On Monday the closet at your medical office account holds eleven cases of center-pull towel. On Friday it holds three, nobody submitted a supply request, and the driver is already scheduled to drop eight more cases because that is what the standing order says.
Eight cases of paper is not a rounding error. On a mid-size account it is the difference between the supply line you bid and the supply line you are actually running, repeated every month for the life of the contract.
Barcode scanning converts supply movement into dated transactions. You scan an item when it is received, issued to a building, transferred, or discarded, and the system keeps a running balance for every closet. That balance is what lets you set par levels, catch shrink in a week, and bill consumables at actual cost.
What follows is the build order. Seven stages, each with a defined input, a defined output, and a condition you have to satisfy before you move on. Run them in sequence and you finish with one number: consumable cost per account per month, backed by scans instead of memory.
The build, stage by stage
| Stage | Input | Output |
|---|---|---|
| 1. Scope | Purchase history, closet walkthrough | Master item list split into scanned and unscanned |
| 2. Symbology and label | Item list, closet conditions | Written labeling standard |
| 3. Hardware | Labeling standard, crew phone inventory | Scanner decision and one test scan per item type |
| 4. Baseline | Physical count sheets, 8 to 12 weeks of usage | Opening balances plus reorder point and par per item per closet |
| 5. Transactions | Route schedule, who is on site when | Four scan events assigned to named roles |
| 6. Reconciliation | Scan ledger, monthly physical count | Variance in units and dollars, with a threshold |
| 7. Costing | Consumption data, unit costs, bid assumptions | Cost per account per month and per 1,000 sq ft |
Stage 1: Decide what gets a barcode and what never should
Pull twelve months of distributor invoices and sort every line by extended annual spend. In most janitorial operations a short list of items carries the majority of the dollars: towel, tissue, can liners, hand soap, and one or two disinfectant concentrates.
Barcode those first. The 40 items at the bottom of the list, the toilet brushes and the odd bottle of stainless polish, can wait for phase two or stay off the system permanently.
Then split the list into two categories, because they behave differently:
- Consumables: paper, liners, soap, chemical. Scanned in and out. The balance drops to zero and gets replenished.
- Assets: backpack vacuums, autoscrubbers, burnishers, propane buffers. Scanned as a transfer between locations or people. The balance never drops, the assignment changes.
Do not mix them in one list. An asset tag answers "where is it and who has it," a consumable barcode answers "how many are left." Those are different reports and different label types.
Move on when: you have a written item list with a unit of issue defined for every line. "Case" and "roll" are not interchangeable, and picking one per item now prevents months of nonsense math later.
Stage 2: Pick a symbology and a label that survives a wet closet
Symbology just means the barcode format. Most cleaning operators only need to make two decisions: use the manufacturer's existing barcode where it works, and pick one internal format for everything else.
| Symbology | What it holds | Use it for | Watch out for |
|---|---|---|---|
| UPC-A / EAN-13 | 12 to 13 digits | Sealed manufacturer cases and bottles | It identifies the product, not your unit of issue. A case and an inner pack can carry different codes. |
| Code 128 / GS1-128 | Variable length, letters and numbers | Your own item and closet labels | Needs horizontal room. Short codes keep labels small. |
| Code 39 | Letters and numbers, low density | Legacy scanners already in the building | Wide labels for very little data. |
| QR Code / Data Matrix (2D) | High capacity in a small square | Equipment asset tags, small bottle labels | Requires a camera or 2D imager. A 1D laser scanner will not read it. |
Now the label stock, which is where most janitorial rollouts fail. A paper label printed on a desktop thermal printer will curl off a chemical drum within a month and turn to pulp on anything stored near a mop sink.
| Where the label lives | Facestock | Adhesive |
|---|---|---|
| Dry stockroom shelf, closet door | Coated paper | Standard permanent acrylic |
| Janitor closet, splash zone, dispenser cabinets | Polypropylene or polyester | Aggressive permanent acrylic |
| Equipment wiped down with quat disinfectant | Polyester with overlaminate | High-tack, textured-surface rated |
| Chemical secondary containers | Polyester, chemical resistant | High-tack, placed clear of the GHS label |
One legal note that matters more than any of the above. Under the OSHA Hazard Communication Standard, shipped containers and workplace containers of hazardous chemicals must carry the required labeling, and your barcode is not a substitute for it.
