All things IoT

Speaker 1: This is ClientSide from Fox Agency.

Nathan Anibaba: Sylwia Kechiche is the Principal IoT Analyst at GSMA Intelligence. She leads IoT product development and is also responsible for sizing the IoT opportunity, with a focus on enterprises, including the role of operators, to researching and analyzing IoT technologies, applications, and business models. The GSMA represents the interests of mobile operators worldwide, uniting more than 750 operators with almost 400 companies in the broader mobile ecosystem. The GSMA also provides industry leading events such as MWC Barcelona and Shanghai, and Mobile 360 series of conferences. Sylwia Kechiche, welcome to ClientSide.

Sylwia Kechiche: Thanks, Nathan, for having me.

Nathan Anibaba: Absolute pleasure having you on the show. Let’s start with your background before we get into the domain expertise. You have a sociology and communications policy academic background. How does that influence shape the way that you think about technology, human beings, and the work that you’re doing today?

Sylwia Kechiche: That’s a good question, Nathan. When I think about myself, I’m a bit different to other analysts because I don’t have the mathematics, statistics, very much numerical background when it comes to being an analyst. On the other hand, what I’m really interested in is a bit of doing investigation. So the way I look at analysis and doing research is more being a bit of a detective, finding a few data points here and there, and then building on the story. And by talking to people, you actually get a better understanding of what’s going on, you get to test your assumptions, and understand what is happening around us. And I think the sociology degree did help me because that’s where you actually learn how you do surveys, how you talk to people, you do one-to-one interviews, and so on. And communication policy studies, that was more around regulation and what changes there are. So interestingly enough, I found that this degree, this background helped me in my work.

Nathan Anibaba: Really interesting. Let’s talk a little bit about the work that you’re doing now, and especially let’s start the conversation at IoT, because I think there’s still a lot of confusion when it comes to IoT. There are lots of different use cases and applications. Explain what IoT really is. Why do people deploy it? And why should we care?

Sylwia Kechiche: That’s actually one of my pet topics. And when you think about IoT, obviously we have our definition that we’ll put across as GSMA Intelligent, GSMA, and the internet of things, so IoT, describes the coordination of machines, devices, and applications connect to the internet. And this could be connected by multiple different networks, this could be wifi, and other [inaudible], ZigBee, Z-Wave, satellite become more popular, cellular networks, and so on. And so multiple different ways of connectivity. But those connected devices include everyday objects and machines from across very different verticals, industries, and so on. And when you think about it, you can think about consumer IoT, so consumer electronics, devices such as wearables, connected cars, all sorts of sensors that are equipped with the technology to send and receive data. So when you think in your home, probably you have Alexa, Google Home, so smart connected speaker. This is IoT. Smart watches and so on, that’s IoT.
And when you think about actually knowing when the next bus or train arrives or knowing how long it’s going to take for the Uber to arrive, that’s also IoT and that’s enabled by telematics. But that’s putting in the consumer perspective on the enterprise side. It’s all about data. It’s not only about connecting devices, because connecting as the read the first step is about what’s happening after you get the data. And then data, some call it the new currency. That’s what’s really important. And it’s all about enabling digital transformation. So what we have heard through COVID-19 pandemic, the term digital transformation really came out as one of the top catch all phrases. So it’s all about making sure that enterprises that deploy IoT can achieve cost saving, increase the revenue, generate new revenues, but also comply regulation.
So on the enterprise side, IoT is all really about application focused on all the control management of supply chains, assets, and fleets, and so on. So that’s why should we care, because it does help. On the enterprise side, it helps with achieving the business goal and business objective, on the consumer side, it really helps us with making our life easier. And when you think about the future, potentially in a couple of years, couple of decades, we can think about just entering in home and everything happening. As soon as you arrive home, your kettle will know you want a tea, your TV will come on, the lights will come on, that’s the vision. But obviously for that-

Nathan Anibaba: That’s a beautiful future.

Sylwia Kechiche: That’s a beautiful vision. It does take time, it does require a lot of collaboration between different ecosystem players, but that’s the vision.

Nathan Anibaba: Really exciting. And so from a business point of view, the business outcomes that enterprises are looking at right now are, as you said, cost savings, new sources of revenue, and also compliance with regulation. Talk about what some of the most attractive opportunities are from a business perspective for enterprises. And how are the best enterprises deploying IoT at the moment?

