Episode 41

Killer Combos & Finding the Fit - Johan Stokking - TTI

Johan Stokking, co-founder of The Things Network and The Things Industries and CTO of The Things Stack, joins the show to talk about why LoRaWAN works best when it’s combined intelligently with other wireless technologies rather than treated as a standalone answer to every problem.

The conversation starts with why The Things Conference deliberately expanded beyond LoRaWAN, and what Johan is seeing as LoRaWAN matures. He explains why developers now understand both what LoRaWAN is good at and where its limits are, and why the real momentum comes from combining LoRaWAN with cellular, Wi-Fi, Bluetooth, and other radios to solve practical deployment problems.

Johan walks through his “niche of a niche of a niche” fridge monitoring example, using cold chain as a way to explain where LoRaWAN fits exceptionally well and why these highly specific use cases can still represent multi-billion-dollar markets.

The discussion digs into real bottlenecks like battery life, basement connectivity, lack of Wi-Fi credentials, and compliance requirements that make LoRaWAN the right tool in the right context.

The episode also explores what’s coming next at the silicon and modem level, including multi-radio devices and why cloud platforms will need to manage multiple connectivity options seamlessly.

Johan shares how network metadata and design data can be used to optimize deployments, improve battery life, and drive real ROI, and where data itself may become more valuable over time.

The conversation wraps with what Johan is most excited about next, including the next Things Conference and upcoming improvements in the LoRaWAN ecosystem focused on better interoperability and plug-and-play deployments.

Johan's LinkedIn

The Things Industries

  • Helium Global IoT Coverage - Want to know if Helium coverage exists where you need it? Check out this map!
  • Helium Foundation - The Helium Foundation's IoT Working Group (IOTWG) has generously provided support for the first 6 months of shows, please go check them out and consider using the Helium LoRaWAN as a primary or backup on your next deployment. With over a quarter million gateways deployed worldwide, it's likely that you have and can use Helium coverage.
  • Support The Show - If you'd like to support the MetSci Show financially, here's where you can donate on a one-time or an ongoing basis. Thank you!
  • MetSci Show - If you'd like to use our IoT or AI Data Value calculators, or you'd like to contact me, the MetSci Show site is the best way to do it.
  • MeteoScientific Console - Use LoRaWAN - The MeteoScientific Console allows you to use LoRaWAN today. As long as you have Helium coverage (and you probably do, about 90% of populated areas in the world have a gateway within 2 miles), you can onboard a sensor. You can always check coverage at https://explorer.helium.com and switch to the "IoT" tab in the top right.

Transcript
Speaker:

Today's guest on Meteo

2

:

Scientific's The Business of LoRaWAN,

and is Johan Stokking, co-founder

3

:

of the Things Network and The Things

industry's and the CTO behind

4

:

a lot of the real world infrastructure

that helps LoRaWAN deployments scale.

5

:

In this conversation, Johan breaks down

why The Thing's conference expanded

6

:

beyond LoRaWAN

into a broader community of communities

7

:

and what that says about where

the wireless IoT market is headed.

8

:

We also dig into his niche of a niche

of a niche fridge monitoring example

9

:

why LoRaWAN fits cold chain so well,

and what it reveals

10

:

about the real bottlenecks

in IoT adoption.

11

:

And he walks through the 0 to 1 versus

one 2 million path.

12

:

Why open source transparency and developer

first product choices still matter

13

:

when customers are trying to scale from a

proof of concept into serious production.

14

:

This episode is sponsored

15

:

by the Helium Foundation and is dedicated

to spreading knowledge about LoRaWAN.

16

:

Normally, at this point, I would recommend

you try the globally available LoRaWAN

17

:

run by Helium,

which you can find at metsci.show/console.

18

:

But since this is about TTN,

I think you should go to the Things

19

:

Network.org

and try their's instead for this show.

20

:

Now let's dig into the conversation

with Johan Stokking.

21

:

Johan, welcome to the show.

22

:

Thanks.

Come on man. Thanks. Thanks for having me.

23

:

I'm psyched to have you.

24

:

Obviously, you're a big part

of what's going on in LoRaWAN.

