ShipBob review — good at its job, and its job stops at the Amazon fee

What ShipBob is, and what it actually publishes

Everything in this section was read from ShipBob's own site on 20 August 2026.

From their about page: founded in 2014 by Dhruv Saxena and Divey Gulati, with "50+ Fulfillment centers across the world", "200+ Retail channels supported", a "99.97% Accuracy rate in fulfilling orders", and "1,000s of brands partner with ShipBob". They name their investors openly — Y Combinator, Bain Capital, Menlo Ventures, SoftBank and Hyde Park Venture Partners. The company is privately held.

From their pricing page, the fee structure is named but no figures are attached. Standard fees cover implementation, receiving inventory, warehousing products, and "picking, packing, and shipping each order". Software access is included for all customers, and standard packaging — brown boxes, poly mailers, tape and labels — is included. Custom kitting is "at an additional cost", B2B, wholesale and EDI orders are "at an additional cost", returns management is charged as "a flat-fee", and custom packaging storage carries fees. The page states plainly that "all quotes are customized for each customer" and that "shipping costs are variable in nature, based on weight, dimensions, destination, service, and more".

Say that neutrally: ShipBob quotes on a call. For physical fulfilment that is more defensible than it is for software, because the cost genuinely does depend on your cube, your weight, your order profile and where your customers live. Two brands with identical revenue can have fulfilment costs that differ by half.

It is also, for the avoidance of doubt, our own position. Dr. Stock publishes no price either. Cin7 publishes three exact monthly figures and Sellerboard publishes its tiers. On pricing transparency both of them are ahead of ShipBob and ahead of us, and that is worth conceding before anything else on this page is read as a sales argument.

What ShipBob is genuinely good at

Four things, and a customer would recognise all four.

Software that is actually part of the product. Most third-party logistics providers hand you a warehouse and an email address. ShipBob's differentiator from the beginning has been a merchant-facing platform — inventory visibility, order status, distribution analytics — included rather than sold separately. If you have ever run fulfilment through a provider whose stock count arrives as a weekly spreadsheet, this is not a small difference.

Onboarding speed and channel coverage. 200+ supported retail channels means the integration you need probably already exists, and connecting a store is generally a short exercise rather than a project. For a growing DTC brand that has outgrown a spare room, this is the single most valuable property of the whole offer.

Distributed inventory. With 50+ facilities, splitting stock across regions to shorten the average shipping zone is available as a standard capability rather than a bespoke arrangement. Zone reduction is one of the few genuine levers on parcel cost, and it is a real reason to prefer a network over a single warehouse.

They publish an accuracy figure at all. Very few fulfilment providers put a number on their own error rate where anyone can read it. Whatever you conclude about the number, publishing it invites the comparison, and that deserves credit.

What none of that tells you is what it will cost, which is the next problem to solve.

The 99.97% figure, worked out per order

Published accuracy rates are almost always quoted and almost never converted. Convert this one.

99.97% accuracy means 3 errors in every 10,000 orders — or, put the way an operator feels it, on average one order in every 3,334 goes wrong.

Now scale it to real businesses:

• At 500 orders a month, that is one error roughly every seven months.
• At 2,000 orders a month, about 0.6 errors a month — seven a year.
• At 10,000 orders a month, three a month — 36 a year.
• At 30,000 orders a month, nine a month — 108 a year.

Read one way, that is an excellent operational number and better than a lot of in-house warehouses achieve. Read the other way, a brand shipping 30,000 orders a month will have roughly 108 customers a year receiving the wrong thing, each of whom needs a replacement, a refund or both, and some of whom write about it.

Neither reading is dishonest and both matter. The useful consequence is a question rather than a verdict: who pays for an error, and how is it resolved? Ask specifically whether mis-picks are credited, whether the replacement unit's cost is borne by the provider, whether the reshipment is free, and what the resolution time is. Get it in the agreement, not in the sales conversation. A 0.03% error rate with the cost falling on you is a materially different purchase from a 0.03% error rate with the cost falling on them.

And ask what the denominator is. An accuracy rate measured on lines picked, on orders shipped, or on orders delivered are three different numbers, and the difference between them is larger than the difference between most providers.

The quote you cannot compare, and the protocol that makes it comparable

Custom quotes are not a problem in themselves. They become a problem when two providers structure them differently and you try to compare a total against a total.

