A Comparison Table for the Calculators This Site Ships
Data this note rests on: Nine instruments, split by what they actually do: five settle a cost, three read a record, and one places a week on a calendar, with every rate on the cost side editable and two stated data gaps left unfilled.
Side-by-side table
Nine instruments ship on this site, and they divide into three jobs rather than nine. Five settle a cost: chargeable weight, agent fee comparison, parcel merge, insurance and landed cost. Three read a record: link health, size conversion with its evidence table, and the QC checklist. One places a plan on a calendar: the shipping window calendar. The table below lists all nine with their inputs, their outputs, whether their rates are editable, whether they are original to this site, and the honest limit that applies to each.
| Instrument | What you enter | What you get | Editable | Original | Stated limit |
|---|---|---|---|---|---|
| Chargeable weight /instrument/chargeable-weight/ | Dimensions, actual weight, divisor, rate per kg | Volumetric weight, billed weight, and the extra cost that volume adds | Yes | No | Applies no rounding rule, because lines round differently. |
| Agent fee comparison /instrument/agent-fees/ | Item subtotal, item count, domestic rate, exchange rate | Totals per agent, with unverified fee fields excluded | Yes | Yes | An unpublished fee is excluded rather than counted as zero. |
| Link health /instrument/link-health/ | A link | Platform, item identifier, tracking parameters, link form | Not applicable | Yes | Makes no request, so it never reports whether a listing resolves. |
| Size conversion and evidence /instrument/size-evidence/ | A size and the scale it is printed on | Equivalent sizes, plus a measured insole evidence table | Not applicable | No | The evidence table is empty: no brand has reached a publishable measured sample. |
| Insurance decision /instrument/insurance/ | Declared value, premium, cover limit, excess, loss band | Expected loss with and without cover, and the break-even value | Yes | No | No measured loss rate is published, so the loss band is your input. |
| Shipping window calendar /instrument/shipping-window-calendar/ | A planned order week | Where that week falls against five annual windows, with three recommendations | Not applicable | Yes | Structural rows are calendar facts; no observation is held for the current period. |
| Parcel merge decision /instrument/parcel-merge/ | Item weights, line weight limit, rate per kg, base cost | One parcel against two, with both totals | Yes | No | Compares weight and cost only, not customs treatment of a larger declared value. |
| Landed cost /instrument/landed-cost/ | Item subtotal, item count, commission, payment rate, freight, declared value | A six-line cost waterfall and the threshold decision | Yes | No | Every rate is an input; your own checkout screen remains the authority. |
| QC review checklist /instrument/qc-review/ | A category | A twenty-item checklist, stored in your browser only | Not applicable | No | A reading aid: it cannot judge authenticity or see what the photos missed. |
- Source:
- Instrument configuration, parameter defaults and boundary text as published on each instrument page of this site.
- Sample:
- n = 9 instruments. Their default values are site constants rather than observations, so no sample size applies to the defaults themselves.
- Recorded:
- 2026-W40
- Known gap:
- Two instruments state an open data gap. The measured size evidence table holds no rows, and the shipping window calendar holds no observation for the current period. Neither gap is filled with an estimate.
The editable column is the one worth checking before trusting a total. Every rate on the cost side is an input, which means the output describes your figures rather than ours. An instrument with four editable rates and one with none are different kinds of tool, and only one of them can be wrong about your order.
What each instrument is good at
The five cost instruments are best understood as a chain rather than a menu. Chargeable weight comes first, because it turns dimensions and a scale weight into a billable weight and prices the difference that volume creates. Parcel merge uses the same weights to compare one parcel against two, which is a decision about the base cost per parcel and about line limits. Agent fee comparison handles the buying side, where commission is a percentage of the subtotal and a fixed per-item charge is a different shape entirely. Insurance prices a bet against a small probability and names the declared value at which covering the parcel starts to pay for itself. Landed cost assembles the six lines in the order they are charged and applies the destination threshold at the end.
The chain matters because two of the five consume another output. Landed cost needs a freight figure, and the natural source is the chargeable weight instrument rather than a guess. Insurance needs a declared value, which is usually close to the item subtotal plus freight, so it sits downstream of both. Running them in the wrong order produces a total built from two guesses that could have been one calculation.
