IoT Insider Podcast | Episode 11: Why IoT projects fail

Hello, everybody. Welcome to the IoT Insider podcast, today, I’m joined by Pangea, MD, Dan Cunliffe, and Operations Director, Chris Romeika. And we’re all here today to talk about why many IoT projects fail and why some of the predictions made in the industry have kind of fallen short in terms of the numbers.

Dan Cunliffe
Chris, I guess we’ll start with a history lesson for some of the people talking about where things go, I’ll kind of run to this one. But I learned recently that apparently IoT traces routes all the way back to 1843, where a certain Sir Charles Wheatstone attached copper wires to balloon and successfully captured and transmitted temperature data from high in the sky, so that meteorologists could predict the weather, pretty interesting. It’s come a really long way from there hasn’t it mate?

Chris Romeika
It definitely has, you know, and obviously, the internet didn’t exist in 1843. The theory was there, you know, getting data from remote sites and bringing it back to somewhere in this case, the grounds where it could all be processed.

Dan Cunliffe
But, you know, how does IoT measure up to the predictions made by the analysts years ago? I think that’s probably most thing we’re going to talk about today.

Chris Romeika
Yeah, I mean, we’ve seen a lot of things been, you know, sent around on the media, billions and billions of devices. So for example, Ericsson and Cisco said, 50 billion IoT devices would be connected by 2020. Big numbers, you know, more accurate sort of estimates now put that closer to 9 billion. So you know, a lot less devices than a lot of people thought there would be. Also, a lot of IoT projects are failing a bit early, according to Microsoft, 30% of them are not getting past the proof of concept stage.

Dan Cunliffe
And I think that’s what we’re trying to discuss really is like, what has caused some of that? I mean, look, from my perspective, I still think 9 billion is not too bad. It’s better than, you know, a much lower number. And actually, it shows that there is something going on, you know, one of the questions is sort of why have some predictions, been over estimated? You know, why do certain issues arise? So we’ll try and we’ll try and sort of tackle that. And I know you did a blog recently, you know, talking about sort of how or the reasons why some things might not have gone to plan. You maybe want to kind of reference some things there?

Chris Romeika
Yeah. So a recent blog post, I did, reasons why IoT projects might sort of fail or stumble, or fall short. Uncertainty, the time taken to run projects, deployments, sort of forgetting that the IoT connectivity part, right till the end, bit of stuff about skills, perhaps security as well, integration problems and challenges and budget as well, or maybe going back to, you know, things people are more comfortable with. So yeah, there’s a lot of there’s a lot of reasons out there. So maybe we can dive into some of those in a bit more detail.

Dan Cunliffe
Yeah, well, maybe let’s let’s look at something like uncertainty, maybe time taken to run projects. I mean, I know from my conversations with all of our partner channels, it’s taken time you and I started this business five years ago, literally, trying to take an IoT ecosystem play to the to the UK partner channel five years forward. Well over 200 partners now, and many of them are actually on the journey, but in the beginning, it was pretty tough, right? I mean, people weren’t really up for it? Yeah, yeah. I think it was a lot of what does it even mean, you know, is it M2M, is it IoT? I remember those blog posts and conversations we used to have about what the differences are. And I think that sometimes people are more driven by the what is the wind right now? Or what is the thing that I can see in the market physically, rather than thinking well, IoT is actually a lot about creating return on investment? I think that’s a big part. And maybe that uncertainty isn’t always picked up.

Chris Romeika
Yeah, I think a good example. I mean, yeah, looking back into those those five years, you know, a lot of time we or maybe you more had to just explain IoT, fundamentally, you know, to what, you know, what it is, you know, to see level people in decision makers in all the companies we spoke to, a lot of the times using some examples, I think a classic one you’d use was like the wine bottle being half full and reporting for another glass and stuff like that. I don’t know, if you still use it now or not. But

Dan Cunliffe
I probably always have used that one, in every job I’ve done.

