How it works · Deployment
Live in five days.
Calibration and driver training in the first two days, quality checks and sign off by day five, then four weeks of hypercare. Multi site rollouts run to the same schedule, one site at a time.
The five days
What happens, day by day.
Driver training is part of the week, not a project that starts after it. Nothing in the sequence waits on your IT team.
- Unwrapping
- Calibration
- Initial recordings
- LPN translation validation
- Racking navigation validation
- Driver training starts
- Quality adjustments
- Admin training
- Acceptance test preparation
- Acceptance test sign off
- On site availability
- Retrospective
- Four weeks of hypercare
Calibration on day one teaches the unit your racking geometry, your aisle widths and your shelf levels. Day two is where label reality gets tested: LPN translation is validated against your actual stock, and racking navigation is checked aisle by aisle, at the same time drivers begin training on the unit. By day four to five, your team runs the acceptance test and signs it off, so the accuracy figure the site starts with is one your own people measured. How that measurement works is on the accuracy page.
What this asks of you
Short list to bring, short list to skip.
Four things to have ready, and six things you do not have to arrange. Nothing in the sequence waits on a procurement cycle or a facility change.
What Sentispec needs from you
- A forklift you already own. It picks the recording unit up like any other load. No vehicle is bought, fitted or commissioned.
- A driver for the length of a run. The driver drives the aisles at normal travel speed, one shelf level at a time, which is why training is two to three days rather than a certification programme.
- Access to the racking during calibration and the acceptance test. Nothing is shut down for it, but the first passes need the aisles available.
- A stock export, if you want a discrepancy report from day one. A file exchange is enough. There is no schema to design and no field mapping project.
What Sentispec does not need
It works out of the box on a stock file, and integration effort is zero to start counting. A standard API is available on the integrations page when you want corrections to sync automatically, and that is a decision you can take months after go live rather than before it.
Proving it before committing
One warehouse first, because there is nothing to depreciate.
The usual capital equipment purchase forces a decision before the evidence exists. A robot or a drone is bought, housed and committed to a building before anyone knows what it will actually find. Sentispec Inventory does not ask for that decision.
- The Site tier is a single warehouse. Setup is €9,950 and the annual fee is €39,550, published in full on the pricing page. There is no larger commitment sitting behind it.
- Nothing arrives on your balance sheet. The recording unit is leased inside the annual fee, so proving the case at one site never becomes a stranded asset if the case does not hold up.
- The evidence is the deciding factor, not a sales cycle. By acceptance test sign off on day five, you have a real accuracy figure, a real discrepancy list and a real recording behind every line, from your own warehouse rather than a reference site.
- Expansion follows the result. Five warehouses or more moves to the Network tier at €8,950 setup and €37,450 per year, per site. Each additional site still gets its own five day deployment, so scaling up never means scaling risk with it.
You are not buying a machine and hoping the case works out. You are running a five day trial at one site and deciding what to do next with the result in hand.
After go live
Four weeks of hypercare, not a handover email.
Acceptance test sign off is the end of deployment, not the end of support. The month that follows is where a new habit either sticks or slips, so it is covered rather than left to your team alone.
Hypercare runs for four weeks after go live, and it is included in the annual fee alongside ongoing support, maintenance, camera leasing and every software upgrade, with nothing billed separately. It is the period where drivers settle into running counts as routine, where admin users get comfortable working the discrepancy list, and where any edge case that the acceptance test did not surface gets resolved quickly rather than becoming a support ticket months later.
What comes out the other side of hypercare is a site running continuous cycle counts as a normal part of a shift, not a stocktaking event on the calendar. Covering 500 to 1,000 pallets takes 20 minutes, typically driven during a break, which is what lets sites hold 99%+ accuracy all year rather than restoring it once a year and watching it drift. Full detail on the number itself is on how accuracy is measured, and the ten deployments we publish, break even included, are on the benchmark page.
Next step
See what day one to five would look like at your site.
A 30 minute demo covering calibration, driver training and what your team would be doing on each of the five days.