Sylwia Kechiche: So it really depends. So we are very lucky that we run enterprise in focused service. So we surveyed, last year, 2,900 enterprises across 18 markets and verticals. So what we find is we cannot ignore the COVID-19 pandemic, that’s all around us, so what we have found out is that the short-term plans were put on hold, but enterprises are still deploying IoT. IoT, it is important for them. And what do they see is that proof of concept and trials, the percentage has increased year on year, so compared to 2019, but enterprises really want to solve the problem. So it’s all about solving problem. 49% of enterprises see IoT as transformation to the company as well as industry. It is a slight decrease over 2019 simply because there is more recognition that IoT, it is difficult. It’s not very easy to deploy simply because as you scale, you actually realize there’s more issues and more problems.
So the current use cases that we see primarily center around cost saving and primarily are about making sure that you can achieve certain benefit. On top of that, we see that revenue generation is actually falling very swiftly, and we’ve seen that, I think last year. 16% of enterprises said that’s how they measure the success of IT deployment. And compliance year on year, we see a growing importance of compliance. So when you think about compliance, it is regulatory compliance because every single different industry has regulations related to privacy, security, we can’t ignore that issues, but also compliance with COVID-19 measures is becoming ever more so important because you want to ensure that the workers’ safety is ensured, to put it like that. And we see that the use cases do vary. So it’s not only about cost saving, it is about generating revenue as well.

Nathan Anibaba: Really interesting. And so from what you’re seeing, what are some of the biggest risk factors or impediments that will stop enterprises or businesses from realizing those cost saving benefits or even new revenue stream opportunities? I imagine that there are several stakeholders involved in implementing IoT from procurement IT, security, marketing, everyone within the enterprise will have a say. What are some of the reasons why businesses struggle to implement IoT?

Sylwia Kechiche: So IoT ecosystems, so ecosystem of providers, vendors, is extremely fragmented. So it’s very difficult for enterprise to actually understand who they should work with, how to deploy and so on. But again, going back to the Enterprise in Focus survey, we find for the past three years, the top three challenges is integration with existing technologies, security and privacy, and cost. So these are the top three challenges, and maybe shift up and down a bit, but that is the key. When it comes to integration, the reason why it’s important is because you already might have existing assets out there. So either information technology, IT assets, or operational technology, OT assets, and you have to make sure that when you bring IoT into the picture, they all work together. So what we found, those that say that the struggled IT integration, 69% find IT integration to be the key challenge and those in retail and utilities find it ever more challenging, while manufacturing companies struggle more with OT integration.
This could be explained by them having to change the business processes and the way they work pretty much overnight during the COVID-19 pandemic, and changing the supply chain, so on. But something really important and I don’t think that’s talked about is that internal, as IoT obstacle has grown, has been important. So in 2020, 40% of enterprises told us that as the key challenge, and that’s up from 26% in 2018. So you can see there’s a strong growth trend. And that actually reflects the fact that education on the benefits of IoT needs to extend beyond the C-level. So you don’t need to only convince people that are in charge of money to deploy IoT, but you also need to explain to people that going to be working with the new processes, they’re going to probably working … Well, let’s think about a connected factor amongst the robots, that this is benefiting them.
And it is all about change management. So you need to make sure that there’s going to be the change management process put in place to get the buy-in from people that actually are on the ground and operate those services. Otherwise, you won’t be able to implement a successful project. And when it comes to that, and we talk about privacy, security, cost, continually one of the key issues, but it is also important to remember that IoT, even though it’s been around for quite a while, it’s still relatively new when you thinks about enterprise thinking, and 48% of enterprises say that their IT technology departments are driving the investment in IoT solutions. Even though it’s decrease and the deployments are moving away from being intolerant of IT, this poses a challenge, because as I said early on, IoT needs to bridge this very different culture of IT and OT and very different languages that those departments are talking.
And what we see quite often, I mean increasingly, we see actually departments being put together that the sole job is to make sure that they translate between the different departments, and what they have to translate is not only different needs of line of business but also what requirements on the security side there are because they are very different. So when you think about it, there are lots of challenges, lots of obstacles, but it is also important to ensure that there’s a trust between the provider and the supplier, and also trust between the business owners and employees.