Chris Romeika
I think, you know, you can also kind of pair up uncertainty a bit with budget integration as well. So for example, once again, you know, going back four or five years, you know, a lot of the people we spoke to that might have been familiar with mobile, you know, so sort of mobile consumer type SIM cards, you know, and we get questions like, you know, okay, well, you know, it’s got, it’s got unlimited SMSs, why can’t I have that now? Or unlimited data? Or, you know, it’s going to roam? Isn’t that going to be expensive? So, you know, it takes time to to sort of educate people and give them the right information about this kind of stuff. But yeah, you know, those those type of things. You know, I think that’s a combination of uncertainty and also wanting needing to take a leap as well.

Dan Cunliffe
Yeah, I think there’s a, one of the points you mentioned earlier is about, actually the IoT skills gap. I don’t think we’ve, I mean, personally, I probably haven’t seen enough roles being advertised for that market. And I think that’s probably one of the reasons why things haven’t moved as fast as possible. Look, in our business, we’ve grown sometimes triple digits, maybe double digits every year for the last five years, right. I guess from our perspective, we’re pretty pretty, pretty buoyant about the market. And that’s because we’ve really been focused on not just one particular way of delivering your service, we’ve been quite broad, very bespoke, sometimes very non bespoke, and making things happen quickly for the client. So we really, you know, we we really put a lot of effort in, but the skills gap could be evident, because, to be honest, I suppose you and I, and the team, we kind of learn our skills over the years.

Chris Romeika
Yeah, that’s right. We pick things up as we go along.

Dan Cunliffe
We haven’t just turned up and known everything.

Chris Romeika
That’s right. And also, in this changing environment, like over though, if we look at over the last five years, you know, in the beginning with, with the SIM cards, we were offering most of them were 3G. And at that time, the providers and the operators were saying like, you know, you don’t need 4G, because we’re defining IoT as low bandwidth, you know, sensor data, a couple of packets or bytes a day. That’s all you need, which, you know, and that has changed a lot. And now, you know, you look at any IoT device, and it’s all about, you know, it might be about low data, but it might also be about, you know, streaming, high definition video, like, you know, a CCTV, no one would argue that’s not an IoT device anymore. Yeah, well, that needs huge amounts of data. So you know, 3G to 4G, 5G, now, also our new Narrowband technologies like NB IoT and LTE category M, and BLE, you know, stuff like that. Yeah, you know, there’s there’s always constant change, just like any environment, but right now, in the IoT space. Yeah, it’s, I think it’s a lot faster than a lot of other similar areas.

Dan Cunliffe
I’d be, really interested to know, the stat between, of course, you know, Gartner probably went for the biggest one, which was 50 billion devices connected, I think that was the biggest stat that came out in terms of volume. We’re now at 9 billion, I remember seeing, I can’t remember the publication, but someone else was more around the 25/28 billion. So it’s maybe horses for courses, in terms of what they chose. But I would also say, or would like to know, what was the predicted average revenue per device, versus then versus now, because to your point, the revenue generated from a 3G device is nowhere near the revenue generated from a 4G device. Its definitely very, very different. So it’ll be quite interesting to see how that fits up. So maybe the market isn’t that bad, actually, and the revenues are there, but it’s just less devices, you know, these are the kinds of things that we go and again, I’ll bring it back to our business, the way we trade with our partners and provide an ability to connect devices, right, from sort of kilobytes to very low megabytes, you know, single digit megabytes right through to single digit terabytes. Yeah, is what we do you know, largest stuff we do is well over two terabytes a month for some people. There is definitely something to be said about the revenues per device in the prediction. Yeah. So we’ve covered the skills gap. I think probably one of the other ones is a lot of the projects run and take longer to deliver than probably expected. What did you kind of take away from that?

