Answers for district IT

How do schools run year-end device collection?

Year-end collection is the operation that determines whether your inventory is real. Every custody record you kept during the year gets tested in a two-week window, and whatever you fail to reconcile in June becomes a discrepancy you carry into the next audit. The districts that do this well treat it as a scheduled logistics exercise with a named owner per building, a fast station-based check-in, and a written escalation path for the devices that do not come back — not as a week when the technology department stands in a hallway hoping.

Decide first whether you are collecting at all

Collecting every device every June is a default, not a requirement, and it is worth re-examining. The arguments for collecting are real: you get a physical count that reconciles your records, a chance to repair damage before it compounds, a summer window to reimage and update, and devices that are not sitting in hot cars for ten weeks. The arguments against are equally real: a full collect-and-redeploy consumes several weeks of staff time twice, it interrupts summer learning access, and it produces exactly the handling that causes the damage you are trying to prevent.

The common middle position is to collect from students who are leaving — graduating seniors, transfers, anyone changing buildings — and to let continuing students keep their assigned device through the summer. That preserves the custody chain for most of the fleet, cuts the volume dramatically, and focuses your effort on the population where the device would otherwise become unrecoverable.

If you keep devices out over summer, you still need a verification event. An attestation campaign where families confirm possession, or a short check-in window where students bring the device to be scanned and handed straight back, gives you the count without the storage and redeployment cost. The point of collection is the reconciliation, and the reconciliation does not strictly require the device to change hands.

Whichever model you choose, decide it in March and publish it, because principals, summer school staff, and families all plan around the answer. A decision made in May is a decision made too late to communicate.

Scheduling, staffing, and the physical setup

Collection volume is concentrated and unforgiving: a high school of two thousand students cannot pass through one station in a day. Work backward from throughput. A practiced two-person station — one scanning, one inspecting — moves a student through in under a minute when nothing is wrong, so plan station count from your enrollment and the number of collection days you can get, and then add a station, because the estimate is always optimistic.

Schedule by period or by homeroom rather than opening a general window. Named blocks give you predictable arrival rates, they make no-shows immediately visible as a list rather than as a vague sense that some people did not come, and they let teachers hold the class accountable. Seniors should be collected before the last week, ideally tied to a checkout process they already have to complete.

Stage the room so the line only moves one direction: students arrive with device, charger, and case; the scanner reads the asset tag; the inspector checks condition and completeness; damaged or incomplete units divert to a second table so they do not stall the line. Have bins and carts labeled before the first student arrives, and decide up front where devices physically go, because a pile in the corner of a library becomes an inventory problem by itself.

Staff it with more than the technology department. Collection is a whole-building activity, and the buildings that run it well borrow paraprofessionals and student aides for the scanning role while a technician handles inspection judgment calls. The technology department's scarce resource is the person who can decide whether a hinge is normal wear or a chargeable repair.

The check-in station: scan, inspect, decide

The station does three things per device, and the software should present them as one screen. Scan the asset tag to identify the device and pull up who it is assigned to. Record condition, including photos of any damage. Close the custody record so the device is no longer showing as in the student's possession.

Completeness matters as much as condition. Charger, case, and any stylus or accessory issued with the device are the items most often missing, and they are cheap individually and expensive at fleet scale. Make them explicit checkboxes on the intake screen rather than something the inspector is expected to remember, and give the accessory its own fee line if your policy charges for it.

Condition assessment needs a short shared vocabulary or it will not be consistent across buildings and inspectors. Something like good, minor wear, damaged, and non-functional, with a photo required for the last two, is enough. Compare against the condition recorded at issue — that comparison is the entire basis for a defensible damage fee, and it only exists if someone captured condition when the device went out in August.

Damaged devices should enter the repair queue at the station, not later from a pile. Scan, mark the symptom, and it is already a repair record with a device, a photo, and a date. The most common way a June collection turns into a September crisis is a cart of damaged devices nobody triaged until the week before school started.

The reconciliation view is what you actually came for: for each building, what was expected, what was scanned, and what is missing. If your check-in produces that list live, you can chase during the collection window instead of after it.

Chasing what does not come back

There will be a tail — typically a small percentage of the fleet, concentrated in a few buildings and a few grade levels. Work it while school is still in session, because your leverage collapses the moment students leave for the summer, and it collapses permanently for graduating seniors.

