Scores by kind of claim
Each kind of claim is scored on its own, out of 100: the share that held up, with the range it likely falls in. Fewer resolved claims give a wider range.
-
Factual claims
Statements about the present or the past, checked against the record: did they hold up?
84 79 88 -
Predictions
Statements about events outside the person's control, checked once they fell due: did they come true?
Insufficient record: 2 predictions resolved, 5 needed for a score.
-
Promises
Commitments to the person's or organisation's own future action, with something to deliver: was it delivered?
Insufficient record: 3 promises resolved, 5 needed for a score.
-
Guidance
A company's own figures for its results over a period, checked against what it then reported.
Insufficient record: 1 guidance claim resolved, 5 needed for a score.
All types combined: 84 (likely 79 to 88, high confidence), from 90 claims: 90 factual claims. It mixes kinds of claim in different proportions for different people, so compare people kind by kind.
Jeff Bezos's claims
165 claims · page 2 of 2
“We consider a large number of factors in search relevance to predict the probability of a customer’s interest and optimize the ranking of results.”
Jeff Bezos, CEO, in shareholder letter 2011 2010 Letter to Shareholders (SEC (Amazon))
Amazon's search considers many relevance factors to predict customer interest probability and optimize result ranking.
Unresolvable The public record could not settle it.
Said for an organisation they lead or work for; it counts on both records.
What would count as true or false
Judge as of 2011. True: Amazon documentation, patents, engineering descriptions or independent technical reporting from around 2011 show that Amazon's product search ranked results using many relevance factors (e.g. text match, sales, customer behaviour, availability) to estimate how likely a customer was to be interested in each result. Mostly true: search used many factors to rank results, but the 'probability' framing is only loosely supported or a minor detail differs. Mixed: search used several factors, but evidence suggests ranking was driven largely by one or two signals instead of a model predicting interest. Mostly false: ranking relied mainly on simple text matching or a single factor, with the claim overstating its sophistication. False: evidence shows Amazon search did not rank results by relevance factors in this way.
Evidence
- Amazon's A9 product ranking algorithm: Your guide to Amazon SEO for maximum visibility Search Engine Land · 5 March 2020
“To do that, the A9 algorithm considers “factors such as degree of text match, price, availability, selection, and sales history.””
- Amazon Ranking Algorithm: The Internal Ranking Explained - SISTRIX SISTRIX · 6 August 2021
“The greater the probability that a product will be bought, the higher Amazon will rank that item, in order to maximise sales. The algorithm takes into account both the behaviour of users, e.g. page views, purchases, but also the structure and content of the individual product pages.”
- Amazon SEO: Improve your Amazon ranking IONOS Digital Guide · 21 July 2017
“even if A9 evaluates significantly fewer factors than, for example, the Hummingbird algorithm used by Google.”
- Amazon A9 Algorithm: How Amazon Search Ranks Products AMALYTIX · 12 October 2020
“Amazon does not publish its ranking formula.”
- Amazon SEO: Improve your Amazon ranking IONOS Digital Guide · 21 July 2017
“The previously mentioned A9 algorithm takes into account two particular types of factors when classifying the offerings on the e-commerce platform:”
- Amazon SEO: Improve your Amazon ranking IONOS Digital Guide · 21 July 2017
“The algorithm evaluates how well the selected products have performed thus far, with user responses playing an important role. Among others, A9 takes into account the click and conversion rates as well as the customer satisfaction.”
- Amazon SEO: Improve your Amazon ranking IONOS Digital Guide · 21 July 2017
“If the Amazon algorithm has filtered out all articles from the product catalog whose product data matches the user’s search, the second step is to show off the performance factors.”
- Amazon SEO: Improve your Amazon ranking IONOS Digital Guide · 21 July 2017
“Since the A9 algorithm is just as random as Google’s algorithm with its placement of results, the Amazon ranking can be actively improved - even if A9 evaluates significantly fewer factors than, for example, the Hummingbird algorithm used by Google.”
- Jeff Bezos (Amazon.com) - "Earth's Most Customer-Centric Company: Differentiating With Technology” | Infinite MIT Infinite History / MIT Own record
“So what we're trying to do is, we look at-- for personalization we basically calculate for every customer, a probability that they will-- what's the most likely product that you will buy.”
Reasoning
The independent sources describing Amazon's A9 ranking factors (text match, price, availability, sales history, click and conversion rates, and purchase probability) date from 2017–2021 [E1, E2, E6, E7], not from around 2011 as the criteria require. One of them also says A9 evaluates significantly fewer factors than Google's algorithm [E8], and another notes that Amazon does not publish its ranking formula [E4]. The only Bezos source describes probability-of-purchase calculations for personalization, not search ranking [E11]. None of this evidence shows how Amazon's search ranked results in 2011, so the claim cannot be judged under the criteria as written.
Cite this claim
Jeff Bezos, “Amazon's search considers many relevance factors to predict customer interest probability and optimize result ranking.” (said 2011). Verdict: unresolvable. Reliability API, claim c_ac337e638a7a, method 0.1.0+c876030c. https://epistacklabs.com/claims/c_ac337e638a7a
Claims without a verdict
Is a verdict here wrong, or do you speak for Jeff Bezos? Reply on the claim's own page, with the form at its foot (how replies work). Verdicts are made automatically; how verdicts and scores are made.