25

:

You just had this conference in September.

26

:

I think 1900 people, almost 2000

people came.

27

:

Yeah. You got to meet and greet

and talk to a lot of them.

28

:

What's going on, dog? What's latest?

29

:

This year we decided to expand

30

:

the scope of our conference,

not just be a conference about LoRaWAN.

31

:

The reason why we did

that was that it's is getting harder

32

:

and harder, actually,

to fill two days with just LoRaWAN content

33

:

because everything has been said

and right there.

34

:

So there's so much stuff already,

35

:

and we are now really in a phase

where the developers

36

:

know what it is and also know

what it's not good for, right?

37

:

And the solutions are being built.

38

:

This whole ecosystem is growing.

39

:

And the second reason is really that

we now have also the confidence really

40

:

to expand the scope

and to invite other wireless

41

:

technologies to the conference

without creating a lot of confusion.

42

:

Right.

43

:

That's also important.

44

:

And third is that we see that

lots of our users and customers,

45

:

but also a partners and people we talk to,

46

:

combine LoRaWAN

with other wireless IoT technologies.

47

:

Sure.

48

:

And so this could be device makers

that have variants of their devices

49

:

that are using cellular or Z-Wave,

or where LoRaWAN and similar

50

:

are complementary because the gateway

has set up a black hole.

51

:

And so yeah, it makes it makes a lot of

sense to have this ecosystem in one place.

52

:

Really? Yeah. Yep.

53

:

And you guys

54

:

and I don't know

if you done this on purpose or not, but

55

:

it was never kind of the LoRaWAN network.

56

:

It was always the things network.

57

:

And so this this allows you to expand into

a ton of different connectivity options.

58

:

Yeah that's true.

59

:

Yeah. Let's see.

60

:

So you talked about this idea

61

:

of a niche of a niche of a niche

with these 8 million commercial fridges.

62

:

And they're not really worth

that much each.

63

:

Walk me through what that means

for someone maybe coming into

64

:

or when who maybe doesn't understand.

65

:

I know a lot of us do, and a lot of us

been around for a while, but someone who's

66

:

just learning about this,

they found this podcast.

67

:

You're like, okay, what are some of the

problems or bottlenecks or issues here?

68

:

So the story we try to tell with

the fridge

69

:

is that we have users and customers

who found it interesting that, right,

70

:

they found a use case which works

71

:

where LoRaWAN is really a good fit.

72

:

And for example fridges. Right.

73

:

Cold chain in general could be fridges

you know in restaurants or towns.

74

:

But it can also be a farmer.

75

:

And so when we talk about numbers

and you know we have for example

76

:

we say that we have this many million

and devices on a network.

77

:

People are like, oh, but it's the tens

of billions that Gartner was expecting.

78

:

You know, ten years ago, right.

79

:

Yeah I promised yeah, yeah, yeah, yeah.

80

:

And so where are they?

81

:

And then oh come back,

you know, we have this, we go, you know

82

:

in this niche of an issue for niche

where we say, okay, this is this is IoT.

83

:

And within IoT

you have constrained devices.

84

:

You know, they have to work on a battery

in a cold environment

85

:

for a few years,

ideally five years or something like that.

86

:

And you need to have compliance reports.

87

:

And so there's a use case

of putting sensors in fridges

88

:

and also those sensors

because of the battery constraint.

89

:

But also the connectivity

there's the difference is in a basement

90

:

you can't connect it to cellular.

91

:

You can't you don't have Wi-Fi

92

:

credentials of the restaurant where you're

deploying this things like that.

93

:

And so LoRaWAN is a really good fit.

94

:

And so we we always use this use case

or a similar use cases to explain

95

:

how hard it is

to to build a solution like this,

96

:

but also to show the potential

because it's, it's

97

:

there are so many fridges

around the world. Right.

98

:

And there is there's very little reason

to monitor them.

99

:

And so we have,

you know, customers doing 200,000 devices

100

:

and they are focusing on on cold chain.

101

:

And the main gist of the story

102

:

is the addressable market

for these types of use cases is huge.

103

:

And the investment is also very big.

