Recon Analytics Acquires Atom Insights adding Device Data dimension

Full Transcript

Don Kellogg 0m10s

Hello, and welcome to the two hundred and eighty seventh episode of The Week with Roger, a conversation analysts about all things telecom, media, and technology by Recon Analytics. I'm Don Kellogg, and with me as always is Roger Entner. Roger, how are doing?

Roger Entner 0m23s

I am excellent. I'm so excited. This is, on one hand, the two hundred and eighty seventh episode, on the other one, it's a first. I'm really, really excited to let our listeners know that we have acquired ADAM Insights, and with it, Pani Spatia and his team, and we're expanding into the device area. So this represents for us the first acquisition, it represents for us an expansion into Canada, and Anish has a whole team in India as well that is supporting him.

Roger Entner 0m56s

And so, Anish and I were talking for a long time, and the logic of us putting these things together is just undeniable. When Don and I started building our pulse service, one of the things that we put in was a plug for somebody like Hanish to come in and fill that and connect it and make it bigger. As our listeners know, and I'm sure you're sick and tired of it, you know, we're now serving over 25,000 people every week. And in our mobile, home Internet, English, Spanish, business, AI, and airline surveys. And we collect here passively the device hardware ID of everyone who takes this survey, and 80% of our respondents do that on mobile.

Roger Entner 1m44s

So we know what device they have, and they don't have to tell us. And so we get, like, device information down to the SKU level. For, like, Samsung devices, we have four, five, six different SKUs of the same phone. The customer may know he has a or she has a Samsung 25 Ultra. We can tell what international SKU it has.

Roger Entner 2m6s

Then we have here over 22 NPS factors on it, for everything from the screen to sturdiness to you name it, every which way dimension. And each of the 22 dimensions is mapped to a component in the device. We can tell you how a device with a certain baseband chip behaves differently on a different carrier network. And not all devices behave the same on an AT&T, Verizon, T Mobile, Comcast charter network. It is mind blowing.

Roger Entner 2m43s

But the missing piece was the device shipment, was the device component? Going even deeper in components than we could. So, I'm so excited to have Hanish with us and joining us, and making this richer than I could have even imagined when we built this capability. And Hanish brings to this his unique expertise, and it allows us to do things that nobody else can do. But with this, Hanish, I'm so excited to have you with us.

Roger Entner 3m14s

Take it away.

Guest Speaker 3m15s

Thank you so much, Roger. This is very exciting for me as well. Been doing this for almost twelve plus years now. Even while being on the device side, I was always referring to your podcast to learn more on the telecom side. So here we are doing this together now from being an avid listener of your podcast to now being on your podcast.

Guest Speaker 3m35s

So it's a very exciting day for me as well. Yeah, I mean, think you have briefed everything pretty well but just talking about statistics, we'll now be able to track the shipment side, the sales side, just giving a broader view of overall market. But overall, I mean, now we are expanding our capability across all the countries, regions. I have been tracking more than 18 countries at model level. So just to give the context of a scale of the data that we'll be producing across the globe, 1.1, 1,200,000,000 devices sold each year.

Guest Speaker 4m5s

So we track that at model level. So all that capability along with all the analytics that you provide on the telecom side, it will all combine together to track the whole journey of the customer right from when and where the customer is buying something to how the experience the consumer is having on the device and the network to what the customer is going to do next, the next consumer journey also. So, I think we'll now be able to offer the whole package right from the consumer experience from buying and then the experience over the years on the device with the device with the operator. So together, I think this is the most comprehensive analytics anybody could get in the market right now.

Roger Entner 4m46s

Yeah, and as you pointed out, with your data set, we are also laying the foundation to go international. Our survey, we already do it in English and Spanish. We have it in German, we have it in French. This allows us to go basically on demand to Canada, The UK, France, Germany, Spain. We have the capability to do that.

Roger Entner 5m13s

We can launch any of these markets. And so our niche capabilities and customer base gives us this push pull part of where do we wanna go next. New verticals, new countries, where do we wanna go. We will extend this further out. We're having speed test data very soon integrated into our platform, So we can tie not only satisfaction, we can tie speeds by network to this.

