Intake Throughput: Receiving, Testing and Listing 200 Phones a Day Without a Warehouse Team

A reseller moving from occasional pallet buys to a standing wholesale relationship hits the same wall at a predictable volume: intake, not floor space, becomes the constraint. Two people grading and listing 40–60 units a week manage fine on an ad-hoc basis. Getting the same small team to receive, test, grade, wipe and list 200 phones a day is not a bigger version of that process — it is a different one, built around fixed stations and handoff rules instead of one person following a unit from box to listing. This is the throughput problem resellers hit as order volume scales, and it is solvable without adding a warehouse crew. It needs a workflow that removes decision points, not more square metres.

Why intake is the bottleneck, not warehouse space

Storage scales linearly: double the stock, double the shelving. Intake does not scale the same way, because every unit has to pass through the same sequence of human judgement calls — does it power on, does the battery read correctly, what grade does the screen get, has the data been wiped — and each of those steps has a fixed time cost per phone regardless of how much shelf space you have. A reseller who adds pallet racking but keeps the same two-person intake bench has bought storage for stock they cannot process fast enough to sell against current demand. The real limit on daily volume is minutes per unit through the stations, multiplied by station count and hours worked — not square metres.

That makes the first planning question not "how much space do we need" but "how many units can move through intake in a working day, and where does that number actually come from."

The four-station model

Splitting intake into four fixed stations, each with a narrow job, is what gets a small team to 200 units a day without chaos. A single generalist handling every step on one phone before starting the next is the slowest possible configuration — it maximises context-switching and means the whole line waits on whichever step that person is slowest at.

• Station 1 — Receiving and IMEI check. Unbox, photograph the shipping condition, scan every IMEI against the GSMA blacklist registry and the purchase order before the unit moves further. A unit that fails here never consumes diagnostic or grading time.
• Station 2 — Functional diagnostics. Battery health, screen digitiser response, cameras, speakers, charging port, buttons and sensors, run through a fixed diagnostic script or app rather than manual spot-checks. This is the station where automation saves the most time per unit.
• Station 3 — Cosmetic grading. A trained grader assigns the cosmetic grade against a written rubric, independent of the functional result from Station 2. Separating cosmetic from functional grading stops one grader's fatigue from affecting both judgements on the same unit.
• Station 4 — Data wipe, cleaning and listing. Certified erasure, a physical clean, and the catalogue entry — grade, battery percentage, storage, photos — created directly from the data captured at Stations 2 and 3, not re-keyed from memory.

A unit flows through all four in sequence, but four different people can be working on four different units at the same time. That parallelism, not faster individual grading, is what turns a two-person ad-hoc process into a team that clears 200 units a day.

Staffing ratios that hit 200 units a day

Rough per-unit timings at a station running smoothly: receiving and IMEI check, 1.5–2 minutes; diagnostics, 3–4 minutes with a scripted tool; cosmetic grading, 2–3 minutes; wipe, clean and listing, 4–5 minutes. Listing is consistently the slowest station because it is the one step that cannot be batched — each unit needs its own photos and its own catalogue entry.

Against an 8-hour day with realistic breaks, one person at a station clears roughly 90–110 units through receiving, 70–90 through diagnostics, 100–120 through grading, and 55–70 through listing. To hit 200 units a day without any station becoming the drag on the rest, that points to roughly two people on receiving and diagnostics combined, one dedicated grader, and two people on wipe-and-listing. Five people, not a warehouse team — but five people each doing one job well, not five people each doing a quarter of every job.

The software layer that prevents rework

The stations above only hold together if data moves with the unit instead of being re-entered at each step. A diagnostic tool that outputs a structured result Station 3 and Station 4 can read directly — rather than a paper slip someone transcribes later — removes the single biggest source of rework in an intake line: a grader or lister having to walk back to Station 2 to ask what a result actually was. Barcode or IMEI-linked records that follow the unit through all four stations, visible to whoever is handling it next, are what let four people work in parallel without a supervisor coordinating by hand. For buyers assessing what to check for at Station 2 and Station 3, see our separate guide to phone testing and quality control — that post covers the criteria; this one covers the throughput built around applying them fast.

Batch sizing and running mixed stock without losing speed

Running one model and storage variant through the line at a time is faster per unit than mixing SKUs, because graders and diagnosticians stay calibrated to one set of known-good reference points instead of recalibrating with every unit. In practice, intake rarely arrives that clean — a single trade-in or buyback lot can mix five iPhone generations and a dozen Samsung models. The practical compromise is batching by brand and generation where the incoming lot allows it, and accepting a small throughput penalty on genuinely mixed lots rather than trying to force an unworkable batching scheme onto stock that does not sort that way.

Order sizing has a direct effect here too: a reseller buying in smaller, more frequent batches gets more consistent daily intake volume than one working through one large quarterly pallet, which makes staffing the four stations predictable rather than feast-or-famine. Our guide to reorder cadence covers the buying-frequency side of this same operational question.

When to add headcount vs when to add a station

The instinct when throughput falls behind incoming volume is to add a person. The better first question is which station is actually the constraint — adding a second grader does nothing if listing is the slow point, and adding a second lister does nothing if diagnostics is backed up. Track completed units per station per day for two weeks before adding headcount; the station with the lowest daily output relative to its target is where the next hire or the next diagnostic tool should go, not wherever feels understaffed.

Scaling past roughly 300–400 units a day on this model usually means doubling specific stations rather than the whole line — a second grading station is cheaper and faster to bring online than a second full line, and keeps the handoff discipline that makes the four-station model work in the first place.

Current graded, tested and wiped stock from SmartChoice is available at shop.smartchoice.ee/stock, sourced in batches sized to match a predictable intake rhythm rather than one unmanageable pallet.

FAQ

How many phones can a small team realistically process in a day?

With four dedicated stations — receiving, diagnostics, grading, and wipe-and-listing — a team of five can typically clear 180–220 units in an 8-hour day. The exact number depends on how automated the diagnostic step is and how mixed the incoming stock is by model.

What's the single biggest throughput killer in phone intake?

Re-touching the same unit more than once — a lister who has to go back and ask what the diagnostic result was, or a grader re-checking something Station 1 should have already confirmed. Data that follows the unit through each station, rather than being re-entered or re-asked for, removes most of this.

Should one person handle a phone from unboxing to listing, or should stations be split?

Split stations outperform a single generalist per unit at any meaningful daily volume. One person following each phone end to end maximises context-switching, and the whole line's speed is capped by that person's slowest skill. Four people each doing one job in parallel is consistently faster for the same headcount.

Is it worth investing in diagnostic software, or can manual checks keep up at 200 units a day?

At 200 units a day, scripted diagnostics pay for themselves quickly. Manual spot-checks are slower per unit and produce results that are harder to pass cleanly to the grading and listing stations, which is where most rework originates. The diagnostics station is usually the best place to automate first.

How does order batching affect intake throughput?

Running one model and storage variant through the line at a time is faster than mixing SKUs, because staff stay calibrated to one set of reference points. Buying in smaller, more frequent batches — rather than one large quarterly pallet — also keeps daily intake volume predictable, which makes staffing the four stations consistent rather than alternating between idle and overloaded.