahead and figure out and once the territory is done, through the entire process, they are able to figure out what is the potential budget and what is the spend likely to be and therefore, whom should I go ahead with. Now, when we relate back to the gaps and the processes that you're talking about? And also in terms of the information hops across sources, and especially in technology intelligence, there are several such parts that you have to connect the dots. How do we connect the dots?
Dorai Thodla: The thing is this right, like, when do you know when a technology is mature and has arrived like so let us take the innovators, early adopters, late adopters, and then we'll forget about majority and the laggards later, right? So let’s take one example. What do these innovators read? Where do they hang out? What do these early majority read? Where do they hangout? Where are these, like, you know, the late majority, right? Or the massive larger number of users? So if you take that, so these guys in the tech world, for example, the early majority are Hacker News, you know, Techmeme and a whole bunch of these things, right? In the current generation, of course, gets it from Instagram and all that kind of stuff. The next one will be getting a lot of information from referential information from their associations like skip, we talked about, right? So say they have you, what is your trade association? What is in your industry kind of thing? So they in addition to getting information, but they tend to read, let's say Business V, and, you know, Inc magazine, or Fortune, or some of those types and then you go to the next level, there they read New York Times, or Washington Post or blog. So if you see a product being covered in one of these little magazines, then hopping into a Business Week or Time kind of thing, and hopping into Yahoo News, Microsoft News, Google News, that kind of thing, and then hopping into New York Times, right? If there's a column about New York Times, it gets a very, very different, this is in the USA.
Pravin Shekar: Dorai, I love it that all our conversations end up coming back to this influencer in various forms here this case the influencers in innovator, early adopter, but I love it that it keeps going back to that. But please continue.
Dorai Thodla: Yeah, there are various types of influencers, right? You use the time, you know, micro influencers, which is awesome, I think micro influencers so you I'm likely to listen to my neighbor, or my closest contact friends or inner circle, for certain advice that I can trust. However much I like some hero, I'm not going to listen to take, you know, Shahrukh Khan's recommendation of what car I should buy next is not relevant to me. I can't afford. So yes, appearing in an advertisement of some car is “Meh, okay, fine.” You know, like, he got paid for it. You know, that car I can't afford, for example. But if you come, actually, you're also a little too above my pay grade kind of thing, but let's say that, so you relate, right? And I say, okay, he is making the decision. And he's very similar to me, in personality kind of stuff. And then so micro influencers tend to be very closely attached. So in that hub, draw a network diagram, you see these little hubs and little spokes, you guys are clustered fairly quickly. All of us live in Adyar or in Chennai, or in this industry or all members in TiE. So there are all these connection spots where we are all peers in TiE Chennai, right? So for example, we listen to each other and have conversations because we are all doing the same thing to some extent, not different businesses, but in the same communities helping same types of people. So yeah, but at some level previously, before internet, you know, who the gatekeepers were? The editors of the magazines in the journals.
Pravin Shekar: Yes, yeah.
Dorai Thodla: Just to get in front of them was like, you know, VCs, imagine VCs, yo, there's no way you can, you can tweet them, and, you know, no way, you know, Twitter has changed that, you know, internet has changed that. internet changed access, web 2.0 has changed the interaction model, you know, you can follow anybody on LinkedIn, you can send in a request to follow, you can send a message, you know, and, you know, if it is nicely worded interesting, and something relevant to that what that person is doing. So, which also means that you need to study them. So let us take the technology intelligence to a slightly different level, right. There's so much of this, like there's a people in this, right. So if we both share the same passion in a technology, you know, we were more likely to be talking, have opinions about the direction of the technology or the uses of the technology and then we'll get into conversations and these are Twitter threads now, right? Or they’re Reddit, subreddit conversations that are taking too. So in the kind of magazines that I talked about earlier, I'd missed out some very influential magazines like Wired. MIT has an emerging tech magazine, and then you go into a little more professional engineers, they're all like the IEEE magazines and those kinds of things. And you know, it's kind of academic plus industry kind of stuff. And so that's where people will look for information. So you have to be visible there in all those processes. And they all provide you as to what they publish, what are the topics, conferences, in your space. Conferences, always, they do a lot of research on what are the most popular topics, to attract people to attend the conferences, to find these speakers. They've already done all that work. So you can just go there and say, Hey, who are the speakers in the conference? What is the bios of the speaker? Which companies do they work for? Which are the companies they are associated with? You know, where is their presence in the web? You know, some of them may be on Twitter and LinkedIn, and some of them may not be, but it doesn't matter. Academicians tend not to be on any of the social media, but they are, they hang out in an entirely different space.
Pravin Shekar: Beautiful. So a whole bunch here. We did miss out speaking about artifacts, but we'll come to it in the next part. One source leading to the next the info hops across sources. What are the gaps? The missing nothing. Inside in between the gaps, what do they share? How do they share? Connecting it back to our innovators, influencers as a part of the process? So, Dorai, let's continue our conversation tomorrow, where I'm pretty certain you're going to be mixing process, innovators, the missing nothing and some tools.
Dorai Thodla: Yeah, so I'll just give you one key word of artifact, ‘Knowledge Graph’. And then chew on it. It is enough for you to go and search. Ask me many questions.
Pravin Shekar: Lovely, Knowledge Graph. Well, here we come tomorrow with another conversation between Dorai and Pravin. Martech, Martech, Martech. Thank you, Dorai.
Dorai Thodla: Thanks, Pravin. It's fun. Yeah.