Why RFID cycle counts are faster than manual counts, and more frequent
A manual count, by hand or with a barcode scanner, records one item at a time, each label in view. A UHF RFID handheld reads the tags of many items in one pass, through fabric and packaging, so a count becomes a walk along the fixtures.
In a 13-week University of Arkansas study at two Bloomingdale’s stores, published in August 2009, scanning 10,000 items in the men’s and women’s denim departments took two hours with RFID and 53 hours with a barcode reader: an average of 4,767 items an hour against 209, a 96 percent cut in cycle-counting time. Readers and tags have changed since 2009, so treat it as an order of magnitude and time your own count in a pilot.
Because each count takes so little time, a store can count more often, and that is where the accuracy comes from. The same research group’s 2010 JCPenney study describes RFID making frequent cycle counts practical in place of large-scale inventories once or twice a year. Associates counted the RFID test departments twice a week, and the authors conclude that weekly RFID counts used to update the stock ledger improved accuracy even in departments that started with high accuracy, while accuracy in the control stores declined over the 15 weeks.
GS1 US, summarising Auburn University RFID Lab studies in 2018, cites inventory accuracy rising from an average of 63 percent to 95 percent, retail out-of-stocks falling by up to 50 percent and cycle count times cut by 96 percent. These are summary figures from retail studies (an average, a maximum and a count-time cut), not a forecast for your stores; the example uses the accuracy figures only to show the arithmetic.
That is why the calculator asks for the count frequency with RFID separately. The time saved per count is often spent on counting more often, as in both studies, so the labour line can shrink to almost nothing: in the example, 52 RFID counts a year take 104 hours, about as long as 2 barcode counts (106 hours). At today’s frequency the same RFID counts would take 4 hours. When the hours are reinvested, the return comes from what accurate stock is worth, which you enter as the sales uplift and the shrink reduction.
What makes up RFID total cost of ownership
The first-year cost of an RFID inventory programme is more than its tags. The University of Arkansas’s JCPenney paper lists what a pilot should put a figure on: hardware, software, tags, changes to legacy systems, labour reallocation and changes to business processes. The table sorts the costs into one-off and running ones:
| Cost | One-off or running | Supplied by | What sets it |
|---|---|---|---|
| Tags, labels or inlays | Running: every item you tag, each year; plus the stock on hand at the start | The tag maker; Proud Tek makes UHF labels, hang tags and dry and wet inlays | Chip and inlay, format and face stock, printing, encoding and quantity; your quotation sets the price |
| Readers, handhelds and antennas | One-off, with repairs and replacements later | Reader makers, through your integrator | How many handhelds and fixed read points, such as dock doors, fitting rooms and exits |
| Software and integration | One-off setup, often with a yearly subscription or support fee | Your integrator or software vendor | Middleware, inventory software and the connection to your ERP, WMS or point of sale |
| Labour to apply tags | One-off for the stock on hand at the start; running for new items, unless suppliers tag at source | You, or your suppliers under a retailer’s tagging programme | Whether the tag rides on a ticket or label the item already carries |
| Maintenance and support | Running | Your integrator and the reader makers | Device repairs, software support, training new staff |
| Pilot and process change | One-off | You and your integrator | Testing tags on your goods; changing count, receiving and replenishment routines |
The calculator takes the same split. Its one-off cost is the tags for the stock on hand, readers and handhelds, software and integration, and other one-off costs: the labour to tag the stock on hand, installation, the pilot, training and process change. Its running cost a year is the tags for new items and other running costs, such as subscriptions, support, repairs and the labour to tag new items.
Tags are the line that grows with volume: they are bought for every item, every year, while readers and integration are mostly paid once. If suppliers deliver goods already tagged under a retailer programme, those tags are their cost, not yours; enter only the items you tag yourself. For shipments between distribution centres, the Project Zipper study reduced the return to one comparison: the cost of RFID tags plus RFID scanning equipment against the cost of the claims (chargebacks) on that distribution centre’s shipments (GS1 US, 2018).
Where the tag price comes from
For this programme, Proud Tek supplies the tags, labels and inlays; UHF readers, handhelds, software and integration come from your system integrator. We publish no tag price list, because the price depends on the chip, inlay, face stock, printing, encoding, quantity and trade term: How to Read an RFID Tag Quotation, Line by Line explains each line of a quotation. For what else an inventory programme needs, from read points to the pilot plan, see RFID Inventory and Asset Tracking for Warehouses and Distribution Centres.
How the calculator works
Every result is one of these formulas, and the result panel repeats them with your numbers. Use one currency for every amount; results are rounded to two decimals.
- Counting hours a year today = sites × labour hours per count × counts a year.
- RFID hours per count = the hours you enter, or today’s hours per count ÷ the speed-up factor.
- Counting hours a year with RFID = sites × RFID hours per count × counts a year with RFID (today’s frequency if left blank).
- Labour saving a year = (hours today − hours with RFID) × loaded labour cost per hour. It is negative when RFID counts are frequent enough to take more hours than today.
- Sales benefit a year = annual sales × sales uplift × gross margin.
- Shrink benefit a year, at cost = annual sales × shrink today × shrink reduction, × (1 − gross margin) when the shrink rate is measured at retail value.
- Tags at the start = items tagged at the start × tags per item × price per tag; tags a year = items tagged a year × tags per item × price per tag.
- One-off cost = tags at the start + readers and handhelds + software and integration + other one-off costs.
- Running cost a year = tags a year + other running costs.
- First-year cost = one-off cost + running cost a year.
- Net benefit a year = labour saving + sales benefit + shrink benefit − running cost a year.
- Simple payback = one-off cost ÷ net benefit a year × 12 months, with the full benefit from the first month. There is no payback when the net benefit a year is zero or negative.