ShipBob names four standard cost components — implementation, receiving, warehousing, and pick-pack-ship — plus separately priced kitting, B2B and EDI handling, returns and custom packaging storage. Only one of those genuinely has to vary with the destination: the outbound parcel cost. The others are, or can be, fixed rates you can put side by side.

So here is the protocol. Before any call, build a ten-row table of your actual top ten SKUs by unit volume, with real dimensions and weights, real monthly order counts, and your real customer geography split by region. Then ask every provider to quote against that table, itemised as:

• Receiving — per pallet, per carton, or per unit, and which applies to your inbound profile.
• Storage — per bin, shelf, pallet or cubic foot per month, and how a partially-used unit is charged.
• Pick and pack — for a one-item order, and the incremental cost of each additional item.
• Outbound parcel — as a rate table by zone and weight break, not a blended average.
• Returns — the flat fee, and what it covers.
• Implementation — one-off, and what happens if you leave inside twelve months.
• Minimums — monthly minimum spend, storage minimums, and any long-term storage surcharge with its day threshold.

That last one is where most of the surprise sits, and it is the direct analogue of Amazon's own aged-inventory rules. Ask the day count and the rate, and ask it of every provider.

Then, when the quotes come back, calculate one number for each: fully landed cost per order, for your own SKU mix, at your own volume. That is the only figure that compares. A provider that looks 8% cheaper on pick-and-pack can be 15% more expensive per order once storage minimums and zone mix are in.

Apply the same discipline to contract shape. Term length, notice period, whether rates are fixed for the term, and what written notice you get before a rate change — get all four in writing. From any provider, including us.

Reviews of 3PLs split hard, and here is how to read them

Reviews of any fulfilment provider divide more sharply than reviews of software, and there is a structural reason worth understanding before you weigh a single one.

A software review is mostly about a product that behaves identically for everybody. A fulfilment review is about a specific building, a specific team and a specific SKU profile. Two brands using the same provider, in different facilities, with different products, will have genuinely different experiences and both will be telling the truth. That is why the ratings look bimodal, and it is not evidence of anything about the provider's quality overall.

We are not going to quote star ratings or review counts here. The major review platforms block automated reading, and search-result snippets of those pages have contradicted the live pages in our own checking more than once. Printing a number we could not open ourselves would be the sort of error this page exists to help you avoid.

What to do instead:

• Ask which facility you will be in, and get references from brands in that specific building — not from the company overall.
• Find a reference with a similar SKU profile. Heavy, fragile, multi-piece, temperature-sensitive and apparel-sized goods behave completely differently in a warehouse.
• Date every review you read. Networks expand, facilities open and close, and account teams change. A two-year-old review may describe a building you will never see.
• Ask what the escalation path is, by name, and ask a reference how long a genuine problem took to resolve.
• Ask about peak. November and December are when fulfilment either works or does not, and a reference's answer about last peak is worth more than any average rating.

None of that is hostile to ShipBob. It is the diligence any brand should run before handing a third party physical custody of its inventory, and we would expect it to be run on us.

What a 3PL does not touch: the Amazon-side economics

Here is the boundary, stated flatly. We do not fulfil orders. Dr. Stock is not a 3PL, not a warehouse, not a system of record and not an ERP. If you need someone to hold and ship your inventory, you need a provider like ShipBob and nothing on this page changes that.

But the reverse boundary matters just as much and is stated far less often: a 3PL cannot fix the money you are losing inside Amazon, because those units are in Amazon's buildings under Amazon's rules. Five specific places, all of them Amazon-side:

• Size-tier misclassification. Amazon assigns a fulfilment size tier from its own recorded dimensions. If those are wrong, you pay the wrong fee on every unit indefinitely — and nothing in your fulfilment provider's system can see it, because their measurements are not the ones Amazon is billing from.
• The 181-day threshold. Amazon's FBA page states that units held in a fulfilment centre beyond 181 days attract an aged-inventory surcharge, monthly, on top of ordinary storage. That is a per-unit calendar problem, and it is the moment a slow SKU converts from an asset into a recurring cost.
• Claim windows. Lost and damaged units, shipment discrepancies and fee errors are recoverable, and the windows are short — commonly sixty days from the triggering event. Identified late is not paid.
• The 2025 reimbursement change. Amazon's updated FBA inventory reimbursement policy took effect on 31 March 2025, moving valuation toward manufacturing and sourcing cost. That shrank the recoverable pool for every recovery service in the market, and it dates every reimbursement comparison written before it.
• Cash trapped in dead SKUs, and the removal-versus-liquidation decision that follows.

