Here is a pattern that repeats across portfolios: a property team runs a competitive tender, scores five vendors on a weighted matrix, picks the highest scorer — and eighteen months later the platform is quietly abandoned because the occupancy data never reconciled with the leasing team's numbers, or the integration with the ERP stalled in month four. The tender was rigorous. The procurement was still a failure. The gap is almost never in the scoring matrix. It is in what the matrix never asked.
This guide covers the questions that separate a working property operations software deployment from an expensive shelf license — written for people who sign the purchase order and then have to live with it.
Most RFPs open with a features inventory: dashboards, alerts, reporting, mobile access. Every serious vendor ticks every box, which means the exercise discriminates between nothing. Flip the document. Start with the three to five operational decisions the software must improve, and demand vendors show — with data from a comparable property — how their platform changed those decisions elsewhere. Examples that actually appear in asset management workflows:
If a vendor cannot connect a feature to one of your named decisions, the feature is decoration and should carry zero weight in scoring.
If occupancy or footfall data feeds lease negotiations, turnover-rent verification, or service-charge allocation, the data will eventually be challenged — by a tenant, an auditor, or a joint-venture partner. At that moment, a marketing claim of "99% accurate" is worthless unless it appears in the contract with a measurement methodology attached.
Push vendors on this distinction. A serious counting platform will commit to a contractual minimum accuracy — Vemco Group, for example, guarantees a minimum of 96% in contract, with typical real-world performance of 98–99% where lighting, entrance layout, and visitor behaviour allow. That structure is exactly what a tender should demand: a floor the vendor is liable for, and a realistic expectation above it. A vendor who will only quote the best-case number, and refuses to define how accuracy is audited (manual count validation, sample size, remediation timeline if the site falls below threshold), is telling you where the risk sits. It sits with you.
Many property operations platforms are quietly coupled to a specific sensor line. That coupling has three consequences procurement teams rarely price in:
Specify device independence as a hard requirement. A sensor-agnostic platform lets you standardise the analytics layer across a mixed portfolio while choosing hardware site by site on merit and price. This single clause changes the negotiating dynamics of every renewal for the life of the contract.
Every vendor claims open APIs. The useful question is different: name three integrations with the specific systems we run — our BI stack, our ERP, our tenants' POS environments — that are live in production today, and put us in contact with the customer who owns them. Vemco's VemFusion layer, for instance, connects counting and occupancy data into POS, BI, ERP, and CRM systems; whichever vendor you evaluate, the test is the same: production references, not API documentation.
Also settle hosting before contract signature, not after. If your fund's data governance or a joint-venture agreement requires data residency or a private environment, confirm the vendor offers a private cloud option and that it appears in the order form — not just in a sales conversation.
A field observation from people who commission these systems: the number one cause of degraded counting data in year two is not sensor failure — it is the building changing around the sensor. A tenant fit-out moves an entrance. Marketing installs a promotional display in a doorway. A new mat changes reflectivity under an overhead counter. None of these trigger an alert in a poorly designed platform; the counts just drift. In your contract, require automated data-health monitoring and a defined recalibration process after layout changes, with named response times. Vendors who deploy at genuine scale — Vemco processes over 85 million counts per day across its customer base — have had to solve drift detection operationally; ask any shortlisted vendor to demonstrate theirs on a live account, not a slide.
License fees are typically a third to half of true cost. Build a five-year model per asset that includes:
On exit terms specifically: you should own the historical data outright, in a documented format, exportable without a professional-services engagement. A vendor who resists this clause is planning to hold your history hostage at renewal. Walk away or price the hostage-taking into your scoring.
Pilot on your hardest site, not your easiest — the mixed-use asset with three entrances and a shared lobby, not the single-tenant office with one door. Validate accuracy with independent manual counts across different day-parts. Have your actual operations team, not the vendor's solutions engineer, build one report end to end. And test scale honestly: a platform that works for one building must prove it can consolidate a portfolio into a single view, because that consolidation is where asset managers extract the real value, and it is the hardest thing to retrofit.
If you are drafting an RFP or tender for property operations software and want a reference point for contractual accuracy guarantees, sensor-agnostic deployment across a mixed portfolio, and integration into your existing BI and ERP stack, talk to the Vemco Group team at vemcogroup.com/contact-us — bring your requirements document and they will show you exactly which clauses they can commit to in writing.