104

:

The total cost of ownership

of a device is,

105

:

it, you know, it

runs, it runs into billions.

106

:

The investment

that the LoRaWAN community as a whole did,

107

:

as far as putting up

108

:

gateways and sensors and just making sure

the whole thing is, is working.

109

:

Let's see.

110

:

You've had this focus on open source,

which I think is super cool.

111

:

I'm not a very mature nerd.

112

:

I've been doing this

for like 4 or 5 years.

113

:

And at first when I heard about open

source, like, oh, it's silly.

114

:

Like people just rip your idea off

and you've had a very different

115

:

take on it and said, like, look,

this is one of the best ways to do this,

116

:

and you've got this idea of 0 to 1

and then 1 to 1 million

117

:

for someone

that may not have heard that before.

118

:

Can you walk us through

what you're thinking is on that

119

:

and some of the challenges going from

for 0 to 1 and then 1 to 1 million?

120

:

Yeah.

121

:

So 0 to 1 is where people try to build

their first working solution.

122

:

And 1 to 1 million is to scale too. Right.

123

:

And we always try to to address

both audiences with the same product.

124

:

And that's always the challenging bit.

125

:

But we managed to do that.

126

:

And open source really is very useful

for that.

127

:

0 to 1 where you have a developer,

he has a laptop

128

:

and a gateway and a device,

and they want to make it work.

129

:

And in a controlled environment,

they don't want to spend money.

130

:

They want to know exactly what's going on.

131

:

So they don't want to sign up

for a black box

132

:

and add it into their credit card

information.

133

:

So that's one reason to do open source.

134

:

Another one is to be very transparent

with how the technology works.

135

:

And our audience are developers. Right.

136

:

So we we mainly our users are developers.

137

:

So they are as technical as we are

just in a different kind of technology.

138

:

They are more on the enterprise side

or network operators.

139

:

And we also think that developers

140

:

increasingly decide which products

are being used in their company.

141

:

And that goes from, you know, databases

to, you know, hardware and other things.

142

:

It's that if the developer really thinks

this is a good product,

143

:

then there is a good chance

that it will be adopted.

144

:

And so we invest a lot in that

and to make it freely available.

145

:

So we have two things network

a community network which is also free,

146

:

but it's hosted.

147

:

So it's more of a black box.

148

:

But at least you have to set up a lot

less.

149

:

You don't have to run things that locally.

150

:

And then we have the same stack,

open source, which is also free.

151

:

But then you have this full control

so you can choose.

152

:

But we have this path forward to

if you're scaling from one to your

153

:

from your working doc

to a million that you can basically use

154

:

the same products, the things that cloud

which is managed or two things

155

:

like enterprise that you can run on prem

or in private cloud.

156

:

Okay, it's super cool to see.

157

:

I mean, some of the stuff

158

:

just from my perspective,

on the open source side,

159

:

it's already complicated

enough, like people are going to come

160

:

and rip it off because they got to learn

how to use it first.

161

:

And it's just such a pain

in the butt that you're like,

162

:

I just want to get this thing working.

163

:

And it was super cool

to read about your approach

164

:

to that and say, like, hey,

this is how we get people in the door

165

:

is we give them the choice to either

do it themselves or try it out,

166

:

and then let them build from there

and rely on the fact

167

:

that we're going to run this

pretty awesome thing, and we're going

168

:

make it a lot easier

for them to do the thing they want to do.

169

:

Yeah. Let's see.

170

:

So this conference that you guys just put

on, you said it was the biggest one ever.

171

:

Open it up outside of LoRaWAN.

172

:

What are you hearing

kind of what's the the word on the street.

173

:

What are people doing in this world. Yeah.

174

:

So it was a big success.

175

:

There was a lot of new insights

for everyone.

176

:

I think joining people

who came with a Bluetooth

177

:

or a Wi-Fi or a Z-Wave background,

they got to learn about LoRaWAN.

178

:

They got to run into people

that they knew already.

179

:

You know, from the IoT industry,

but they didn't know that they

180

:

were also working in other IoT tech.

181

:

You know, it was very,

very good to to connect and to reconnect.

182

:

People.

