Introduction: The Build vs Buy Decision Every IoT Team Faces
Every IoT product team reaches the same fork in the road. You need a battery-powered remote sensor node — something that reads a 4–20 mA pressure transmitter, an RS-485 flow meter, or a pair of digital state inputs, then transmits the data over NB-IoT or LTE-M to a cloud dashboard. The question is whether to build a custom PCB from scratch or buy a production-ready module like the NORVI EC-M12.
On the surface, building looks cheaper. You pick an STM32L0 MCU, a SIM7070 cellular modem, design the power supply for a pair of ER34615H lithium cells, route the RF traces, and spin the board. The BOM cost at volume seems lower than buying a finished unit. However, that surface comparison misses the full cost picture — and in most real deployments, it gets the answer wrong.
This guide compares the two routes across five dimensions: BOM cost and NRE, RF design and antenna work, regulatory certification time, firmware bring-up, and total time to field. The goal is to give you a clear, number-driven framework — not a sales pitch — so you can make the right call for your project volume and timeline. The battery IoT node you choose determines your project’s cost, schedule, and risk profile for years.
The True Cost of a Custom Battery IoT Node PCB
BOM Cost at Volume
Hardware engineers often calculate build cost as BOM cost multiplied by unit volume. That calculation is correct but incomplete. Before you manufacture a single unit, you incur non-recurring engineering costs that the per-unit BOM figure never captures.
A custom battery IoT node built around the STM32L072, SIMCOM SIM7070 modem, dual ER34615H battery holders, M8 connector, DIP switch input selection, and an IP67 enclosure runs roughly $45–$80 per unit at 100-unit quantities, depending on component sourcing and enclosure specification. At 1,000 units, the BOM cost drops toward $30–$50 per unit. These figures look competitive against a finished module at first glance.
However, the BOM figure assumes you already have a working, certified design. It does not include the cost of getting there.
Non-Recurring Engineering (NRE)
Designing a battery IoT node PCB from scratch involves schematic capture, PCB layout, signal integrity review, power supply design for ultra-low-power operation, and design-for-manufacture review. For a hardware engineer billing at $80–$120 per hour, a competent four-layer board with an STM32L0 and a cellular modem takes 150–300 engineer-hours to design correctly. That puts NRE between $12,000 and $36,000 before you order a single prototype.
Add two or three prototype spins to fix layout issues — particularly around the modem RF section — and NRE climbs further. Most teams underestimate this figure by 40–60% because they plan for the first prototype to work and it rarely does.
RF Design and Antenna Work
The SIMCOM SIM7070 modem is a well-documented part but routing it correctly on a PCB requires controlled impedance traces, proper ground plane management, antenna matching network design, and clearance zones around the RF section. Getting these wrong costs you 3–6 dB of sensitivity, which translates directly into reduced range and higher power consumption as the modem retransmits packets.
Most product teams hire an RF specialist for the antenna matching and board review. RF consulting rates run $150–$250 per hour, and a proper review plus iteration takes 40–80 hours — adding $6,000–$20,000 to the project cost. On the EC-M12, this work is already done. The battery IoT node ships with a certified, tested RF section. You connect your sensor, write your application firmware, and deploy.
Regulatory Certification: The Hidden Schedule Killer
Regulatory certification is the single biggest underestimated cost in custom IoT hardware development. Any battery IoT node that transmits on licensed or unlicensed spectrum requires type approval before it can be sold or deployed commercially.
CE and FCC Certification for a Custom Design
CE marking under the Radio Equipment Directive (RED) requires radiated emissions testing, conducted emissions testing, receiver blocking tests, and SAR evaluation if the device operates near the human body. FCC Part 15 and Part 22/24 approvals add further test suites. A full CE + FCC certification run at an accredited test lab costs $10,000–$30,000 in lab fees alone and takes 4–9 months from first submission to grant.
If the lab finds a failure — which happens in roughly 60% of first-submission tests for new designs — you redesign, re-spin the PCB, and retest. Each re-spin adds 6–12 weeks and $5,000–$15,000 in additional lab costs. A single certification failure on a custom battery IoT node can delay a product launch by six months.
EC-M12: Pre-Certified and Ready to Deploy
The NORVI EC-M12 ships with certifications already in place, including EN 61131-2:2007, EN 61010-1:2010+A1:2019, EN IEC 61010-2-201:2018, and EMC compliance under the 2014/30/EU directive. Because the battery IoT node is a pre-certified module, you deploy it commercially the day it arrives. There is no lab queue, no test schedule, no re-spin risk. For projects with a fixed launch date or a contract delivery milestone, this advantage alone justifies the buy decision.
