The handheld computers get budgeted. Somebody built a five-year total-cost-of-ownership model, compared two vendors, negotiated a service contract, and signed. The cases did not get budgeted. They get bought the way copier toner gets bought — reactively, in small quantities, at whatever the reseller charges, out of a maintenance line nobody reviews. That is the gap that makes rugged PDA and barcode scanner fleet protective case sourcing worth an operations buyer's attention: it is one of the few remaining places in a distribution centre where a five-figure annual line item is being spent without anyone having modelled it.
If you run several thousand handhelds across multiple sites, the question you are eventually asked by finance is simple and rarely well answered: is it cheaper to keep buying the device vendor's accessory, or to tool your own?
Why the accessory spend hides
Three structural reasons, and they compound.
The device is capital, the case is consumables. Handhelds are usually acquired as capex or through a leasing arrangement, with a defined refresh horizon. Cases are ordered against a facilities or MRO code, often per site, often below the threshold that triggers a purchasing review. Nobody consolidates the spend because no single purchase order is big enough to notice.
Damage is recorded against the device, not the accessory. When a unit comes back from the floor with a cracked screen, the ticket says "device damage." It rarely says whether the unit had a case fitted, whether the case was the right one, or whether the case was already torn. Without that field, you cannot prove that accessory spend reduces device spend — so it never gets defended in a budget review.
The refresh cycle resets everything. New device generation, new form factor, old cases scrapped. Any tooling investment has to be justified inside the remaining life of the fleet it fits, which is why the timing of this decision matters as much as the arithmetic. The same logic drives the payment-terminal side of the category, covered in why the Android POS refresh cycle drives case demand.
The number you actually need
Not a market size. Your own damage rate, split by whether a case was fitted. If your service desk cannot produce that split, adding one checkbox to the repair intake form will produce a defensible business case within two quarters — and it costs nothing.
Branded accessory versus your own tooling: how the comparison really works
The instinct is to compare unit prices. That comparison is almost always wrong, because it ignores amortisation, freight, spares depth and the cost of a fit failure across a fleet.
Build the comparison on lines, not on a single price. Fill each line from your own records; the point of the table is that every figure in it is yours, not a benchmark somebody sold you.
| Cost line | Buying the branded accessory | Tooling your own |
|---|---|---|
| Unit price | Reseller list, less whatever volume discount your contract carries | Ex-works factory price at your annual volume |
| Tooling | None | One-time mold cost, amortised across units you will actually order |
| Freight and duty | Usually bundled and invisible | Yours to manage; classification and landed cost become your problem |
| Lead time | Off the shelf until it isn't | Confirm per order; longer on first article, shorter on repeats |
| Spares depth | Whatever the reseller stocks | Whatever you order — you control the buffer |
| Fit risk | Low, it was designed for the device | Yours; a bad cutout is a fleet-wide problem |
| Design control | None | Colour coding by site, logo, strap geometry, thickness where you want it |
| End-of-life exposure | Reseller carries the stock | You carry it — unsold cases for a retired fleet are a write-off |
The break-even is not a number anyone can hand you. It is the point where (branded unit price minus your landed unit price) times the units you will genuinely consume before the fleet retires exceeds the tooling cost plus the cost of carrying inventory. The two variables people get wrong are consumption — they use fleet size instead of fleet size times replacement multiple — and remaining fleet life, which is shorter than it feels once a refresh is already scheduled.
Three questions decide it:
- How many cases does one device actually consume per year in your operation? Pull it from purchase history per site, not from an assumption.
- How many months of fleet life remain before the devices are refreshed?
- Are you willing to hold inventory, or do you need somebody else's warehouse?
If the answer to the third is "somebody else's," own tooling is a harder sell regardless of the arithmetic.
What a rugged case for a scanner has to survive that a phone case does not
Insider details, and each one is a place where a cheap case fails and a well-specified one does not.
The attachment point fails before the case does. Hand straps, tethers, holsters and pistol grips attach to the case body, and the load path runs through a small area of moulded material. In a picking operation the strap is loaded thousands of times per shift. The case body can be perfect and the strap boss still tears out. When you evaluate a case, pull on the strap anchor with the case fitted on the device, not on a loose part. And ask the molder directly whether strap, holster or grip attachments are part of what they supply, or whether the case is a body only — factories that mould cases do not all make the accessories that hang off them.
The charging cradle is a second fit test. Fleet handhelds live in multi-bay cradles overnight. A case that fits the device but not the cradle means a night shift of staff removing cases at charge-in and refitting them at charge-out, which in practice means the cases end up in a drawer. Test in the cradle, and test in a full rack where bay-to-bay clearance is tightest.