The three reading instruments answer different questions and share one property: they never contact the outside world. Link health reads the structure of a link, identifying the platform, extracting the identifier and flagging tracking parameters added by whoever shared it. Size conversion translates between labelling conventions and carries a measured insole evidence table that stays empty until a brand has a sample worth publishing. The QC checklist turns a glance at warehouse photographs into a review, grouped by what you are looking at, with its state kept in the browser and nothing submitted. None of the three needs a request, an account or a purchase, and none of them claims to know anything about a listing that the record does not contain.
The calendar stands alone among the nine. It holds five annual windows that change dispatch and transit, each with three recommendations: when to order, when to consolidate, and which weeks to avoid. Its rows are labelled as either structural, meaning the holiday calendar that repeats every year, or observed, meaning something recorded with a week and a sample count attached. No observed row is currently shown, and the page says so rather than estimating the disruption from the shape of previous years.
| Question | Why it is out of reach here | Nearest instrument |
|---|---|---|
| Does this link still resolve | Would need a request to the marketplace for every link checked. | Link health, which reads form and explicitly stops there. |
| How long will this parcel take | Would need parcels followed from dispatch to delivery, on more than one route. | Shipping window calendar, which describes windows rather than durations. |
| Is this seller dependable | Would need order outcomes attributed to sellers over time. | None. No seller-level figure is published on this site. |
| Will this parcel be assessed | Would need assessed parcels with their declarations and outcomes. | Landed cost, which applies the threshold rule without predicting the outcome. |
| Where will prices go next | Would need at least a second observation of the same price field. | None. One extraction date exists, so no movement can be measured. |
- Source:
- The boundary statements published on each instrument page, read against the two measurement passes described in the method page of this site.
- Sample:
- No qualifying sample yet. Every row in this table names a measurement that has not been taken, so no sample can be counted against any of them.
- Recorded:
- 2026-W40
- Known gap:
- The nearest-instrument column points at the closest available tool rather than at a substitute. In three of the five rows the closest tool answers a different question, which is stated rather than glossed.
Which one to open first
Order matters more than choice here, because the instruments pass figures to one another. The sequence below follows a single order from a link pasted out of a sheet to a parcel opened at the door, and it puts each instrument where its output is still useful.
- Start with link health for any row you did not write yourself. It answers what the link is, which platform it belongs to and whether somebody added tracking parameters, and it does so without claiming anything about the listing.
- Size conversion and its evidence table come next if the order contains footwear. The conversion block describes what a label means, and the evidence table is empty by policy rather than by neglect, so a tape measure beats both.
- Chargeable weight comes before packing, not after. Dimensions, actual weight and the divisor for the intended line produce the billed weight and the dollar cost of the volume rule, which is the figure the next two steps consume.
- Parcel merge decides one parcel or two while the packing decision is still open, using the same weights and the line limit. It answers a cost question and says nothing about customs treatment.
- Agent fee comparison sits on the buying side, where the question is the shape of a fee rather than its headline rate. A fixed per-item charge grows with item count and a commission grows with value.
- Insurance comes before shipping, and only if the declared value is above the point where cover pays for itself. The instrument names that point instead of assuming that cover is always worth buying.
- Landed cost goes last among the cost tools, because it takes the freight figure from step three and the declared value that step six used, and applies the destination threshold to the result.
- The shipping window calendar is a timing instrument and can be consulted at any point. Its value is highest when the order week is flexible, because two of the five windows are worth moving a plan to avoid.
- The QC checklist arrives when the warehouse photographs do, before consolidation, which is the last point at which a problem can be corrected cheaply.
A reader with a single question rather than a whole order can ignore the sequence and open one instrument. The map is simple: shipping cost points to chargeable weight, order cost points to landed cost, buying options point to the agent fee comparison, link confusion points to link health, timing points to the calendar, and anything about what arrived points to the QC checklist.
Four of the nine instruments accept your own numbers as inputs. Their output is only as good as what goes in, and a default value left in place is an assumption that the page cannot flag for you. The two instruments with stated data gaps, the size evidence table and the shipping window calendar, are the ones where the honest answer is currently that no data exists.
Taken together, the nine cover the arithmetic of an order and the reading of a record, and they stop at the boundary of what can be checked without requests, purchases or outcomes. That boundary is the reason the cost side is fully editable and the reading side is deliberately blind: a calculator can be honest about a total built from your figures, and it cannot be honest about a listing it has never contacted.