183

:

And like you say, I think where we are

called to things, network

184

:

things, industries

and we are not exclusive with LoRaWAN,

185

:

we released a, a gateway to Things

into Gateway Pro last year.

186

:

We see that more than 80, 85%,

something like that uses

187

:

the cellular backhaul exclusively.

188

:

And so we

189

:

really see that there is that cellular

and LoRaWAN are very complementary.

190

:

And yeah, what we were going to do

191

:

in the same stack, we're undecided yet,

but we have some ideas

192

:

also because we see that also on this

on the radio

193

:

and the silicon and the modem,

you know, the module level things

194

:

start to converge in the sense

that there is not

195

:

necessarily one IoT technology

that's going to be dominating

196

:

LoRaWAN really has its niche,

and it also has its limitations.

197

:

Sure.

198

:

But we see that there are

is going to be multi radio silicon.

199

:

And that is very very interesting.

200

:

And that will naturally

I think result in supports

201

:

that users and customers of that silicon

expect from a cloud platform

202

:

that if there if are devices support

you know they have long range and LoRaWAN

203

:

for instance that they're able to manage,

you know both connectivity options.

204

:

Yeah. At the end of the day

it's very customer focused piece.

205

:

Instead of saying like, hey,

this is just the one technology we'll use

206

:

and we're going to be the best at it

or whatever the messages is like, hey,

207

:

the problem you're trying to solve

208

:

is to get the data

from where it is to where you want it,

209

:

and Lora wins a big part of that.

210

:

But it's certainly not the only part

and a put up enough.

211

:

So for sure. Yeah. Challenge. Yeah.

212

:

Yeah indeed. Yeah.

213

:

So you know, we strategically decided

to go for operational excellence,

214

:

you know to be the best.

And in terms of availability.

215

:

Yeah.

216

:

And price for the markets and also product

leadership to have the best

217

:

LoRaWAN service stack.

218

:

But it doesn't mean that we don't believe

in any other IoT technologies.

219

:

Right.

220

:

We still have continuously listened

to our users and customers.

221

:

And yeah, and that's I think that the conference is really just the start of that.

222

:

Was there

223

:

anything either a use case or just an idea

224

:

that you heard that was surprising to you

at that conference?

225

:

Yeah, I'm always surprised

by what people use our platform for.

226

:

I don't have specific instances.

227

:

I, I actually have to admit.

228

:

Also, I forget about those examples

and use cases for like 600 conversations.

229

:

I'm like,

just tell me the best way to like, do it.

230

:

I had so many.

231

:

Yeah, I know, it's really endless.

232

:

And I have a three year old.

233

:

And he was also at the conference

that takes up here.

234

:

We have this.

235

:

Yeah, yeah. No not not really.

236

:

But we have this wall of fame and there's,

you know, this this wall of IoT devices.

237

:

And he was looking at that

and he saw this cow.

238

:

Right. Something to cuddle with

and some device maker.

239

:

Use that as an example to, to show

that there is this,

240

:

this cow tracker GPS tracker for cattle.

241

:

So he ran immediately to the cow

and was like, oh wow.

242

:

This is.

243

:

And then, you know, he he also understands

a lot better now what we're doing.

244

:

That's that's always very inspiring.

245

:

And it really ranges from cattle tracking

to, to also very boring things like,

246

:

you know, water metering and

and other utility stuff, just sending data

247

:

once in a while, but also sometimes

really societal impact use cases,

248

:

you know, very small things

from crowdsourced network

249

:

to measure the air quality

during New Years, for instance, where,

250

:

you know, people have lots of fireworks

and other things

251

:

and they make a map for that which

and they use the things network for that.

252

:

That's really great.

253

:

And then, yeah, fortunately we have a lot

of paying customers to pay our bills now.

254

:

And those use cases

are not that exciting, to be honest.

255

:

It sure may seem like that to you because

you guys have been doing it for so long.

256

:

Saga.

257

:

Oh, another whatever,

10,000 customers on water meters

258

:

from the kind of outsider's perspective

where I'm coming from, it's

259

:

just the coolest thing to think

that we spent whatever it was 40, 50,

260

:

80 years with these,

you know, dumb devices and thinking

261

:

like they will never figure out

what's going on.