The scan window is a no-material zone. Imager and laser scanner apertures have an exit window with a defined clearance. Silicone that intrudes even slightly into the optical path produces intermittent no-reads that get logged as device faults. Same for the camera used for direct-part-mark reads. Any case for a scanning device needs the aperture cleared and the edge chamfered away from the optical path.
Cleaning chemistry eats the wrong material. Cold-chain, food and pharma sites wipe handhelds with quaternary ammonium or alcohol-based agents, several times a day. Different elastomer families respond very differently to that regime — some go tacky, some chalk, some stay stable and just lose printed graphics. Specify the actual cleaning agent your sites use and ask what the factory recommends against it, then ask for a soak sample rather than an opinion.
Colour is an operational tool, not decoration. Fleets that colour-code cases by site, shift or department recover devices faster and reduce inter-site drift. It costs nothing at the moulding stage and is one of the few genuine advantages of tooling your own.
The sourcing path if you decide to tool
The sequence below is what a competent programme looks like from the buyer's side. It is deliberately front-loaded: everything expensive happens after step five.
- Consolidate the spend. Pull twenty-four months of accessory purchase orders across all sites. Most fleet operators discover the total is larger than they assumed and more fragmented than they assumed.
- Fix the fleet picture. Exact device models and variants, quantity per site, months of remaining life, and whether a refresh is already contracted.
- Send a device, not a drawing. A molder can work from 3D data, but a physical unit — ideally two, one to be destroyed in fitting — removes an entire class of variant error. If you have CAD or a scan, send that too.
- Ask what already exists. Before paying for tooling, ask whether the factory has an existing mold for the model, and what the sample turnaround is from it.
- Sample against the real environment. Fit on device, fit in cradle, fit in rack, strap load, scan-window clearance, wipe-down with your cleaning agent, drop from the height your floor actually produces.
- Pilot at one site before fleet-wide rollout. A pilot of a few hundred units in one DC surfaces problems that no lab test does.
- Then tool and order, with the spares buffer sized from your own replacement multiple rather than a round number.
Steps one to five cost time, not capital. Skipping them is how fleet-wide accessory programmes fail expensively — the same pattern the CMH note on hidden costs when sourcing from China describes in a broader context, and the reason MOQ negotiation is easier once you know your real consumption. On the engineering side, cutout and fit tolerance is its own discipline: see port cutouts and fit tolerance on PDA and scanner cases, and for evidence handling, drop-test and ingress evidence a case supplier can actually show.
Common questions
Is tooling our own case sensible for a fleet of a few hundred devices?
Usually not on price alone. It can still be right if you need something the branded accessory does not do — a strap geometry that suits your picking method, colour coding across sites, or a case for a device the accessory market has ignored. Decide on the requirement first; if the requirement is only "cheaper," a few hundred units rarely amortises tooling within the remaining fleet life.
Can one case design cover several device models?
Rarely, and attempts to force it produce loose fits that fail the strap test. What can be shared is the family: common material, common surface texture, common strap hardware, separate mold per model. Ask the factory whether the strap and hardware can be standardised across a family even where the body cannot.
How do we stop a good sample turning into a mediocre production run?
Write the acceptance criteria into the purchase order rather than the email thread: dimensional checkpoints on the cutouts, hardness range, colour reference, and the specific fit tests above repeated on production samples. Agree what happens if a lot misses them before the lot exists.
What to ask a case molder before you commit
- Do you have an existing mold for this exact device variant, and can you sample from it against my unit?
- Do you supply hand straps, holsters or grip attachments, or the case body only — and if the body only, who makes the rest?
- What is your annual capacity for a single high-volume SKU, and what happens to lead time if I double the forecast mid-year?
- What is the minimum order quantity, and how does a second and third colour change it?
- Which elastomer do you recommend against my named cleaning agent, and can you supply a soak sample?
- What fit tests will you run before shipping — device, cradle, rack, strap load, scan window?
- What does the first article inspection report contain, and who signs it?
- Who owns the tooling, where is it stored, and what is the process if the fleet moves to another supplier?
As a reference point for what that supplier conversation looks like against an actual plant: WJM Silicone in Longgang, Shenzhen makes rugged cases for PDA handheld computers, industrial tablets and barcode scanners — reinforced corners, raised bezels for screen protection, precise cutouts for cameras, buttons and I/O ports — from a 12,000 m² site with 200+ staff, a 12-person R&D team, silicone compression and plastic injection lines in-house, and 10,000+ completed mold development projects on file. Payment terms are T/T, confirmed per order. Treat all of it as claims to be evidenced during your own audit, including capacity for the SKU you care about.