Escalate on a written ladder with dates attached, so that it is a district process and not a personal negotiation between a secretary and a family. A first notice home, a second notice with the replacement cost stated, a hold on whatever your district can legally hold, and finally a lost-device determination that assesses a fee. Know your state's rules before you write it — the ability to withhold records, transcripts, or participation in activities over an unreturned device varies by state and is genuinely restricted in some.

Separate real loss from records error, because the two look identical on the missing list and have completely different resolutions. A device that shows in the Google Admin console as having synced last week is not lost; it is with someone. Comparing last sync time and last known user against your missing list resolves a meaningful share of the tail before you send anyone a bill.

For devices genuinely gone, close the record with an outcome rather than leaving it open. Lost, stolen with a police report attached, or written off with a date and an approver. An open unresolved custody record is the thing that makes next year's count wrong, and an asset that quietly disappears from the inventory is indistinguishable from an asset that was stolen — which is exactly the distinction an auditor will ask you to make.

Summer work and the August redeployment

Collection is only worth the disruption if you use the summer window. The device work is straightforward: repair the damaged units while parts lead time does not matter, update and reimage, verify enrollment and policy, and physically clean. Do the repairs first — parts ordering in June means devices ready in July, while parts ordering in August means a September backlog.

Summer is also when the fleet-level decisions get made with real data. You now know how many devices you have, how many are beyond economical repair, how many fall past their auto-update expiration before next June, and therefore how many you actually need to buy. That count is far more trustworthy in July than the one you estimated in the spring budget.

Stage redeployment by building and by roster before students arrive. Rosters change over the summer — students move, transfer, and enroll — so the assignment list should be built from a current SIS sync in August rather than from the June list. Districts that pre-assign in June spend the first week of school undoing it.

The August handout is the mirror image of the June station and deserves the same design: scan the tag, confirm the student, capture condition at issue, and get the signed agreement attached to the checkout. Condition captured now is what makes next June's damage assessment defensible, and the signature captured now is what makes the fee collectible. Skipping either one saves ten seconds in August and costs the district real money the following spring.

Common questions

Should we collect Chromebooks at the end of every school year?

Not necessarily. Many districts now collect only from students who are leaving — graduates, transfers, and building changers — and let continuing students keep their assigned device through summer. That preserves the custody chain, cuts volume sharply, and focuses effort where devices would otherwise become unrecoverable. If you keep devices out, replace collection with a verification event such as an attestation campaign or a scan-and-return window.

How long does year-end collection take?

It depends on station throughput, not on fleet size directly. A practiced two-person station processes a clean return in under a minute, so plan station count and days from enrollment, then add one station — the estimate is always optimistic. Scheduling by homeroom or period gives predictable arrival rates and makes no-shows visible as a list.

What should we check when a device comes back?

Identity by scanned asset tag, condition with photos of any damage, completeness of charger, case, and accessories, and the custody record closed. Compare condition against what was recorded at issue — that comparison is the entire basis for a defensible damage fee, and it only exists if condition was captured at handout.

What do we do about devices students never return?

Work the tail while school is in session, then escalate on a written ladder with dates: notice home, second notice stating replacement cost, whatever hold your state law actually permits, and finally a lost determination with a fee. Check state rules first — withholding records or transcripts over an unreturned device is restricted in some states.

How do we tell a lost device from a records error?

Check last sync time and last known user in your management console against the missing list. A device that synced last week is with someone rather than lost, and that check typically resolves a meaningful share of the tail before anyone gets billed.

What should happen over the summer?

Repair first, because parts lead time is free in June and expensive in August; then update, reimage, verify enrollment, and clean. Use the resulting real count of working devices to size the purchase order, and build the August assignment list from a fresh SIS sync rather than from June's roster.

How Chalk approaches this

Chalk runs collection as scan-to-reconcile: scanning a building's returns produces the found, misplaced, and missing lists live, so you can chase during the window rather than after it. Check-in captures condition and photos against what was recorded at issue, damaged units drop straight into the repair queue with a symptom, and lost or stolen outcomes are recorded with police-report capture so the write-off has a paper trail. Attestation campaigns cover the districts that keep devices out over summer, and phone-camera scanning means an extra station costs nothing but a person.

See the full feature catalog →

Run this playbook on your own fleet.

Chalk installs to a populated inventory from your SIS and Google Admin in about 30 minutes. Self-host free forever under AGPL-3.0, or let us host it — priced by fleet size, never per seat.