Chris Romeika
Yeah, I probably see that well especially in you know, the area we operate in the channel yeah there’s there’s just a long you know, usually there’s a long type of process from from start to finish when you actually have a SIM cards life so you know, starting out it’s the negotiation phase. You know, we may sit on that for a while and I’m sure a lot of people listening in who are in the channel as well may may go Oh, yes, it is definitely how it is for me negotiate, negotiate, and then after like, you know, weeks of nothing, you might might hear something. The customer saying I need it, I need it tomorrow. Yeah, you know, I need the SIM cards tomorrow or the devices tomorrow, you know, which is nothing new to us, you know, we’ve we’ve been in this space for a long time now. But then yeah, you know, it’s there’s always sort of that that sort of sprint, then sit down and wait, the waiting part is you know that there’s isn’t that decision making the integration as well. So you’re probably so you know, there may be a lot of times when the actual requirements may not have been totally met, for example, they might have thought they didn’t need a public static IP on the SIM, but now they need public static IP. So they have to change all that around. And you know, and change the orders and stuff like that. Or maybe something like the end customer has suddenly told them they need CDR feeds they can’t operate without CDR feeds.

Dan Cunliffe
They need to build the product to actually employ someone. I agree with you, I think it’s kind of touching one of the points that we wanted to cover, which was that deployments within IoT, often leave the intelligence around the connectivity and how I’m actually going to do the internet part in the IoT to last, unfortunately, or kind of throw it away as a that’s easy to do. I don’t think it’s easy to do. And I’ll stand by that. And I’m not trying to be kind of because we provide a business that does that kind of thing. But it isn’t easy to do, because quite often people will forget that, you know, newer technologies have to be rolled out before you can access them. Yeah, things like LoRaWAN have to be deployed, you know, things like Sigfox have to be, the device has to be ready for it. And there’s a lot of these things that people kind of forget, when they come right to the last bit and go, Okay, how are we going to deliver the final piece to it. So we quite welcome people to bring their IoT projects to us early, you know, we have always run a business, that focus on the connectivity, the device and the application. Because we can help to steer or guide or even just support really your IoT deployment, in terms of the successes.

Chris Romeika
Yeah, and we’re set up in a way as well, you know, we’re pretty friendly. I think, you know, when it comes to partners, who have someone who wants to, you know, work on a new project, you know, I’ve seen many, many times, that’s, you know, one of the key things people aren’t sure about, especially when you’re developing an IoT solution from a, you know, from an application perspective, is just how much data the solution needs. Sometimes a lot of partners come along and say, Oh, you know, we want we want IoT, we want connectivity. But then, you know, the next question from us, our side usually is we usually it’s like, you know, how big is the the opportunity? How many SIMs do you need? but also how much data do you need? Where do you need it to connect? Yeah, so yeah, and that’s when it’s, you know, so we’re quite happy to help out with, you know, figuring out helping the customer figure out how much data they need, but like you were saying, like, you know, and then we go into things like, where do you need it to connect? And then, you know, it gets even deeper, like, you know, do you want to be fully resilient? Or do you just need basic connectivity? Like, in other words, how many networks do you need? Do you need it in more countries and stuff like that? Do you need to send SMS messages? voice? So you know, a lot of the time it’s the person thinking, they just need a single thing I need, I need IoT. I need connectivity. But, you know, going back into it, you know, the data, the countries, the add on services, like SMS, like, do you need that? Have they thought that out? That’s when things start to get complicated. And things slow down?