An example of how small the first of those can be. On a fishing tackle brand, a product weighing 1.119 lb was billed as 2.00 lb, because its box — 6.77 by 6.38 by 6.06 inches — triggered a dimensional weight of 1.883 lb. Getting the packed unit under 1.75 lb was worth about $1.94 to $2.56 a unit, and on the best-selling variant that fee was the difference between roughly 20% margin and the 44–53% its siblings earned. The whole gap was packaging. That is one account rather than a rule of thumb, but it is why we treat fulfilment fees as a packaging-engineering problem far more often than a negotiation problem — and it is invisible in any fulfilment provider's system, because Amazon bills from its own recorded measurements, not from yours.

There is a sixth item, and it only shows itself at peak. During one Prime event we watched a sports nutrition brand have 97% of its reserved units — 22,712 of 23,506 — sit frozen in fulfilment-centre processing: Amazon physically held the stock and had not yet made it sellable. The deal on the affected product drew 3,664 glance views and sold six units, because no delivery date could be shown. Its sibling product ran the identical deal and sold thousands, which is what made the diagnosis certain — the demand existed, the fulfilment did not. That was one account on one event, and the part worth carrying is the method rather than the figure: running a comparable sibling SKU as a control is the cleanest way to show a shortfall was operational rather than commercial, and it is not a test any fulfilment contract gives you.

The practical consequence for anyone reading a ShipBob review: if your inventory sits in FBA, a 3PL decision does not touch any of the five. If your inventory sits in a 3PL and you fulfil Amazon orders from it, some of them stop applying and different ones start. Know which world you are in before you shop.

The split case is the one worth thinking hardest about, because it is where most growing brands actually land: a pool of stock in FBA serving Amazon, and a second pool in a 3PL serving your own store and other channels. That arrangement is usually correct, and it doubles the number of places your inventory economics can go wrong — two sets of storage rules, two ageing clocks, two definitions of a damaged unit, and a transfer decision between them that nobody owns. The single most useful question to answer before signing a fulfilment contract is therefore not about the provider at all: which pool does each SKU belong in, and who decides when a unit should move? Get that written down first, and the quotes become much easier to read.

Where Dr. Stock sits, and who should ignore us

Who should ignore us. If your problem is that orders are not going out of the door reliably, or that you are personally packing boxes at ten at night, buy a fulfilment provider. That is a real and urgent problem and we do not solve it. Nothing about Dr. Stock reduces your need for a warehouse.

What we do. Dr. Stock is the operator for the Amazon-side economics above: reorder timing and stockout risk, cash trapped in slow-moving SKUs, storage fees and the aged-inventory threshold, the removal-versus-liquidation call, FBA fee errors and dimensional-weight misclassification, shipment discrepancies and reimbursement recovery inside the claim windows, and the true cost of returns per SKU.

Fable 5 does that work daily, with operators from Full Circle supervising — a full-service Amazon management company with more than $500M in managed revenue across 100+ brands. You choose the autonomy level: every change waiting on your approval, routine work automatic with the larger items queued, or fully autonomous inside agreed guardrails. Inventory purchasing decisions always come to a human regardless of setting. Orbit is included at no additional cost — inventory, finance, ASIN profitability, plus BSR, buy box, price and fee trackers.

Where we lose. We publish no price, and Cin7 and Sellerboard both do. We do not hold your stock, ship your parcels or handle your returns. And if your catalogue is small and slow-moving, there may simply not be enough Amazon-side leakage to justify a managed service — in which case the honest answer is that you should keep the money.

Two places to go next, and only if the description fits. If what you actually need is one authoritative stock number across several channels and locations, that is a system-of-record purchase, and our breakdown of Cin7 pricing covers what that costs and where the band edges sit. If you are comparing fulfilment providers rather than reviewing this one, our ShipBob alternatives page sets out the categories and who each suits.