Nathan Anibaba: As you look at the landscape now, what sort of businesses, what sort of industries are benefiting most from IoT right now? And which ones, in your opinion, do you feel would benefit but for whatever reason currently aren’t?

Sylwia Kechiche: So when you think about M2M and IoT typically looking at the enterprise, utilities have been deploying M2M for a very long time, and very simply, smart metering, making sure you don’t send people out there just to read the meter. It is important. So this has been going on for a long time. We’ve seen COVID-19 pandemic put a little bit of a pause on that because engineers couldn’t go out there and replace the meters, and so on, but this has picked up. So utilities, smart metering definitely is one of the key industries that’s been embracing IoT for very long time. When you look across other sectors, retail. So retail, if you think about point of sell, so every single time you pay with your credit card utilizing point of sale. So retail already has an adoption of IoT. And this actually has increased because contactless payment is a thing right now. You barely can pay with cash.

Nathan Anibaba: Sure. In some places in London, you can’t at all.

Sylwia Kechiche: Exactly. But also you think about the way their warehouses are working, the way the manufacturing and the supply chain is operating, these were the industry verticals that had to adopt the so-called new normal, whether or not we like this term, it’s the new normal. They had to pretty much ensure that they know what’s happening on the entirety of the supply chain. So whether or not the products are coming, how long they are, when you think about vaccines, are they being stored at the right temperature? So everything to do with asset tracking, asset monitoring, workers monitoring for their own health and safety, that will happen. So health and safety is ever so more important, but we also have seen when you think about smart buildings, the overall adoption of IT would have decreased because people don’t go to offices and so on, but again, the face of it is changing simply because you see thermal cameras just ensuring that people don’t have a high temperature when they enter and so on.
You see more healthcare use cases. So hospitals had to make sure they know where the assets are, because when you don’t know where your bed is and you need to put somebody on the hospital bed, you need to find it very, very quickly. So there was a very, very fast adoption across a range of verticals, even smart cities, they had to actually reallocate the budgets that had, let’s say to connect streetlights to, let’s say, implement more [inaudible] and so on. So there was definitely a change in the way IoT has been deployed and it has stopped for a little bit. It was a bit of a pause, but overall, as I think I wrote a while back in short-term disruption, longterm business benefit, it is because enterprises across the world realize that they need to transform. There’s a need for them to be connected and agile, and be flexible in order to respond to very sudden changes.

Nathan Anibaba: Well, Let’s talk about something else that you wrote recently. One of your posts, you write the overall size of the IoT revenue opportunity in 2025 will be $900 billion, a 2.6 time increase on the 2020 figure. That has been revised a little bit due to COVID-19 and the impact of the pandemic. But how do you come up with that number, first and foremost, and what factors go into those calculations?
Sylwia Kechiche: So the new figure is 906 billion. That’s down from 1.1 trillion in 2025.

Nathan Anibaba: Amazing.

Sylwia Kechiche: So we have reduced by 200 billion. So it is rather a tactics … It’s an interesting approach because first, we need to understand how COVID-19 … we had to understand how COVID-19 pandemic has affected IoT adoption levels. So we look at it across households and businesses. So whether or not adoption of IoT devices across households will increase, decrease, that’s number one. What’s happening on the enterprise size? Adjust the forecast based on the potential impact of COVID depending on the vertical geography and so on. And that depends on the length of the lockdown, fiscal stimulus packages in place, and so on. And we also look at what IMF was forecasting in terms of the outlook of the depth of the economic crisis, and they have revised the outlook a few times as well. So that doesn’t make our job very easy because nobody knows. At the end of the day, it is very uncertain market. There’s massive uncertainty about what’s going to happen when things going to go back to normal, if they’re ever going to come back to normal.
But again, we had to come up with a baseline. What we realize is whenever there are government measures in place, their impact is lesser. So let’s say in America, Western Europe, Asia Pacific, that had help from the government, they did not see as much effect. I mean, obviously there was decrease, but it was very, very different. And the SME sector, so the small and medium enterprises, particularly susceptible to all of those changes. So if there is no help to SMEs the most likely going to go bankrupt, do not survive. So that was something that we had to look at as well, but we assume that IoT’s enterprises will recover after initial contraction. And what we see is obviously whatever impact COVID-19 has had or will have on the connection side is going to be reflected in the revenue generated. So less connections, less revenue, and so on. But that’s not the really important factor to think about is that quite a lot of new contracts or the contracts being renegotiated, canceled, and so on, they have to be very attractive when it comes to price.
So we’ve seen vendors offering all sorts of prolonged trials, pay over a longer period of time, and so on. So, we’ve seen in general contraction of the revenue markets simply because we see that the competition is very, very tough and enterprises have less money, so you have to ensure that you are priced competitively. When it comes to revenue, the biggest decrease we’ve seen on the professional services side. So what we see is consulting. So new projects were affected the most, therefore the consulting revenue that deals with sizing those new projects and so on, and then system integration and so on, these are the revenue that decrease the most, but it is rather a long process trying to figure out what’s the impact of COVID. And some of the revenue will never be recovered because some of the business lines will have to be combined into one.

