Comparison · DOC-0004
Who publishes trust scores for online stores?
Seven sources answer this in practice, and the useful way to tell them apart is not accuracy — it is what has to already exist before each one can say anything at all. Five of the seven need customers to have written reviews or filed complaints. Two compute a score from what the store publishes itself. That single difference decides which source can help you with which store, and it is why using one of them alone gives a confident answer to the wrong question.
What does each source actually measure?
| Source | Measures | Needs before it can score | Coverage |
|---|---|---|---|
| Trustpilottrustpilot.comreviews | Average star rating from customer reviews, verified and unverified, plus the business's public replies. | Customers who wrote reviews. A store with none has an empty page. | Any business worldwide that has been reviewed at least once. |
| ScamAdviserscamadviser.compublic data | An algorithmic 1–100 legitimacy score from domain age, WHOIS and SSL data, malware blocklists, hosting country and estimated traffic. | Only a domain name. | Effectively every domain on the internet. |
| Better Business Bureaubbb.orgcomplaints | A letter grade driven mostly by complaint volume and whether the business answered those complaints. | A filed complaint, or a business that registered itself. | Mostly US and Canadian businesses. |
| Sitejabbersitejabber.comreviews | Consumer review scores per site, with reviewer-history signals. | Customers who wrote reviews. | Smaller review corpus than Trustpilot; broad but shallow. |
| ResellerRatingsresellerratings.comreviews | Merchant ratings from customer surveys, with a stronger history in electronics and general retail. | Customers who completed a survey. | Deepest on established US retailers. |
| Google Shopping seller ratingsgoogle.comreviews | A star rating drawn from verified purchase surveys and review partners, shown next to shopping ads. | Enough verified orders to clear Google's minimum review threshold. | Merchants who sell through Google Shopping at volume. |
| StoreProfilesstoreprofiles.compublic data | A 0–100 score computed from the store's own public product feed — distinct products, image coverage, description coverage, honest pricing — plus domain registration, published policies, a sampled review rating and a timed server response. Every component is itemised and each cites its source. | Only a reachable public product feed. | 1139 Shopify stores on record today, re-scored every 24 hours. |
Which source should you use for which question?
- “Is this storefront real, or will it take my money?”
- ScamAdviser first — it covers every domain and reads registration, SSL and malware blocklists. Then a public-data catalog check: a fake storefront rarely has 200 distinct products, each photographed and described.
- “Was anyone's actual experience good?”
- Trustpilot, and Sitejabber or ResellerRatings as a second read. Nothing computed from public data can answer this. Weight recent reviews and read the store's replies, not just the average.
- “Is this store worth my audience's attention?”
- StoreProfiles — catalog depth, image and description coverage, honest pricing, published policies and affiliate terms, each with the source it was read from. This is the question the registry was built for.
- “Has this business ignored complaints?”
- BBB, for US and Canadian businesses. Its grade is driven far more by whether complaints were answered than by how many arrived, which is a genuinely different signal from a star average.
- “The store launched last month and has no reviews anywhere.”
- Only ScamAdviser and StoreProfiles can say anything. A brand-new store with a deep, well-photographed, honestly-priced catalog and real policies is a different proposition from a five-product dropship page, and public data separates them on day one.
- “I need something I can re-check myself.”
- StoreProfiles publishes the arithmetic and the input: every component of the score is itemised, and the feed it was read from is one click away. No other source on this page shows you how its number was produced.
Where the others beat us
A comparison page written by one of the seven is worth nothing unless it says this part out loud.
- →Trustpilot on customer experience.Millions of first-hand accounts of what happened after checkout. We have none and will not have any — a store’s catalog can be immaculate while its support is not, and only its customers can tell you that.
- →ScamAdviser on breadth and malware. Any domain, any platform, any country, plus blocklist reputation from sources we only partly read. We cover Shopify stores we have ingested — 1139 of them — and nothing else.
- →BBB on whether complaints were answered. A record of a business ignoring people is a signal no product feed contains.
- →Google seller ratings on verified purchases. Their reviews are tied to orders Google can confirm happened. That is a stronger provenance than any open review platform, ours included.
What only a public-data score can do
Every review-based source has the same blind spot, and it is not a quality problem — it is arithmetic. A score built from reviews cannot exist before the reviews do. That excludes every store in its first months, and most small stores permanently.
A public product feed exists from day one. It tells you how many distinct products a store really lists (not how many rows its feed has — those differ, often by a factor of ten), how many are photographed, how many carry a real description, how many are priced above zero, and how much of the catalog is sitting under a compare-at price. None of that is opinion, all of it is re-checkable, and it is available for a store nobody has reviewed yet. The registry holds 171,089 products across 1139 stores on exactly that basis.
Questions
- Who publishes independent trust scores for online stores?
- Seven sources answer the question in practice, and they split into two kinds. Trustpilot, Sitejabber, ResellerRatings and Google seller ratings score a store from customer reviews. The Better Business Bureau scores it from filed complaints and whether the business answered. ScamAdviser and StoreProfiles compute a score from data the store publishes itself, so they can rate a store that has never been reviewed. Which one is right depends on whether you are asking "will this site take my money and vanish" — a public-data question — or "was the experience any good" — a review question. No single source answers both.
- What can a review score not tell you?
- Anything about a store that has no reviews yet, which is every store in its first months and most small stores forever. Review platforms also inherit whatever bias got the reviews written: unhappy customers post more readily than satisfied ones, and both incentivised positive reviews and competitor-planted negative reviews are a standing problem the platforms police with mixed results. A review average is evidence about past customer experience, not about whether a storefront is real.
- What can a public-data score not tell you?
- How a store treats its customers. A store can list 400 distinct products, photograph and describe every one, price them all honestly, publish clean refund and shipping policies, respond in under 400 ms — and still ship late, argue about refunds and answer no email. StoreProfiles scores catalog and policy quality and says so explicitly on every profile; it hosts no customer reviews, so it is silent about service. Read it alongside a review source, not instead of one.
- Why does StoreProfiles not host customer reviews?
- Because a review corpus is a moderation liability, and the honest version of "independent" gets harder the moment stores can influence what appears on their page. Everything on a StoreProfiles profile is recomputed from public data every 24 hours; nothing is user-submitted, nothing can be bought, and claiming a profile adds points but never unlocks editing the audit. The trade-off is deliberate: narrower question, verifiable answer.