Roger Entner 5m42s

Our AI arena capabilities are massive because using AI, we can now already tie together our proprietary datasets and the questions that our clients ask every single week. Where are we, Don? Now 150 databases? We're adding so many databases.

Don Kellogg 6m0s

As of last night, it was 149. Almost. I'm sure we'll breach one fifty soon.

Roger Entner 6m6s

By the time this airs, we will have more than a 150 databases.

Don Kellogg 6m9s

Probably. Yep.

Roger Entner 6m10s

Wanna go through some of the things that we've been working on that side?

Don Kellogg 6m14s

Yeah. Sure. So, I mean, like, we're wrapping our heads around, you know, there's a lot of really good government databases out there, but a lot of times they're large and they're hard to work with and they're difficult to normalize, etcetera. And so what we've been doing is really kind of collecting all this data, curating it, and then creating kind of a a layer that then helps the AI with discovery. And so for example, you know, you could look at something like, how does an increase in inflation impact willingness to switch among wireless customers.

Don Kellogg 6m46s

Right? And so we can use the macroeconomic indices that are produced by, you know, BEA or BLS, and then we can, like, match that back to our data. Right? We also have all of the FCC data loaded in, and the FCC is a really good example of really valuable data, but that is hard to work with because it's so massive. So for example, at its most granular, the FCC broadband coverage data is over 450,000,000 records per release, and it's released every six months.

Don Kellogg 7m17s

We're going back now three time periods. Right? So we can see how that data is changing over time. The other thing that we've done that goes back to some of the work that I did earlier in my career around geospatial analysis is we've created lookup tables and kind of geographic lookups that allow us to translate between different data types. Right?

Don Kellogg 7m36s

So you might have one database that's looking at something at ZIP code or at census block level, and you might have another database that's looking at MSA or BTA or RSA. We've built up all the component blocks to rebuild all those things with their subcomponents. Right? And so, for example, we could look at the FCC data by MSA or by DMA or, you know, any other definition you can draw on a map. We've built those crosswalk files for that.

Don Kellogg 8m1s

And so that's one of the fundamental problems that a lot of times you'll come across when you're trying to mix and match datasets is you may have different geographic definitions. We've solved that problem as well. And it all, you know, is orchestrated via AI, and it's auditable. Every analysis produces an audit trail. So it's you know, you're not worried about throwing a bunch of data AI and having it hallucinate.

Don Kellogg 8m22s

We have the receipts on all this stuff, and it's easily outputable alongside the final analysis that this generates. So it's pretty exciting. I can only imagine, and I'm I'm kinda salivating even thinking about this, but adding the device data in on top of that will be really the icing on the cake. It's already such a rich data source, but there's so much optimization, I think, on the carrier side around devices that still remains to be done. So we're looking forward to helping facilitate that.

Roger Entner 8m49s

Yeah. And we're gonna take Kanish datasets, putting them into our data repository, making it all AI queryable. And we have a three tier AI offer now. One is outside the firewall,

Don Kellogg 9m5s

and we

Roger Entner 9m5s

have our first customer with that, that integrates with all of this. We have one inside the firewall, and then we have a standalone, much more richer version that we're working on. We're investing north of $2,500,000 this year on the reconnaissance platform alone. And the data sets that we have, and with Hanish coming in, will add international databases, we'll include the ones that he has, and this is all a comprehensive package. It's insanely powerful.

Roger Entner 9m36s

I feel like I'm living in the future. We have shown this, we've shown it to quite a few of our carrier clients, and one feedback from one provider was like, Dawn and Roger, you're showing us here things live that are better than any of the canned demos we have seen from other providers, because we live this, we breathe this. I was talking with Srini Kopalan a couple of weeks ago, and I told him that Recon Analytics' third largest expenditure after salary, after healthcare, is AI. AI tokens, AI subscriptions. It's insane.

Roger Entner 10m12s

We're so deep into this. And Hanish, with his data, makes this all so much more powerful. I'm so excited to show the data that we have to our clients, show Hanish to our clients, show what we do already to Hanish clients, which will be quite a thing.