Nathan Anibaba: Let’s talk a little bit about the enterprise and IoT specifically, because I know you spend a lot of your time working with enterprise businesses. When looking at the adoption of IoT across the enterprise, you mentioned industries such as retail and manufacturing earlier, what does success look like? And what are the main metrics that they’re using or other businesses should be using to measure the success of an IoT deployment?

Sylwia Kechiche: We do ask question in our enterprise survey about how do you measure success by IT? And it is measured in terms of cost-saving, revenue generation, and compliance, which we already discussed. But compliance is something that keeps on increasing as a success metric from 31% in 2018 to 52% in 2020. So that’s quite a fast growth, and that’s mostly driven by compliance requirements with data security and privacy regulation. And revenue generation, which I found quite surprising, but then later on it was not, became a top metric in 2020. 68% of enterprises measure it as such, just ahead of cost savings, 65%. So companies are under the pressure to grow the top line. So obviously cost saving is something that naturally they do, but they need to think about how else they’re making money. But I love a good use case. It’s one thing that enterprise server tells us, but what do actual enterprises telling us?
So this year at [CES], Caterpillar, which is a mining company, mining equipment manufacturing, it’s debut, which typically you would see all sorts of consumer electronics, device manufacturers, because it used to be a consumer convention only, and what did they say is that thanks to their autonomous equipment and the benefits it delivers, it delivers 30% increase in productivity, 20% cost reduction, and what is very, very important, thanks to their equipment there is 0 incident. So very much helps with safety. And another example that we have when you think about manufacturing, you mentioned manufacturing one of the industries, is [Accent’s Panda Nanjing] factory, which is one of the largest industrial factors involved in the manufacturing with 3D product, one day deployed IoT to automate production. Again, it resulted in savings from increased efficiency, reduction in tenants cost, increased flexibility, which is very important in product line design. And they managed to get a 50% return on investment during the first year and they plan to break even in less than two years
And that could go on and go on because I love those things samples, but I was listening to a panel with a gentleman from [Dow], which is a chemical corporation manufacturing plastic, he said for him is really important the safety, so not on compliance, but in general, ensuring safety. So just in time prediction, making sure the assets and people are safe, and that’s anything to do as in process safely, so vibration monitoring, leak detection, this all contributes to safety. And we don’t often think about it in that terms, that IoT helps with that. And he said that over the past 20 years, there was 100% increase in safety performance and there are 0.1 incidents per 10,000 hours.

Nathan Anibaba: Amazing.

Sylwia Kechiche: So you cannot measure the actual benefit of making sure that your employees are safe and secure in what they do. And that’s something that, from measuring the success of IoT, it’s very hard to translate into monetary or very hard to quantify, right?

Nathan Anibaba: It is. So I hear what you’re saying as far as top line performance. I mean every enterprise is under pressure to deliver more top line performance, but who tends to champion IoT within the enterprise because does that sit with IT? Is it security that tend to champion it? I mean there are many other competing priorities that executives at the C-suite have M&A, bottom line growth, COVID-19, go down the list. Where does IoT sit in their list of priorities and who tends to champion it within the enterprise?

