The Unsung Hero of Micromobility: How Cellular IoT Keeps E-Scooters from Becoming Expensive Sidewalk Litter

Shared e-scooters have quietly become one of the most successful connected-device deployments in any consumer-facing industry. The reason they keep rolling – through winters, permit reviews, and fleet relocations – is cellular IoT.

The ride ends. The work doesn’t.

When it comes to eScooters, the process can seem simple. A rider pulls up to their destination, taps “End Ride” in the app, and walks away. From their perspective, that’s the end of the story.

For the operator, it’s the beginning of another one. That scooter now has to report its location, prove it was parked within a permitted zone, confirm its battery level, stand by for the next rider, and – if it stays idle long enough – signal a charging partner to come pick it up. If any step in that chain breaks, the entire economic model of shared micromobility wobbles.

A modern shared e-scooter is, at heart, a cellular IoT device with wheels. Its connectivity is what turns a parked vehicle back into a ready-to-ride asset the moment the next rider taps unlock.

This post is about the layer almost no one talks about: the always-on cellular connection that makes every ride, refund, and relocation possible.

---

Micromobility isn’t a trend anymore – it’s infrastructure

The first wave of shared scooters arrived fast and a little chaotically. The second wave (the one we’re in now) is mature, regulated, and deeply integrated into the way cities move.

The numbers tell the story. According to NACTO (National Association of City Transportation Officials)’s 2023 Shared Micromobility Snapshot (published July 2024), shared e-scooter systems in the United States and Canada logged 65 million trips in 2023 – a 15% rebound from 2022, recovering from a post-pandemic dip and re-establishing e-scooters as a core mode of urban short-trip travel. Usage is concentrated where density is: 40% of all shared e-scooter trips took place in just 10 major cities, such as Los Angeles, Washington DC, and Austin.

Shared e-bikes grew even faster (close to 50% year-over-year) signaling that the broader shared micromobility category is expanding, not converging on a single vehicle type. And the market is anything but uniform: San Diego saw shared e-scooter trips drop from roughly 3 million in 2022 to 360,000 in 2023 as permit conditions changed, a reminder that city-level policy can reshape a local market almost overnight. That volatility is exactly why real-time fleet visibility, compliance data, and the ability to reconfigure a fleet in response to permit changes are now baseline operator requirements.

Operators have consolidated, margins have tightened, and city relationships have matured from “permit pilot” to “core transit partner.” The use cases have expanded well beyond the stand-up kick scooter: e-bikes, shared mopeds, delivery robots, corporate and campus fleets, and last-mile logistics all ride the same underlying connectivity stack.

That maturity has come with a new floor for what “good” looks like. Modern e-scooter regulations assume the device can be located, disabled, audited, and geofenced in real time – which suggests cellular connectivity is not optional. Cities now require live data feeds via the Mobility Data Specification (MDS). Permit renewals hinge on parking compliance data. Slow zones are enforced by the device itself, not by signs on a pole.

None of that works without an always-on cellular link between every scooter and the operator’s cloud.

---

What a scooter actually has to do between rides

The rider-facing moment – unlock, ride, park – is the visible tip of a much larger operational iceberg. Here’s what an individual scooter is doing in the background, continuously, for its entire deployed life.

Reporting its location, always. GPS provides the fix; cellular carries it home. Operators need both the current position and the history. I.e. where a scooter has been, how it got there, and where it was last seen if it drops offline.

Accepting unlock and lock commands. The moment a rider taps “Start Ride,” a command has to reach the scooter in under a second. The same goes for the other end of the trip. Latency here translates directly into rider frustration and lost revenue.

Enforcing geofences. Slow zones, no-parking zones, permit boundaries, campus areas – the rules are updated from the cloud and enforced on the device. A scooter that enters a slow zone doesn’t ask for permission; it just throttles back.