Never apply an asset tag over a GHS pictogram, a signal word, or an EPA registration number on a disinfectant. Put the barcode on the shoulder, the cap, or the opposite panel.
Move on when: you have printed a test label in each material, stuck it in the worst closet you service, and read it successfully after two weeks of real conditions.
Stage 3: Phone camera or Bluetooth imager? Choose the scanner
Crews already carry phones, and a modern phone camera reads both 1D and 2D codes well in decent light. That is the cheapest possible start and it is the right start for most operators under about 30 accounts.
Where phones struggle is volume and lighting. Receiving a pallet of 60 cases on a loading dock at 6 a.m., scanning a code at arm's length under a single fluorescent tube, or scanning with wet gloves on: that is where a dedicated 2D imager pays for itself in minutes saved per receipt.
- Phone camera: zero hardware cost, works for closet issues and spot counts, slower on bulk receiving, dependent on the crew member's own device and battery.
- Bluetooth 2D imager: one per warehouse or per supervisor vehicle, fast trigger-pull scanning, reads damaged and low-contrast labels, requires charging and a place to live.
- Rugged handheld terminal: only worth it if you run a real warehouse with daily pick and pack. Most janitorial operations do not.
A practical split: phones in the field, one imager at the warehouse. Buy the imager after you have proven the process, not before.
Move on when: every item type on your master list has been scanned successfully at least once with the hardware you actually plan to deploy.
Stage 4: Run the baseline count and set par levels
Barcode data is worthless without a correct opening balance. Pick a date, shut down issues for a few hours, and physically count every barcoded item in every closet and in the warehouse.
Two people per closet. One counts, one records. Count by unit of issue, exactly as you defined it in Stage 1.
Then build par levels from actual usage, not from what the last supervisor felt was about right. The formula:
- Reorder point = (average weekly usage x lead time in weeks) + safety stock
- Par (maximum on hand) = reorder point + one delivery cycle of usage
Worked example: Riverbend Medical Plaza
Assume a 60,000 sq ft medical office building cleaned five nights a week. Over the last ten weeks the site has consumed an average of 6 cases of center-pull towel per week. Your distributor lead time is 10 days, which is 1.5 weeks. You want a half week of safety stock, which is 3 cases.
Reorder point = (6 x 1.5) + 3 = 12 cases. With a weekly delivery cycle, par = 12 + 6 = 18 cases.
That number is checkable. If the closet holds 22 cases, you are financing paper that a supervisor cannot see behind. If it drops below 12 and no order fired, your reorder trigger is broken.
Move on when: every barcoded item has an opening quantity, a unit cost, a reorder point, and a par, recorded per closet and not just company-wide.
Stage 5: Write the four scan events into the nightly route
There are only four transactions worth tracking. Assign each one to a named role and a specific moment in the week, or it will not happen.
- Receive. Input: distributor delivery. Output: quantity added to warehouse or closet balance. Whoever signs the packing slip scans it, at the dock, before the driver leaves.
- Issue. Input: stock moving from warehouse to a building closet, or from closet to cart. Output: quantity deducted from the source and added to the destination. Scan the closet label first, then each item.
- Transfer. Input: a case or a piece of equipment moving between accounts. Output: balance and custody change. This is the transaction most operations skip, and it is why balances drift.
- Waste or damage. Input: a punctured liner case, a spilled drum, a burned-out pad. Output: quantity removed with a reason code. Without this event, every loss looks like theft.
Keep the scan itself under ten seconds. If a cleaner has to open a form, choose a location from a dropdown, and type a quantity, adoption dies in week three. Scan closet, scan item, tap quantity, done.
Go-live checklist for the first two accounts
- Closet label printed and mounted at eye height, inside the door
- Every barcoded item labeled on a panel that faces the aisle
- Opening balance entered and signed off by the supervisor
- Reorder point and par set for each item
- Four transaction types demonstrated by the crew, not just explained to them
- Backup process written down for the night a phone dies
- One person named as owner of the count for that building
Move on when: you have 30 straight days of scans on the pilot accounts with no more than a handful of missing transactions per week.
Stage 6: Reconcile the closet and put a dollar figure on the variance
Once a month, physically count the pilot closets and compare to what the system says. The gap is your variance, and it is the entire point of the exercise.