Sylwia Kechiche: It really depends on the enterprise size and what’s the goal enterprise has in mind. And what we have found out is when we ask about the reason for deploying IoT, 63% of enterprises have told us that they haven’t deployed IoT as part of digital transformation. So that’s increased from 60% last year. So if an enterprise has a goal to transform themselves to become more digital, they will deploy it as part of that. So IoT, AI, machine learning, cloud, and so on. There’s vast of different technologies and very different ways to ensure that you’re moving up on this journey of digital transformation. And when it comes to larger enterprises, that’s quite often driven by the headquarters. So again, I believe that 60% of enterprises, that’s headquarters driven, but quite often you might have a local champion. So you might have a smaller outpost, so you have a smaller company that wants to change one particular problem, because at the end of the day, IoT is all about solving a problem.
You don’t deploy IT just for the technology because you want to have, let’s say, 2G, 3G connected devices. You have a problem and you want to solve it. So there’s really interesting way that we now started talking not only about proof of concept. So what is the concept? It’s proof of value. What’s the value you’re going to get if you deploy such and such technology. So it very much varies. So IT departments are still in charge because typically that’s where all the budgets sat, and we see that IoT budgets are under a little bit of strain, and dedicated IoT budgets, that’s another matter. So we’ve seen that either enterprises have a dedicated IoT budget or they’ll have to repurpose budgets from other sides of the business. And we’ve seen that their purposing has actually increased.
So you have to achieve cost saving in other parts of the business to be able to sponsor or to invest into IoT. But that very much depends. So, as you mentioned, R&D departments, they’ll be in charge of IoT if they want to create something new, but in general, it is about changing the way business operates. It’s not only about just getting a benefit in terms of connecting devices because that’s not the end goal. The end goal is getting the data and then changing the way business operates. And once you change one factor of it, more data you get, you can change more outputs and you get cleverer in terms of what you can do.

Nathan Anibaba: The data point is an interesting one because making sense of your data is I guess what all enterprises have to do in order to achieve the business outcomes that they’re looking for from IoT. But a lot of their data is siloed, it’s lost, it’s uncategorized, or it’s just not collected. How can you get a business or an enterprise to access their data, use their data in the right way in order to get the results that they’re looking for from IoT?

Sylwia Kechiche: That’s a very interesting question, because what we see is that data is really what drives the outcomes. And the question about data is that a very small proportion of data that’s collected is actually being analyzed and being used. And obviously it is still stuck inside of IT, OT silos. Everything to do with ingesting data, you have to ensure that you have the right platform. So if we go to the IoT stack very quickly, you have your devices, then you have to connect them. You connect them using multiple different types of technologies, as I mentioned, [inaudible], solar, whatever other technology. So you have those devices, you connect them, and then all those devices generate data. So what do you do with them? What do you do with that data? You have to translate that data into common language to ingest it and then to make sense out of it, using either very basic analysis about what has happened in the past. So step one, in terms of what enterprises do, they want to understand what if condition A happens? What happens as a result?
So just very brief outline of what’s happening with the data. And then you go a step further, is okay, so if this and this happens, what happens next? So you go through the whole entire variance, what do you do with your data? So you can then do prescriptive analytics, you can do predictive analytics, you can use AI, machine learning to learn what your data tells you, but with very little data, you can do that. But then when you think about it, if you want to add data from different sides of your business, obviously you have to translate it into one language, and then you need a platform. You need to have a platform that helps you do that, and apply all sorts of machine analytics and so on. What we increasingly see, and that’s a very interesting trend is that sending data to cloud, not everybody wants to do that. And quite a lot of decisions have to be made on premise on the edge, so on the device. So you don’t want to send your data to your cloud because it takes time, the journey backwards and forward.
So we see a very strong growth in edge computing, and that’s part of 5G. So we see lots of activities around edge computing, AWS, Azure, Google, the partnering with operators and so on, trying to find out the new use cases for mobile edge computing and 5G. But going back to data, you have data that obviously can be stored, can be analyzed, can be ingested, but then the real value, the business benefit that comes, and that’s the ultimate Nirvana when it comes to data, is when you are able to actually combine data from multiple different sources. So when you think about it, you have your organization data, you have anonymized data, because that’s important as well, from organizations doing very similar things to yours so you can compare and contrast how you perform against those. But you can get third-party data.
So imagine John Deere is the agriculture company. So they have connected tractors and also they’ve connected massive, massive machines, and what they do, they actually help farmers to do several things. A, it’s autonomous. Some of those are tractors autonomous so you don’t have to actually worry about driving it, but also precision. So because you have the data, you can map the field that you own, you’re mapping it, and then you’re able to very precisely drive, and you’re not wasting any of the grains that you have to plant because you know exactly how you’re going, you know the route, and so on, and so on.
And you’re going to do it at exactly right time. So it goes in my head, the whole logic behind it, but if you plant at a certain time, you get a higher yield and higher crop productivity, and so on, and so on. So that’s data coming in. You are informed by data. And then once you have access to all the data, weather data, you know not to do it a certain time because let’s say it’s going to be too hot or it’s going to be too wet, and so on. So that’s when you bring in the third party data. And that’s what really, really helps. When you combine all the different data together, you can really get smarter about not only your own business but how it should perform compared to the best in class.