262

:

You just kind of bill people once a month.

263

:

And we've come into this time

really in the last probably ten years,

264

:

where all of a sudden

we now know so much more about our world.

265

:

And so an individual use case

maybe like right, water meter or whatever.

266

:

But thinking about what impact does it

have, as you're saying on the on society,

267

:

to understand these at a granular level

outside of the lab,

268

:

what you're doing is pretty, pretty rad.

269

:

Yeah. No. Yeah.

270

:

So I, I very much agree with that to you.

271

:

And also I think it's really good for us

and for everyone in IoT to realize

272

:

that we are basically generating data

now also for

273

:

which is very relevant

also now in the whole AI revolution,

274

:

where there's so much need

for high quality, structured,

275

:

timestamped data that we might not even,

276

:

you know, see the impact of that yet,

but just the fact that we are

277

:

generating a lot of stuff

and that we're making it work

278

:

for our customers, that they will be able

to train on that data and they will

279

:

surprise us in a few years

from now with really exciting solutions.

280

:

Yeah.

281

:

And that could be,

you know, the free to use case.

282

:

You know,

it starts with a compliance report

283

:

and a PDF that says, you know,

284

:

this fridge was in the right

temperature range ranges.

285

:

And maybe in a few years from now

286

:

they're able to reduce power consumption

of those fridges by 25%.

287

:

And you know,

those things are are incremental.

288

:

But they are increasingly exciting

for sure.

289

:

Yeah, it makes a difference.

290

:

I was at a conference last year,

the indoor ag agriculture conference up

291

:

in, Vegas, and one of the speakers

there had said like,

292

:

oh, I've heard for years

that AI is going to make my data valuable.

293

:

I've never been able to sell it.

294

:

And I can't help but think in the years

since, like, it must be a lot easier

295

:

to sell data and companies like yours,

if not position to sell.

296

:

The data that you have

are certainly positioned to kind of build

297

:

these bridges between customers

who are generating this data,

298

:

and clients who might want to buy

that data to train their own models.

299

:

Is that something that you guys

are looking at?

300

:

Or you're like, hey, we're just going

to let the market take care of that?

301

:

We are currently not looking at that

on the application level.

302

:

I can imagine that it could become

303

:

very much interesting for customers

to be able to sell their data and

304

:

to use our platform

as a distribution channel,

305

:

because the data is there already,

and that they may that,

306

:

you know, it could be as simple

as having their customers

307

:

subscribe to the data feed

that is in our platform.

308

:

And we may also have play

a commercial role in that.

309

:

I think on the shorter term,

it would be in the in the network

310

:

metadata,

which is a lot of data that we can use.

311

:

And we can also help customers with a,

312

:

you know, a recommendation algorithms

to optimize their network.

313

:

And the ROI on that is almost

always positive, because even if we

314

:

which is recommends, you know, place 1

or 2 gateways there in there.

315

:

Right, that will have this effect

on the LoRaWAN data rates.

316

:

And it will have this effect on the effect

error rates and this effect

317

:

on the battery

consumption of those devices,

318

:

because they don't have to scream

at as of 12. Yep.

319

:

And for that, yeah,

certainly we need a lot of data.

320

:

And this is data that we have.

321

:

And it makes most sense for us

if we take this role

322

:

and build such AI driven solutions.

323

:

Yeah.

324

:

I'm thinking because you guys are European

325

:

based in the privacy laws,

there are pretty strict.

326

:

So you guys have kept the network data

the whole time,

327

:

or is that something that each customer

328

:

has to opt in to and said, yeah,

I've got this gateway up

329

:

and you can use this data,

or is that just a fundamental piece of

330

:

if you're on TTN, like,

we're going to have this data

331

:

so that we can help all customers

figure out how to get the best coverage?

332

:

Yeah.

333

:

So this is not

any personal identifiable data.

334

:

Right. So this is just network metadata.

335

:

This is this is really nothing more than

gateway locations and device locations.

336

:

And and the RFC conditions.

337

:

Okay. It's like RSS I think as last.

338

:

Yeah okay. Yeah.

