Why SmartFactory
The MES built this decade. Priced like it.
Production monitoring isn’t new — but most of the platforms selling it were architected twenty years ago and priced by the module. SmartFactory starts from a different premise: one platform, everything in, your data open to you, and numbers that admit their own limits.
| Capability | SmartFactory | Legacy monitoring platforms |
|---|---|---|
| What you get for the money | ||
| Energy monitoring | ✓ Included in every plan, per machine, live and historic | Typically a paid add-on, per machine per month |
| Energy per unit of output | ✓ Machine-level kWh tied to the job that ran, so energy per kilogram or per part is a query over any period you choose — during a compliance target period, not after it | Site-wide kWh off the bill, reconciled against output in a spreadsheet when the assessment falls due |
| Your production cost per part | ✓ Your energy, material, labour and absorbed overheads resolved to one £ figure on the works order, frozen when the month closes — what a part costs you to make, defensibly | Energy reported in kWh, if at all — production cost per part and the overheads left to a spreadsheet |
| Users | ✓ Unlimited on every plan | Unlimited on some platforms, per-seat on others |
| Modular tiers | ✓ Buy the modules you need — quality-only through to full MES — and switch more on without re-implementing anything | A platform bundle, with the parts you wanted priced into the tier above |
| API & webhooks | ✓ Open REST API and HMAC-signed webhooks on every plan, with published rate limits and one-click replay | Often gated to top-tier plans, rate-limited, unsupported |
| Depth where the work actually happens | ||
| Maintenance | ✓ Fault reported from a QR label in under a minute, routed, escalated on a clock, and confirmed fixed by the operator who reported it | A downtime reason code, and a separate CMMS to buy |
| Tooling & mould PM | ✓ Tools are assets; cycles accumulate per mount and lifetime across machines, and a mould raises its own 250k service | Calendar-based preventive maintenance, because a calendar is all the data supports |
| Tool changes | ✓ Changeovers derive themselves — remove, preheat, install — with preheat stations as a finite resource on their own schedule lane | A fixed changeover allowance on the Gantt, if changeovers appear at all |
| Planning | ✓ ERP-agnostic import with preview before commit, a working calendar that retro-fixes OEE, eight named conflict codes | ERP integration project, or manual entry |
| Process design | ✓ A drag-and-drop canvas that records which operation consumes which material — operation-level consumption that no platform in this bracket records — and refuses to guess the ones nobody has allocated | A routing typed into a grid, with materials attached to the order rather than the operation |
| Traceability & recall | ✓ Lot and serial genealogy walked in both directions off one append-only ledger: which job consumed this unit, and which units came from this lot | Batch numbers on a label, and a spreadsheet for the recall |
| Quality module | ✓ RFT by operator, shift and part, plus SPC: frozen control limits, Cp/Cpk and Pp/Ppk with the estimator named, Nelson rules | Basic quality checks, usually on higher tiers — no SPC |
| Labelling & traceability | ✓ Built-in label designer, Zebra and BarTender printing, a queue that pauses with a stated reason, label-to-inspection traceability | Rarely offered at all — a separate system to buy and integrate |
| People & skills | ✓ Operator identities, a skills matrix that validates the assignment board, and on-site presence from Paxton Net2 | Operator names in a dropdown |
| Can you trust the number? | ||
| Empty windows | ✓ “No production in this window” — never a 0% OEE that looks like a result | A zero, indistinguishable from a bad shift |
| Degraded data | ✓ Sensor dropouts show the figure with an amber caveat attached to it | The gap is interpolated, and nobody is told |
| Two honest numbers | ✓ Raw and capped performance; schedule-late and forecast-late; downtime MTTR and repair MTTR — both reported, never reconciled into one | One headline figure, method unstated |
| Unclassified downtime | ✓ Reported as a first-class metric, so you know what share of your downtime analysis is evidence | Unclassified time quietly absorbed into “other” |
| Imports | ✓ Preview before commit, exceptions naming row, column, value and error code; ambiguous dates are never guessed | Import and hope; fix the data afterwards |
| Getting it in, and keeping IT happy | ||
| Machine connectivity | ✓ OPC UA, relay/pulse, power-signature detection or our edge node — every machine connects, old or new | Proprietary I/O hardware you buy from the vendor, or the machine stays dark |
| Data capture hardware | ✓ Off-the-shelf smart meters and relays up to MID-certified billing-grade metering — non-invasive, clamp-on install | Vendor-locked I/O units, invasive wiring |
| Network & firewall | ✓ One gateway making a single outbound connection on 443, nothing inbound, buffering through internet outages | VPN, static IPs or an inbound rule into the plant network |
| Multiple sites | ✓ Locations within your company roll up into one view, while each site keeps its own boards, shifts, tariffs and users | A separate instance per site, consolidated by spreadsheet |
| AI analytics | ✓ Ask questions of your live data, get cited answers (Intelligence add-on) | Static report packs |
| Trying it | ✓ Instant simulated factory — see the full product with realistic data on day one | Book a demo, then wait for an install before you see your data |
| Architecture | ✓ Modern multi-tenant cloud platform, monthly releases | Single-customer instances of a codebase that predates the cloud |
“Legacy monitoring platforms” describes patterns common across established production monitoring products; check any specific vendor’s current pricing and packaging.
A straight answer
“Aren’t the established players safer?”
They have longer track records.
True — and if a 25-year-old install base is what you value most, you should weigh that. What you inherit with it is their architecture, their pricing model, and their upgrade cadence.
We have a modern platform, moving monthly.
SmartFactory ships on a modern cloud stack with releases every month, energy monitoring live at real UK manufacturing sites, and an AI layer already in production. You’re not paying off anyone’s legacy.
You don’t have to rip anything out.
Run SmartFactory on one site or one cell alongside whatever you have today, and compare after a quarter. Usage-based pricing makes a side-by-side trivially cheap to run: a single cell uses very little, so it costs very little.
Run the comparison on your own floor.
Start with a live demo, then a pilot on a handful of machines. If the numbers don’t make the case, they were never going to.