Nathan Anibaba: So Sylwia, can you tell us a little bit more about GSMA and GSMA Intelligence? How does that fit into the bigger picture?

Sylwia Kechiche: And GSMA, as you explain, is a trade association and amongst other things that we do, amongst the conferences, we also have programs. So there is program focusing on IoT, and in particular focusing on 5G IoT for manufacturing. That’s a very hot topic at the moment. Focusing on drones, automotive and also IoT security, which is a very key topic. Another program is future networks, looking at how the networks going to look at the futures, including 5G, and identity. When it comes to GSMA Intelligence, we are the analyst arm of the GSMA. So we’re providing data analysis and forecast, mostly on mobile operators and networks, and we do quite a lot of research.
So our team authors multiple reports across different topics, and one of the best known are mobile economy reports. So those series actually come out in conjunction with Mobile Congress Barcelona, Shanghai, and so on. So the most recent one is the Mobile Economy China, and you can download it. So I would very much encourage everyone to read it. So obviously looking at how was the status mobile economy. That’s very important for us. We look at other topics, obviously 5G is key, spectrum, what’s happening on the spectrum side, what’s happening on the fixed side, digital consumer, and last but not least, IoT and enterprise. So this is what GSMA Intelligence does. So we will have a team that focuses on forecasting, so data, team of researchers, team of economists, and in general, we do quite a lot.

Nathan Anibaba: Really, really fascinating. And Sylwia, I can speak to you about this all day. We’re going to have to get you back on the show because there are a million questions that we didn’t get to ask you, but I’m going to ask a final question now about your history and background, and specifically about leadership and diversity within technology, because you are a Polish female with a sociology background really on the cutting edge of technology for the enterprise, as you’ve just explained. You don’t see many people with your background doing what you’re doing. How do you reflect on your own career and background? And talk little bit about what you’d like to see as far as opportunities for women in technology is concerned.

Sylwia Kechiche: That’s a very, very interesting question. So typically analyst tech is very much male dominated. Well, I shouldn’t say dominate, it’s not the right word to use. It’s a male industry simply because of background engineering degrees and so on, and being on a couple of different panels, one of the key challenges is women are less likely to be educated in maths and so on, and so the STEM topics and so on, and that’s one of the big barriers. So girls are being told it’s okay that you’re not very good at maths because you don’t need it, and so on. So I think this is all going back to the way you were brought up, that you should be better at languages, you should be better at this and that. So unfortunately this goes back to very much the role models and the way male versus female are being portrayed. But I think yes, definitely, being Polish, living in London, being educated the UK did help. And I think being female also helps with having a little bit different perspective when it comes to things.
So it’s a really hard question. Being bit more intuitive, so thinking about things from a different perspective and bringing this point of view. What I would like to see more, I would like to see obviously more females in the field, but this is not going to change overnight. We see that across C-level, the change is happening, but it’s very, very slow. It’s going to take decades. When you think about when women were allowed to vote, that wasn’t such a long time ago. So everything takes time and we have to be, unfortunately, very patient. But I would say having role models and having somebody that you aspire to be always helps.

Nathan Anibaba: Well said. Sylwia, thank you so much for doing this.

Sylwia Kechiche: Thank you, Nathan.

Nathan Anibaba: If you’d like to share any comments on this episode or any episode of ClientSide, then find us online at fox.agency. If you’d like to appear as a guest on the show, please email Chloe at fox.agency. The people that make this show possible are Chloe Murry, our booker/researcher, David Clare is our head of content, Ben Fox is our executive producer, I’m Nathan Anibaba. You’ve been listening to ClientSide from Fox Agency.

Speaker 1: Join us next time on ClientSide, brought to you by Fox Agency.