Firmware Bring-Up and Time to Field
Custom PCB Firmware Bring-Up
Bringing up firmware on a new PCB starts before the board functions correctly. You validate power supply rails, check oscillator operation, establish UART communication with the modem, verify GPIO operation, and debug the cellular registration sequence — all before writing a single line of application code. For a battery IoT node with a cellular modem and ultra-low-power sleep modes, bring-up typically takes 3–6 weeks for an experienced embedded engineer.
Stop-mode wakeup timing, interrupt handling for the STM32L072, modem power sequencing via the PA1 power key line, and microSD write operations all require careful debugging. Issues that appear only under field conditions — temperature cycling, cellular handover, battery voltage sag — surface weeks or months after initial bring-up.
EC-M12: Arduino IDE on Day One
The EC-M12 runs on the Arduino IDE ecosystem with full library support for the STM32L072 core. GPIO assignments, modem AT command sequences, sleep mode configuration, and SD card write operations are all documented in the NORVI application guides. A developer with Arduino experience writes production-ready sensor acquisition and MQTT publish code within a day or two of unboxing.
Because the hardware is validated and the bring-up work is done, your engineering time goes entirely into application logic — sensor calibration, transmission intervals, alert thresholds, and cloud integration — rather than hardware debugging.
Build vs Buy: Cost and Risk Comparison
The table below summarises the key cost and risk factors across both routes. Custom PCB figures reflect typical 100-unit development projects; EC-M12 figures reflect the buy route with NORVI’s battery IoT node platform.
| Cost / Risk Factor | Custom PCB Route | EC-M12 Buy Route |
|---|---|---|
| BOM cost (per unit, 100 units) | ~$45–$80 | EC-M12 unit price |
| Non-recurring engineering (NRE) | $15,000–$40,000 | $0 |
| RF / antenna design & layout | 6–12 engineer weeks | Done — CE/FCC certified |
| Enclosure + IP67 tooling | $8,000–$20,000 | Included (IP67 rated) |
| Regulatory certification (CE, FCC) | $10,000–$30,000 + 4–9 months | Pre-certified — ships immediately |
| Firmware bring-up time | 4–8 weeks | Arduino IDE — same day |
| First unit to field (time) | 9–18 months | Days to weeks |
| Break-even volume | 2,000–5,000+ units | No minimum — buy 1 or 10,000 |
| Supply chain / MOQ risk | High — component shortages | Managed by NORVI |
| White labelling / OEM | Full control | Yes — supported |
When Does a Custom PCB Make Sense?
The build route wins under specific conditions. If your deployment volume exceeds 10,000–50,000 units and you have 18+ months before the product launch date, the per-unit cost advantage of a custom BOM eventually exceeds the NRE and certification investment. Additionally, if your application requires a form factor, I/O combination, or frequency band that no off-the-shelf battery IoT node covers, custom hardware becomes necessary.
For OEM products where hardware differentiation is a competitive advantage — a proprietary sensor integration, a custom enclosure, or a unique power architecture — the build route also makes strategic sense. In those cases, the NRE is an investment in a defensible product, not an overhead cost.
When the EC-M12 Is the Correct Decision
For the majority of industrial IoT deployments, the EC-M12 is the right call. The IP67 enclosure, 38,000 mAh lithium battery capacity, dual 4–20 mA inputs, RS-485 Modbus, configurable digital inputs, multi-band NB-IoT/LTE-M, and pre-certified RF section cover the majority of remote monitoring use cases without any custom hardware work.
If your deployment is fewer than 5,000 units, your timeline is under 12 months, or your engineering team’s core competency is software rather than hardware, the EC-M12 delivers a working battery IoT node in the field faster and at lower total cost than any custom PCB route. White-labelling is supported for OEM deployments, bulk order discounts apply at volume, and you connect directly to your own cloud platform — AWS IoT Core, Azure IoT Hub, NORVI Cloud, or any MQTT broker — without licensing fees.
Conclusion: Make the Decision with Full Cost Visibility
The build vs buy decision for a battery IoT node is not primarily a BOM cost comparison. It is a total cost comparison that includes NRE, RF design, certification, bring-up time, and schedule risk. When you account for all of those factors, the EC-M12 wins decisively for volumes under 10,000 units and timelines under 18 months.
Custom PCB development makes sense when you have the volume, the timeline, and a genuine hardware differentiation requirement. For every other deployment — remote water monitoring, agricultural sensing, industrial pressure and flow telemetry, environmental data logging — the EC-M12 gets you to market faster, with less risk, and with a total project cost that custom hardware cannot match.
If you are working through this decision for a specific project, NORVI offers a free consultation at norvi.io/free consultation. Bring your deployment volume, timeline, and I/O requirements — and get a clear, honest answer on which route makes sense for your application.
by 
