Carbon6 reviews: an average across a dozen different products tells you almost nothing

What you are actually reading a review of

Search this phrase and you get star ratings. What almost none of them tell you is that the thing being rated is a portfolio, not a product. Listed on Carbon6's own site, the suite spans profit recovery, advertising, inventory and sourcing, data and operations, and multichannel tooling — a dozen or so named tools including a reimbursement recovery service, a chargeback product for first-party suppliers, an inventory forecasting tool, a link and attribution product, a PPC tool, several analytics and alerting products and a marketplace research tool.

Those were built by different teams, at different times, for different buyers, and most of them were independent companies with their own reputations before they were brought together. A four-point-something average across all of them is an average of a warehouse forecasting tool, an advertising link shortener and a claims-filing service. It cannot guide a purchase, because nobody is buying all twelve.

So the first instruction is mechanical: find the review of the module you are considering, and ignore the suite-level score entirely. Search the individual tool's name, not the parent brand. The corpus under each original product name is deeper, more specific and far more useful than anything filed under the umbrella.

This is not a criticism of the company. Roll-ups are a legitimate and often good strategy — buying the strongest point tool in each niche and putting it behind one login genuinely helps sellers who would otherwise juggle six subscriptions. It just makes the review literature nearly useless unless you read it at the right level.

The ownership change that dates every older review

Second thing to check before you trust a review: when it was written relative to February 2025.

SPS Commerce announced an agreement to acquire Carbon6 in January 2025 and completed the acquisition in February 2025, for a total purchase price of approximately $210 million, with roughly forty per cent of the consideration in SPS Commerce stock. Carbon6 is now part of a public company whose core business is retail supply-chain data rather than Amazon seller tooling.

That is not a warning. SPS Commerce is a substantial, long-established business, and being owned by one is generally a stability improvement over being owned by a venture-funded roll-up. But it does mean an ownership change sits in the middle of the review timeline, and reviews on either side of it describe different companies in terms of roadmap, support structure, pricing authority and integration priorities.

The buyer's move here is not suspicion, it is diligence, and it applies to every vendor you use including us: ask what happens to your price, your data and your open cases if the company changes hands again. Ask for price protection for a defined period and a notice period for fee changes, in writing. Those clauses cost nothing to request at signature and are impossible to obtain afterwards.

Ownership in the Amazon tooling market moves constantly — several well-known names in this space have changed hands, changed names, or been folded into larger suites in the past three years, which is precisely why so much of the published commentary about them is out of date.

What the individual pieces are genuinely good at

Read at module level, several parts of this suite have real reputations, and it is worth being specific about who each one suits.

• The reimbursement recovery service. Contingency-priced work with human specialists auditing accounts rather than pure automated filing. If you want claims worked by people who will assemble evidence and chase a stalled case, that model is the right shape, and the human review step is what separates cases that get paid from cases that get closed.
• The inventory forecasting tool. Amazon-native replenishment planning with seasonality and promotional adjustment built in, rather than a trailing average dressed up as a forecast. For brands whose demand is genuinely seasonal, that distinction is the whole ballgame.
• The chargeback product. Aimed at first-party vendors dealing with deductions and compliance charges, which is a different discipline from third-party seller work and poorly served elsewhere.
• The attribution and external-traffic tooling. Built for brands driving off-Amazon traffic to listings and wanting to know what it did.

The consistent criticism across the suite, and it is fair, is the cost of assembly. Modules were built separately, so onboarding several of them is several onboardings, with separate configuration and permissions. A small team that adopts three tools to use one well is paying an operational tax that a single-purpose product would not charge. Buyers who do best here are the ones who genuinely need three or four of the pieces.

What it costs — and what the absence of a price actually means

Carbon6 publishes no pricing on its site. When we read it on 20 August 2026, the calls to action were to book a call, with no figures attached to any module.

Say that neutrally, because it is normal for a portfolio like this. Modules are priced differently, several are sold on contingency rather than subscription, and enterprise deals are banded. A single headline number would misrepresent nearly every buyer's actual bill. Plenty of respectable companies price this way — we do, for Dr. Stock, so a page that treated quote-only pricing as a black mark would be arguing against itself two paragraphs later.

What it does mean is that reviews become your only price signal, and reviews are a terrible one. "Expensive" in a review has no number attached to it, no indication of which modules were bought, no company size and no year. When you find that word in this suite's feedback — and you will — treat it as an instruction to ask, not as data.

For the recovery module specifically, a useful comparison anchor does exist. GETIDA publishes performance-based pricing starting at 25% of recovered funds with no subscription and the first $400 free for new accounts, read the same day. That is not necessarily what any other vendor charges — contingency rates are commonly banded by volume — but it gives you a published number to open a conversation against, which is exactly what a quote-only process otherwise lacks.

How to evaluate any suite whose reviews are noisy

This shape recurs constantly in Amazon software, so the method is worth having rather than the verdict. Six steps, and they take an afternoon.

• Name the module. Decide which one tool you are buying and search reviews under its original product name as well as the suite name. If the original name produces a deeper corpus, that is the one to read.
• Date every review against the ownership timeline. Anything written before an acquisition describes a different company's roadmap and support.
• Read the three-star reviews only. Five-star reviews describe hope and one-star reviews describe billing disputes. The middle band is where people describe the product.
• Ask for a customer in your category at your size. A tool that works beautifully for a 40-SKU private-label brand may behave differently across 4,000 SKUs.
• Ask what you keep when you leave. Historical data, configuration, open cases. Get the answer before you integrate, not after.
• Ask for written notice of fee changes. On every supplier, including us.

Two of those steps are worth more than the other four. Reading only the middle band of reviews removes almost all of the noise, and dating reviews against corporate events removes most of what is left. Applied together they turn an unusable star average into something you can actually decide from.

Where the suite ends, and where we start

Even assembled well, a suite of tools reports and files. Some of the largest supply-side numbers are neither reported nor filed — they are decided.

• Size-tier and dimension errors. Amazon bills fulfilment against the measurements it has recorded. Wrong measurements mean an overcharge that repeats on every unit shipped until somebody requests a remeasure and drives the dispute to a conclusion. There is no claim form for it, which is why it falls between products.
• The 181-day line. 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. A forecasting tool will show you the age. It will not choose between removal, liquidation, a price move and absorbing the charge, and the assessment does not wait.
• Cash in dead stock. The largest line on many brands' balance sheets is inventory that has stopped selling. Releasing it is a decision with commercial consequences, not a report.
• Reorder timing. An alert is not a purchase order placed against a lead time with money currently tied up in something else.

Dr. Stock does not fulfil orders, is not a system of record and is not an ERP. It is those decisions run as a product — Fable 5 on Amazon inventory and supply chain, with human operators supervising at an autonomy level you set, and purchasing always returning to a person. The operators come from Full Circle, a full-service Amazon management company with $500M+ in managed revenue across 100+ brands. There is no published price: a demo, the first 30 days free, and a number agreed on the call. Orbit is included at no extra cost.

The honest redirect: if what you want from this suite is the advertising tooling, then the question you are really asking is about the ad account, and Dr. PPC is the product built for that — not this page.