Earlier this week, Liron Shapira, a self-proclaimed Web 3.0 skeptic, shared an article from The Generalist newsletter that dug deeply into the Helium Network and with it, his skepticism about the business. One of the biggest shockers in the article — and his related tweet thread — was the fact that Nova Labs, the company behind the Helium Network, made just $6,500 in revenue in June.
Nova Labs CEO Amir Haleem didn’t refute that number. But I’d like to put it into context and explain why it isn’t the indictment against the business model others may think it is while also explaining my own concerns about Helium’s future.
First, a refresher. Nova Labs, which changed its name earlier this year in an attempt to distance itself from the low-power long-range Helium Network, has a long history in the IoT sector. The company was started in 2013 and went through several pivots before settling on the idea of using tokens to incentivize people to place Helium hotspots in their homes. The hotspots act as wireless gateways, receiving LoraWAN or Helium’s own proprietary LongFi radio signals and using the hotspot owner’s broadband network to transfer the data to the internet.
To reward users who put the hotspots in their homes, Nova Labs established the Helium Network Token (HNT). A hotspot owner generates HNTs by providing both coverage and transferring data. And companies that want to buy data on the network use the HNTs to buy data credits. Each data credit is worth $0.00001. So the June revenue reported by Nova Labs from data transfers represents 650 million data packets.
As a point of disclosure, I run a Helium hotspot in my home, and have since 2020. I made about $20,000 by selling HNTs in 2021 and currently have 126 HNTs worth about $1,195 at the current price of roughly $9. In the time I’ve had the hotspot, the price of HNTs has ranged from less than $1 to a high of around $55, peaking as interest in cryptocurrency peaked.
Which leads to the first big question anyone interested in the Helium network should be asking: What happens to a decentralized wireless network if people who purchased their hotspots expecting to make $50 a week start making a mere $4 a week? Indeed, is that post-crypto crash rate enough incentive for people to maintain their hotspots? It is for me because I went into this thinking it would be fun to support an IoT network. And I didn’t spend $500 or even a few thousand dollars on a hotspot during the frenzy of 2021 expecting to make bank.
Tens of thousands of people did, however. Kevin, my colleague on the podcast, is one of them. He ordered his hotspot in March 2021 and it arrived in April 2022. He paid $450 for it, and based on the prices of HNTs today he can expect to make $15.75 a month. But while he’s understandably disappointed in the ROI, that wasn’t his main purpose for buying the hotspot. Like me, he’s interested in supporting a decentralized IoT network, so he says he’ll happily keep it operational.
In the meantime, Helium has added roughly 500,000 hotspots to the network in the last year alone. They are operated not just by individuals, but by businesses that were created to buy and distribute hotspots to specific areas and homes in exchange for a cut of their HNTs. How many of those businesses want passive income in the $15 per month range?
Moreover, since many of the inflated income numbers were created by the speculative frenzy around HNT and the heightened ability to mine HNTs during the earliest stages of building a network, what will the true income opportunity be?
I may have bad news on that front. One reason Helium is so compelling to me as an IoT reporter is that it fundamentally changes the cost of delivering data over a LoRaWAN network. Such networks are useful for sending small bits of data over large distances.
Cellular networks can also provide this type of long-range connectivity, but they cost more. For example, some of the best pricing I’ve seen on the cellular side comes from companies such as Hologram or Blues Wireless. Hologram charges 78 cents per MB per month per device (70 cents per month for the device and 8 cents for a MB of data), and Blues lets a customer buy a module with an MCU and connectivity for $49 in exchange for 500MB of data over the subsequent 10 years. Both companies offer different pricing based on volumes.
They also use LTE CAT-M networks or even full 4G, which means they can send a lot more data than a LoRaWAN network, so this isn’t an apples-to-apples comparison. Nova Labs COO Frank Mong told me that Helium’s packets are about 24 bytes in size. But the devices running on the Helium network aren’t going to need multiple megabytes of data. According to previous conversations with Mong, a packet is enough to send a GPS location, time, and temperature, or another bit of data. So if you had a sensor transmit every five minutes, the year-round cost would be about $1.05 on the Helium network.
If I wanted to send that same amount of data using Hologram, I’d pay $8.40 for the device and 8 cents per MB of data. Based on the number of packets I’d send in the earlier scenario, the end user would rack up roughly 2.5 MB of data, leading to an annual cost of $8.64 per device. The discrepancy in costs is partly because the Hologram charge includes an eSIM and device management and partly because the customers of each network have different levels of technical expertise and needs.
But the important takeaway here is that the Helium network is designed for hundreds of millions of devices to send small bits of data really cheaply. Which means that we won’t see amazing revenue from Nova Labs until we start seeing hundreds of millions of devices on the network. Today Nova Labs has Lime, Cisco, MyDevices, and Careband as customers on the Helium network, but many of those are in trial phases or selling relatively few LoRaWAN devices.
This is exactly how the network is supposed to work. This isn’t a network designed for connected vehicle trackers or hot tubs. This is a network designed for the likes of UPS to install a cheap sensor on every package and monitor its progress around the country. It’s for a few hundred million outdoor weather stations or mailbox sensors. As such, it needs to be cheap and it needs to convince customers to put billions of devices on the network.
Which takes us to another challenge the Helium Network faces, and one that could affect its overall $1.2 billion valuation. There are still too few LoRaWAN devices out there. When I was giving a talk at the LoRaWAN World Expo earlier this month, I was dismayed both by the lack of sensors and the subsequent frustration expressed by many attendees, who said it was still too hard to get LoRaWAN devices online easily.
Without LoRaWAN devices, the network will remain empty and won’t ever achieve the volumes it needs to drive revenue into the hundreds of millions it would need to support its valuation. Nor will it generate the data transfers over hotspots that will, in turn, generate more HNTs for the hotspot owners that comprise the bulk of the Helium network. And if those owners aren’t incentivized to keep operating their hotspots, the network will lose coverage and capacity, which will make customers reluctant to use it.
Basically Nova Labs is performing a big experiment in market-making and cryptocurrency at the same time. Each is difficult, but its attempt to do both at the same time is fun to watch.