Guest Speaker 10m31s

I would like to add this, Roger. This is something very unique that other end market analyst agencies, tend to forget. Because the device teams, they sort of work in silos and the analysts who are working on the data, sometimes they are unaware about the macroeconomic elements that are going to impact the market. The best example of this is if you look at the previous forecast provided by agencies last year, they were estimating the market to grow by 1% to 2% going through twenty twenty eight, twenty nine, thirty. But now, all of a sudden, we are hearing about the memory crisis, the prices of memory going up.

Guest Speaker 11m4s

And now, all of a sudden, all the agencies, they have revised their forecast to negative 10% for 2026. So this is how important is just taking into account the larger scheme of things, the macroeconomics, what is happening across industry verticals to understand the impact what is the impact it is going to make on the smartphone side. So it's something very unique which I saw here and I think it's going to add a great deal of value for the clients using the end market analysis data.

Roger Entner 11m35s

Absolutely. And one of the things that we're going to help Hanish with is to bring our trademark speed, Fidelity impact to it. Fidelity here is already this data beats anybody else's in the market. But we will be able to do this faster than anybody else with the same fidelity. We have done this with customer insights here.

Roger Entner 11m56s

We will do it with devices. And this results in greater impact, where it benefits also our other clients. We're moving at a weekly pulse in our surveys. This type of granularity, there are signals that are going both ways. We can signal from The United States what we have already.

Roger Entner 12m15s

And when we launched the other international markets, we have more a wider, wider net. We see here demand signals. We see the signals from Hanish data. The result is just like synergetic. You have one or the other.

Roger Entner 12m31s

If you put the two together, it's gonna be even more powerful. You know, I know a lot of our friends in the industry are listening. I don't wanna give away everything we do. We're showing it to our clients. But here's the thing, this year, '26, is a banner year.

Roger Entner 12m48s

We have a lot of really cool things in stock where we're gonna make this even richer and expand even further. So, I'm so excited.

Guest Speaker 12m58s

One more item, Roger, if I may add. One more interesting aspect of being on the device side is you also get to work with a lot of supply side, chipset companies, semiconductor companies. And the main question or the most worrying question they always ask us is about the demand of particular devices. Now, with your NPS, we already know how consumers are rating a particular device, let's say, based on the chipset, based on the screen size, based on all the different hardware elements of a particular device. And we also know how the device is doing in the market or what is the consumer sentiment or the overall consumer sentiment around the usage of that device.

Guest Speaker 13m36s

So this is going to be the most important aspect of our research, especially for the clients on the semiconductor side because now they are able to get a live gauge of how things are on the ground, sometimes on a weekly basis, and how those consumers are experiencing that device as well as the network on that device on a regular basis. It's a very amazing dataset for the semiconductor side of companies.

Roger Entner 13m59s

It comes together, and their clients, or their clients' clients, have the data already available. It's like, wouldn't you want to know what your client knows, right, about the performance of your chipset that you don't see? Also, there's a great synergy is with our huge sample sizes. We're pretty close on installed base. With Hanish data, we get the buy in, and what gets actually sold at the carriers.

Roger Entner 14m28s

And so this makes it so exciting. You know, we can put all the things together, measurement, attitude, more things to come, right?

Guest Speaker 14m36s

Yep, absolutely.

Roger Entner 14m38s

Excellent. I also want to say thank you. I have been mean to Total Wireless, and Total Wireless have been really good people. You know, their reaction was, to me being mean to them, was like to send Don and me a very thoughtful gift package. And I'm actually talking on the headset that they sent.

Roger Entner 14m58s

So here's the note that Total Wireless sent us. You know, dear Roger, we listen to your podcast and we thank you for your inspiration to do better. Here's a brand new headset for you to use when you record your podcast. Well, here I'm using it.

Don Kellogg 15m11s

I am too.

Roger Entner 15m12s

Brought to you by the fastest growing no contract carrier in The US based on annual distribution growth. Sincerely, friends at Total Wireless. Thank you, friends at Total Wireless. Really appreciate it. Your class act.

Roger Entner 15m25s

Thank you.

Don Kellogg 15m26s

Absolutely. Alright, gentlemen. Roger, we'll talk next week.

Roger Entner 15m31s

Appreciate it. Bye bye.

Guest Speaker 15m32s

Bye. Thank you.