Dan Cunliffe
Yeah, no, I agree with you moving a bit onto kind of like, you know, where are some of the successes, kind of what the key takeaways are? Or, you know, what sort of opportunities can IoT bring? And maybe also, let’s touch a little bit on 5G, because it’s not that far away. I think I think some of the key takeaways are, the market is growing. And you wouldn’t get to 9 billion devices, according to the where we are now, if it wasn’t growing. Yeah. End of story, so it is growing. Is it growing at the rate that Gartner said? No, because they said 50 billion, but actually, maybe the revenues generated are good enough. And when we started five years ago, there were less companies doing it. And now there’s lots of companies doing it. Yeah, not like us as our partners. And we need to be able to work on showing other people how they go forward. So for me, for me, some of the key takeaways are, you know, keep looking, keep being available, be broaden in the way that you’re thinking about IoT. It’s not it’s it’s not a dead pan market. In fact, it is growing rapidly. And what you need to think about is things like what are the devices that need to be used in IoT, because that’s kind of where you begin. Sometimes, if I know the devices that are going into projects, I can kind of understand where my piece will play. And maybe your first step as one of our partners is just to connect the device. The next step could be work with us to help procure the device and there could be margin in that. The final step could be work with us to try and help the application layer. Various things that can be going there’s different ways you can do that. So you can begin where you feel most comfortable. If you’re just working into it. Just recently, we’ve Well, I was going to just make an anecdote about something that I was actually in the gym the other day and when I’d finished just going to go and collect my bag and stuff from the bathroom. There was an interview on the radio in the bathroom and talking about how young drivers are seen as the most susceptible to creating a claim, which is a bit unfortunate, for some young drivers are actually pretty good, better than maybe some normal drivers. However, for the first time I’d ever heard, there was an interview style conversation about telematics making it better for young drivers to get better premiums, because the telematics device can help sort of profile them to improve the cost that they have to pay. And that to me was like mainstream IoT is right there in front of you. And telematics is one of the earliest adopters, that market is probably the earliest besides, you know, Sir Charles Wheatstone from 1843, but it’s probably one of the earliest adopters in IoT market, and to only start to really hear about that now, it’s pretty interesting and shows you where we’re on the journey, there’s no doubt in my mind that everything will need to be connected in order to sustain the way we try and live as a human race it just has to be that way. What do you just just sort of closing off and thinking and then adding to that, what you think 5G is going to try and bring to the, to the mix?

Chris Romeika
Oh, I think it’s going to bring a lot of speed, obviously. So it’s going to be leveraging on what we’re already starting to get from 4G, and things like, you know, high speed 4G, for example, LTE, category six, you know, hundreds of megabits per second. So there’s going to be speed, which is going to enable things such as 8K video, and that, so security cameras will be using a lot more data. 5G will be there to support that. And also, you know, the 5G capability to, you know, have sort of ultra reliable communication channels, and, you know, sort of guaranteed delivery of messages. Yeah, there’s, there’s going to be a lot of a lot of new opportunities in there. I think there’s also going to be, there’s going to be a bit of a, you know, there’s obviously going to be a rollout phase as well, for sure, as 5G takes a lot longer to roll out, because it has to have a lot more cell sites than 4G or 3G, it has to be a lot more closely spaced together. But yeah, you know, I think I think the actual migration path also is going to be quite interesting. We’ll be seeing a lot of 5G ready devices, you know, that currently support 4G, but will just sort of seamlessly switch over to 5G when the time, you know, when it’s available. Yeah. I think that’s gonna be great. Yeah, looking forward to that.

Dan Cunliffe
Yeah, I think we’ll probably bring it to a close. We don’t want to keep going too much more. But you know, from our perspective, we know that the market is definitely growing. We see it every day with partners placing orders every day, looking to grow and do interesting things. The market isn’t just in the UK, it’s a global market. And I think that’s pretty interesting to sort of keep thinking about it is very, very possible to trade. IoT globally. We do it so and our partners do it as well. I think on that note, you know, we’d like to close off here and just thank everybody for tuning in. If you have any questions or you pick this up and you’re not a partner do contact us through the website, Pangea-group.net We also have LinkedIn which is Pangea Connected. And we also have our Twitter which is @Pangea Connected as well. So do join in, if you want to know more. Chris thanks so much appreciate you joining us. Please do read Chris’s latest blog for his guide to successful IoT projects. And as I said, get in touch if you want to know any further, I’m Dan Cunliffe, signing off.