Choosing the Right LPWAN: LoRa vs NB-IoT vs Cat-M1 for Industrial Deployments

After years of deploying connected-worker solutions across oil rigs, construction sites, mining operations, and logistics hubs spanning four continents, I've learned that connectivity choice can make or break an industrial IoT project. It's rarely a clean technical comparison — it's a tangle of coverage, cost, regulation, power budgets, and operational reality.

Here's what I've learned about the three dominant LPWAN technologies for industrial deployments, drawn from real projects rather than datasheets.

The landscape at a glance

Numbers on a comparison table are useful, but they don't capture the nuance of deploying in a live industrial environment. Let me share what the datasheets won't tell you.

LoRa: the workhorse for private networks

LoRa's greatest strength is also its most misunderstood: you own the infrastructure. In environments where cellular coverage is unreliable, non-existent, or where you simply can't depend on an MNO's roadmap, LoRa gives you complete control.

On a mining project in the Middle East, we deployed LoRa gateways across a remote site where the nearest cell tower was kilometres away. Within days, we had a working sensor mesh reporting environmental data back to a central hub. No SIM cards, no carrier agreements, no monthly per-device fees eating into the business case.

The trade-off is throughput and latency. LoRa is excellent for small, periodic data packets — environmental readings, asset heartbeats, threshold alerts. It struggles with anything requiring real-time responsiveness or larger payloads. If your use case involves streaming audio, transmitting images, or requiring sub-second command-and-control, LoRa isn't your answer.

LoRa excels when you need hundreds of low-cost sensors reporting periodically in a defined area. Think campus-scale, not country-scale.

NB-IoT: the carrier-grade metering play

NB-IoT was designed for static, low-power devices that transmit small amounts of data infrequently. Smart meters, water sensors, environmental monitors — it's purpose-built for this category.

Where NB-IoT shines is deep indoor penetration. The narrowband signal can reach basements, underground car parks, and building cores that defeat conventional cellular. For utility metering and building infrastructure monitoring, this is transformative.

However, I've seen NB-IoT struggle in two scenarios that matter enormously for industrial deployments. First, coverage gaps are real. MNO rollout varies dramatically by territory. In the UK and parts of Western Europe, coverage is maturing nicely. In parts of Asia-Pacific, the Middle East, and North America, it can be patchy. We've had projects where NB-IoT was specified based on carrier coverage maps, only to find that actual coverage at the site was insufficient.

Second, NB-IoT doesn't handle mobility well. If your device moves between cells — even slowly — you'll hit re-attachment delays that can cause data loss. For wearable devices on mobile workers, this is a deal-breaker.

Cat-M1: the versatile middle ground

Cat-M1 has become my default recommendation for connected-worker wearables and mobile asset tracking in regions where coverage supports it. It supports full cell handover, meaning devices can move between towers without losing connection. It supports VoLTE for voice. And it offers enough bandwidth for firmware OTA updates, which is critical for devices deployed in the field for years.

During the connected-worker project I led as CTO, we deployed Cat-M1 wearables in ATEX zones across oil and gas installations. The devices needed to simultaneously report GNSS location, gas sensor readings, proximity alerts, and physiological measurements — all while maintaining a persistent MQTT connection for real-time alerts. Cat-M1 handled this comfortably within its power budget.

The caveat is that Cat-M1 draws more power than NB-IoT or LoRa. For battery-powered devices, this means either a larger battery, more frequent charging cycles, or careful power management with aggressive sleep modes. On wearable devices where battery life directly impacts user compliance, this requires careful hardware and firmware co-design.

Multi-territory deployments: the hidden complexity

The most painful lessons I've learned have been around multi-territory certification and MNO compatibility. What works seamlessly in the UK may not work at all in Qatar, Australia, or the United States.

Each territory has its own ISM band allocations for LoRa, different MNO Cat-M1 and NB-IoT band support, different certification requirements, and different attitudes to spectrum usage. A product designed for EU deployment may need significant RF redesign for North America or Asia-Pacific.

My strong advice: make territory planning a first-class concern in your product architecture from day one. Don't treat it as a post-design certification exercise. The RF front-end design, antenna selection, modem choice, and firmware configuration all need to account for target markets from the earliest schematic stages.

Making the decision

There's no universal answer. The right choice depends on your specific deployment context. But after dozens of industrial IoT deployments, here's my framework:

Choose LoRa when you control the deployment site, need many low-cost sensor nodes, can install your own gateways, and don't need real-time responsiveness or mobility.

Choose NB-IoT when devices are static, data volumes are small and periodic, deep indoor penetration is needed, and the deployment territory has confirmed carrier coverage.

Choose Cat-M1 when devices are mobile, you need higher bandwidth or voice capability, firmware OTA is required, and you're deploying in territories with strong LTE-M coverage.

And increasingly, the answer is multi-mode. Modern chipsets from Nordic, Quectel, and others support both NB-IoT and Cat-M1, allowing dynamic fallback. For products targeting multiple territories, this flexibility is worth the marginal BOM cost increase.