01
It won’t guess a carton.
An order containing a SKU with no dimensions gets no recommendation — and the screen names the SKU, the missing fields, and what filling them in would unlock.
Volta · Warehouse
It says when it doesn’t know.
Two SKUs are missing dimensions, so it won’t recommend a carton for any order containing them — and it names the SKUs and the fields. Two tasks have no standard, so those hours aren’t scored. Everything else on the floor is one list, worst first.
7kinds of floor decision on one desk
3scopes a hold can have: item, lot, location
11%travel cut on a named pick resequence
78%standards coverage, reported before the scores
0tasks auto-assigned unless you turn it on
12modules on one platform
These are things Volta does — not results a customer measured.
01
The desk
Eighteen decisions, one list.
Pallets waiting to be put away. Carrier cutoffs you’re about to miss. A pick face that came up short. Counts that don’t match the book. Holds somebody placed on the floor. Seven different kinds of problem, ranked together, because the floor doesn’t experience them as separate systems.
Every row names the workbench that closes it — work it in Putaway, release it in the Pick Queue, inspect it in Holds. You never go looking for the screen that fixes what you just read.
02
Labour
It argues, so you can disagree.
“Move Calderon to zone C” is worth nothing. What arrives instead is the case: zone A is running 97 an hour with the most slack, zone C is your constraint at 66 — twenty-eight below floor average — and moving him puts 138 an hour onto the constraint.
Now a supervisor can argue with the reasoning rather than just the conclusion. That’s the difference between a recommendation engine and a colleague.
03
The next task
It costs the walk before it hands out the work.
When a picker finishes, the candidates for their next task are ranked by how far they’d have to walk to reach it — a pick in the same zone costs nothing, a putaway two zones over costs two — and the reason is printed beside each one.
Auto-assign is off unless you turn it on. The floor sees the ranking and the arithmetic; whether the system hands the task out by itself is your call, not a default.
Why this instead of what you have
The report was confident. That was the problem.
Warehouse software is very good at producing a number. What it rarely tells you is how much of the floor that number actually covers — so you find out which parts were guesses when somebody acts on one.
The carton didn’t fit.
Because two items in the order had no dimensions and the system estimated rather than saying so. You learn this at the pack bench, with the order open.
The productivity report was wrong.
Two task types had no standard configured, so the hours went in unscored and quietly dragged everyone’s average down. Someone got a conversation they didn’t deserve.
Nobody knew which zone was the constraint.
So the extra body went where the supervisor had a hunch, and the bottleneck stayed exactly where it was all week.
A floor system that reports its own coverage before it reports its numbers, declines the answers it can’t support, and shows its arithmetic when it wants you to move somebody.
04
Five reasons it wins
Two of these are the system refusing to answer.
02
It reports coverage before it reports numbers.
Seventy-eight percent of completed tasks have a standard. The other twenty-two aren’t scored, and the missing standards are named. You know what the productivity figure is actually made of.
03
Moves are argued from the constraint.
Which zone has slack, which zone is the bottleneck, by how many units an hour, and what the move is worth. A supervisor can push back on the reasoning, which means they can also agree with it.
04
The walk is costed before the work is handed out.
Next tasks ranked by travel, with the reason on each. And auto-assign stays off until you decide otherwise — the floor gets the ranking either way.
05
A clean count still needs a signature.
Book 140, counted 140, variance zero — and it sits in the approval queue anyway. A register that only surfaces the discrepancies is a register that can be gamed by not counting.
05
What’s in it
Six areas, one module.
The desk
6- Seven kinds of floor decision, ranked together
- Every row opens the workbench that closes it
- Putaway by licence plate, zone and location
- Shorts with the reason they came up short
- Variances typed by cause, with the delta
- Age on every row, so nothing quietly sits
Cutoffs
4- Per carrier, per lane
- Lines picked against lines needed
- Picking now, or still awaiting release
- Release or drive it from the row
Holds and counts
6- Holds scoped to an item, a lot or a location
- The role that placed it, on the record
- Damage and QC held separately
- Book against counted, with the delta
- Clean counts still require approval
- Recount without leaving the queue
Labour
6- Units per hour and accuracy, per worker
- Percentage to standard, where one exists
- “No standard” shown, never scored as zero
- Coverage reported before the averages
- Zone, activity, shift and hours worked
- Below-standard surfaced as its own count
The floor plan
5- Throughput by zone, slack against constraint
- Reassignments argued with the arithmetic
- Projected gain on the constraint, stated
- Next task ranked by travel cost
- Auto-assign off until you turn it on
The intelligence
5- Pick paths resequenced, with the saving named
- The locations, in order, so you can check it
- A deadline to apply it before release
- Blockers name the record that fixes them
- Suggest, draft, act and attest, or auto
Thirty-two of them. There are more, but these are the ones a floor supervisor feels on the first shift.
Next step
See it on your floor.
Ninety minutes on your operation: your zones and locations, your carrier cutoffs, which task types have standards and which don’t, and where your constraint actually is. You keep the configuration.
Volta is the ELYON warehouse module. Every capability here is live in the product today. No figure on this page is a customer outcome.