339

:

The packet loss

and that's already most of it.

340

:

Yeah.

341

:

So we need this data

for us to do our job in the first place.

342

:

And we store it indeed for longer periods

to do very useful analysis.

343

:

Yeah. Yeah.

344

:

I can just imagine

if you have it back to the day to day one,

345

:

you probably have somewhere

this visualization of like,

346

:

this is how the network grew.

347

:

And I know

348

:

the one of the first use cases was like,

hey, this how fast we could cover a city,

349

:

but would be super cool

to see like this so fast.

350

:

We cover the world

and this is like where the

351

:

where you can see it in the same way

that other networks have the,

352

:

the same kind of visualization.

353

:

Like this is where you could put a gateway

in this where you don't need to put one.

354

:

Yeah. Very cool. All right.

355

:

As we wrap this thing up, we're right

at the end of:

356

:

So just a couple days left in the year.

357

:

Anything that you're super psyched on

for:

358

:

we're going to go out and crush

this next year.

359

:

We're always very excited

for the next conference.

360

:

And but that's going to be in September

only.

361

:

But preparations are really start for sure

quite soon.

362

:

So that's that's definitely on our agenda

in the already in a few months from now.

363

:

And we are very much looking forward

to that.

364

:

On the LoRaWAN level.

365

:

There are very nice things in the pipeline

366

:

which I am not able to share a lot

to today.

367

:

It's exciting. It's good.

368

:

We need to wait for the Lora Alliance

to have everything ready. Yep.

369

:

But certainly on the LoRaWAN specification

site there are very nice things coming.

370

:

Not only on the LoRaWAN spec itself,

371

:

but also making stuff work better together

in a in a plug and play manner.

372

:

So I'm spending a lot of time on that,

and I'm expecting that we will see some

373

:

nice releases in the coming,

say 6 or 7 months.

374

:

Super exciting.

375

:

And will we see you at CES, are you not?

376

:

You're not coming to America this year.

377

:

So yeah, besides my three year old,

I also have a six week old daughter now.

378

:

So, yeah, so it's very nice,

but I'm skipping this year.

379

:

Yeah, I know it's a nine hours

time difference and a very long trip,

380

:

so I can't leave my wife alone with two

if I'm not yet.

381

:

Yeah, yeah.

382

:

It's funny.

383

:

I feel the same way.

384

:

And like, the stuff going on in Barcelona,

like, it's a huge time difference.

385

:

And I'm there for two days.

386

:

It's like a

it seems like a two day commitment,

387

:

but it's like a ten day commitment

to, to get ready and recover.

388

:

Yeah.

389

:

So Johan, thanks

a ton for making the time.

390

:

I know you're busy.

391

:

Obviously you got these two kiddos

plus a business to run,

392

:

so I really appreciate you coming on

and talking to us a little bit about

393

:

what are the things, industries

and things network is doing

394

:

and how you guys are

expanding. Thank you. Thanks.

395

:

That's it for

this episode of The Business of LoRaWAN.

396

:

If you want to go deeper

and actually deploy devices,

397

:

the MeteoScientifc

console is the fastest way to do that.

398

:

And honestly, it's

also the best way to support the show.

399

:

When you use the console, you're not just

listening, you're participating

400

:

in the same real world LoRaWAN work

we talk about here every week.

401

:

You can get started with the free trial

at MeteoScientific.com.

402

:

Huge thanks to the sponsor of this show,

the Helium Foundation,

403

:

for supporting open LoRaWAN

infrastructure worldwide.

404

:

Check them out at helium.foundation.

405

:

And if the show has been useful,

a quick rating or review on Apple Podcasts

406

:

or wherever you listen.

407

:

This really helps

408

:

people find it and helps the show grow

so we can help more people.

409

:

I'm Nik Hawks with MeteoScientific.

410

:

I'll catch you on the next episode.

Next Episode All Episodes Previous Episode

About the Podcast

Show artwork for The Business of LoRaWAN
The Business of LoRaWAN
Learn From the Pros

Listen for free

About your host

Profile picture for Nik Hawks

Nik Hawks

Incurably curious, to stormy nights and the wine-dark sea!