Continuing the Riverbend example with illustrative unit costs: the ledger says 13 cases of towel on hand at $42 per case, the physical count finds 11. That is a 2 case shortfall worth $84.
Express it as a rate: variance % = variance value / value of goods consumed in the period. If total variance across the closet is $131 and consumption was $1,574, you are at 8.3%.
A first month in the high single digits almost always means process, not people. The usual culprits are unscanned transfers between accounts, a receipt entered as a case when the delivery was an inner pack, and waste that nobody logged.
Fix those three, run another cycle, and watch the rate fall. Many operators set an internal investigation threshold and hold it: pick a number you can defend, communicate it, and count the closets that exceed it more often.
Move on when: variance has stabilized and you can explain, item by item, what caused the remaining gap.
Stage 7: Produce the cost-per-account number and check it against your bid
Now the payoff. Consumption for the period is a simple identity: opening balance + received - closing balance = consumed. Multiply by unit cost and you have real supply cost for that building, not an allocation.
Riverbend Medical Plaza, one month, all figures illustrative:
| Item | Opening | Received | Closing | Consumed | Unit cost | Extended |
|---|---|---|---|---|---|---|
| Center-pull towel (case) | 18 | 10 | 14 | 14 | $42 | $588 |
| 2-ply bath tissue (case) | 12 | 9 | 12 | 9 | $58 | $522 |
| Can liners (case) | 9 | 7 | 9 | 7 | $34 | $238 |
| Foam hand soap (case) | 4 | 3 | 4 | 3 | $46 | $138 |
| Quat concentrate (case) | 2 | 1 | 2 | 1 | $88 | $88 |
| Total | $1,574 |
Two derived figures matter. Cost per 1,000 sq ft per month is $1,574 / 60 = $26.23. That is the number you carry into your next bid for a similar medical office.
The second is the gap. If you bid this account assuming $1,050 per month in consumables, you are running $524 over, which is $6,288 a year on one building. You cannot negotiate that back from a distributor, and you certainly cannot find it in a monthly invoice total that lumps six accounts together.
That gap is the checkable outcome of this whole build. Sixty days from your baseline count you should be able to state, per account, what supplies actually cost and how far off your bid assumption was.
Frequently asked questions
Do I need a barcode on every bottle, or just the case?
Label at your unit of issue. If you order and issue by the case, barcode the case and let the bottles inside ride along. Break the case open into a closet stock of individual bottles and the case barcode stops matching reality, which is exactly when you need an internal Code 128 label for the bottle as its own item.
Can I use the manufacturer's existing UPC instead of printing my own labels?
Yes for sealed cases of a product you buy in one consistent pack size, and it saves real labeling time. Two cautions: pack sizes change without warning when a distributor substitutes a brand, and a UPC does not identify which closet the item belongs to. You still need location labels of your own.
Does barcode scanning replace the physical inventory count?
No. It reduces how often you need a full count and makes cycle counting practical. A running scan balance lets you count your top five items monthly and everything else quarterly, instead of shutting down for a full wall-to-wall count. Without at least one physical count to reconcile against, the ledger is unverified.
How long does it take to barcode an existing supply room?
Plan the labeling in item types rather than pieces. Once labels are printed and the item list is finalized, a two-person team can usually label and baseline-count a single-building janitor closet in one shift. A central warehouse with a few hundred SKUs is a weekend project, not an afternoon.
Who should own the scan: the cleaner or the supervisor?
The person physically moving the goods scans them, because anything else is data entry from memory. The supervisor owns the monthly reconciliation and the variance explanation. Splitting it this way keeps scanning fast at the closet door and keeps accountability with someone who has time to investigate.
Where CleanTrack360 fits
CleanTrack360 does not scan barcodes. What it does handle is the layer around your inventory system: supply requests submitted from the field, location-based team messaging so a closet shortfall reaches the right supervisor the same night, bulk CSV import for locations so your building list matches your closet list, and reports with CSV export so you can join consumption data to the accounts it came from.
If you are also running scheduling, geofenced GPS clock-in and clock-out in the phone browser, quality inspections with photo evidence, and a client dashboard, plans start at $99 per month for up to 5 team members, $199 for up to 20, and $249 for up to 50. Pricing is per plan, not per user, and there is a 14-day free trial with no credit card.