Streaming battery telemetry. Operators route chargers and swap crews based on live battery data across the entire fleet. Without connectivity, there’s no routing – and without routing, a meaningful portion of the fleet is unavailable by morning.

Running diagnostics and firmware updates. Brake wear, motor faults, controller software patches, security updates — all pushed and pulled over cellular. Over-the-air (OTA) update capability is now table stakes.

Detecting theft and tampering. An accelerometer says “this scooter is moving.” GPS says “in a direction no rider authorized.” A cellular heartbeat lets the platform raise the alarm in real time. That combination is the difference between a recovered asset and a total loss.

Two Soracom engineers walked through exactly this data flow in an earlier post — How Hard is it to Visualize IoT Escooter Data Flow? — if you want to see what the telemetry looks like end-to-end.

---

Why cellular, specifically?

It’s a fair question, and one that gets asked more often than people might expect. Why not Wi-Fi? Why not Bluetooth? Why not one of the long-range, low-power protocols like LoRa?

Because scooters move, and because operators need to talk to them right now.

Wi-Fi depends on hotspots. That works at a depot or warehouse. It falls apart the moment the scooter rolls down the block.

Bluetooth is the right tool for one specific job: the handshake between the rider’s phone and the scooter at the start of a ride. It’s not a fleet-management channel.

LPWAN technologies like LoRa and Sigfox are excellent for low-bandwidth, low-frequency telemetry — meter readings, environmental sensors, asset pings. They don’t carry the bandwidth or latency profile needed for OTA firmware updates, real-time geofence enforcement, or sub-second command response.

Cellular IoT — particularly LTE-M and LTE Cat 1 for scooters, and LTE Cat 4+ for heavier connected vehicles — sits at the intersection of coverage, power, latency, and bandwidth that micromobility actually requires.

Connectivity	Coverage	Power draw	Latency	Fit for scooters?
Wi-Fi	Hotspot only	High	Low	Depot only
Bluetooth	~10 meters	Low	Low	Rider handshake only
LoRa / Sigfox	Wide area	Very low	High	Limited telemetry
Cellular (LTE-M / Cat 1)	City and national	Low–medium	Low	Yes — the default choice

---

The hidden hard part: scaling across cities, carriers, and countries

One fleet in one city on one carrier is a solvable problem. Twelve cities across four countries, with variable coverage quality block by block, is a different animal.
This is where connectivity stops being a checkbox and starts being a strategic decision. A few of the recurring challenges operators hit at scale:

SIM logistics. Swapping SIMs per market — or worse, building devices that only work on a single carrier — doesn’t scale. Every SIM is a manual step, and every hardware SKU is a supply-chain liability.

Carrier fragmentation and roaming. Some carriers dominate one city and disappear in the next. Roaming agreements vary. Data plans and billing structures don’t line up across markets, which makes financial reconciliation an operations headache all on its own.

Coverage dead zones. A scooter in an underground parking garage, near a bridge, in a rural charging warehouse, or on a ferry across a city harbor can drop signal for hours. Operators need a plan for the last block, not just the average-case coverage map.

Security. Millions of connected devices on public cellular networks is an attack surface. Device identity, private networking, and encrypted data paths aren’t optional at fleet scale.

Our earlier deep-dive How Hard is it to Build an IoT Electric Scooter Fleet like Bird or Lime? walks through exactly how these pieces come together in practice.

---

What “good” looks like: a connectivity checklist for micromobility

If you’re evaluating connectivity partners for a micromobility fleet — whether you’re a global operator, a new entrant, or an OEM building the next generation of connected two-wheelers — these are the capabilities that separate a platform that scales from one that becomes a bottleneck.

One SIM, many carriers. A single global IoT SIM that roams across multiple networks in a given market, and across borders, without requiring a hardware change. This alone eliminates a huge class of logistics and SKU problems.

Flexible SIM form factors. Traditional SIM, eSIM, and iSIM — so hardware designers can trade off size, power consumption, and ruggedization without being boxed in by connectivity decisions made years earlier.

Private, secure networking. Device traffic doesn’t need to touch the open internet. Private IP networking, VPN integrations, and encrypted cloud delivery paths are now baseline expectations.

Fleet-wide visibility and automation. Live dashboards, alerts on offline devices, OTA update pipelines, and — increasingly — AI-driven fleet insights that surface patterns across millions of data points.

Transparent, usage-based pricing. A scooter that’s idle overnight shouldn’t cost the same as one that’s streaming diagnostics during peak hours. Flat-rate data plans at fleet scale quietly destroy unit economics.

Backup connectivity paths. For the genuine edge cases — rural charging hubs, cross-border repositioning, remote maintenance facilities — a satellite fallback keeps coverage intact where cellular thins out.

Soracom provides all of the above, built into one cloud-native IoT platform — but you should expect this list from any serious connectivity partner in 2026.

---

Scooters today, autonomous everything tomorrow

Everything operators are learning right now in shared e-scooter fleets generalizes. E-bikes, shared mopeds, delivery robots, connected cargo bikes, last-mile EV vans, and eventually autonomous curbside vehicles all face the same fundamental connectivity challenges: real-time location, remote command and control, OTA updates, geofence compliance, secure telemetry, multi-market scaling.

The operators winning this next phase aren’t necessarily the ones with the flashiest hardware. They’re the ones whose fleets are reliably reachable — whose data pipelines are clean, whose city relationships are built on trustworthy reporting, and whose devices are quietly doing their jobs in the background.

We covered the broader arc in IoT Gets Rolling: Smart Scooters and the New Connected Commute — worth a read if you want the wider view on where connected mobility is heading.

---

The quiet work behind every ride

A shared e-scooter at the curb looks like a product. It’s really a promise: that the moment a rider taps the app, a network somewhere is ready to answer, a cloud platform is tracking the device’s health, and a set of systems are coordinating to turn a parked vehicle into a trip-in-progress.

That promise is cellular IoT. It’s what keeps fleets running, cities permitting them, riders choosing them, and operators funding them.

If you’re building, scaling, or modernizing a connected mobility fleet, we’d be glad to talk through the connectivity layer. Let’s connect →

---

Frequently asked questions

How do shared e-scooters connect to the internet? Shared e-scooters use cellular IoT connectivity — typically LTE-M or LTE Cat 1 — built into the onboard controller. A small IoT SIM or eSIM inside the scooter maintains an always-on connection to the operator’s cloud platform, carrying GPS data, battery telemetry, diagnostics, and remote commands.

What’s the best connectivity option for a shared e-scooter fleet? Cellular IoT is the default choice, because it combines broad outdoor coverage, low enough power draw for battery-powered devices, and low enough latency for real-time unlock, lock, and geofence commands. Most operators choose multi-carrier IoT SIMs to avoid single-carrier coverage gaps.

Why do micromobility companies need multi-carrier SIMs? Coverage quality varies carrier by carrier and block by block. A multi-carrier IoT SIM can automatically attach to the strongest available network, reducing offline time, eliminating per-market SIM swaps, and simplifying scaling into new cities or countries.

How does cellular IoT help prevent e-scooter theft? Each scooter reports its location and motion state over cellular on a regular heartbeat. When a scooter moves without an active rider session, the platform can raise an alert, push a remote-disable command, and track the device’s location in real time — dramatically improving recovery rates.

What’s the difference between LTE-M and NB-IoT for scooters? LTE-M offers higher bandwidth, lower latency, and mobility support (handoff between cell towers), which suits moving vehicles like scooters. NB-IoT is optimized for stationary, low-power telemetry like utility meters. For real-time fleet operations, LTE-M (and LTE Cat 1 where available) is the typical fit.

---

Ridership and market figures cited above are drawn from the National Association of City Transportation Officials (NACTO), “2023 Shared Micromobility Snapshot,” published July 2024. Source.