Rank	Change	Price	Smartphone	Digital Trends	The Verge	TechRadar	Creative Bloq	IGN	Tom's Guide	Pocket-lint	T3	Laptop Mag	ZD Net
1	-	Check price	iPhone 15 Pro Max	1	-	3	-	-	1	1	1	-	6
2	+3	Check price	Samsung Galaxy S24 Ultra	-	4	1	-	-	2	-	2	-	1
3	-	Check price	iPhone 15	2	1	2	-	2	5	-	-	-	-
4	+3	Check price	Google Pixel 8 Pro	4	-	4	-	-	4	2	-	-	2
5	-3	Check price	Samsung Galaxy S23 Ultra	-	-	-	1	-	-	3	10	3	-
6	-	Check price	Google Pixel 7a	-	5	-	-	-	3	-	-	1	-
7	-3	Check price	OnePlus 11	-	-	7	-	-	-	5	6	4	-
8	+1	Check price	Samsung Galaxy Z Flip 5	14	6	-	-	7	-	9	-	-	-
9	+1	Check price	iPhone 15 Pro	-	3	-	2	-	8	-	-	-	-
10	-2	Check price	Google Pixel 8	8	-	-	-	1	-	-	-	-	-
11	New *	Check price	OnePlus 12	3	-	-	-	-	7	-	-	-	7
12	+10	Check price	OnePlus Open	-	-	9	-	-	11	-	-	-	3
13	-1	Check price	iPhone 15 Plus	9	-	-	-	-	10	4	-	-	-
14	-1	Check price	Samsung Galaxy Z Fold 5	13	7	-	-	6	-	-	-	-	-
15	-1	Check price	Google Pixel 7 Pro	-	-	-	4	-	-	-	3	-	-
16	-1	Check price	Samsung Galaxy S24 Plus	11	2	-	-	-	-	-	-	-	-
17	New *	Check price	Asus Zenfone 10	-	-	-	-	-	-	7	-	-	4
18	-1	Check price	Nothing Phone (2)	-	8	-	-	-	-	-	4	-	-
19	-1	Check price	Asus ROG Phone 7	-	-	-	-	5	-	8	-	-	-
20	-1	Check price	Oppo Find X5 Pro	-	-	-	7	-	-	-	7	-	-
21	-10	Check price	iPhone SE	10	-	-	-	-	-	-	-	5	-
22	New *	Check price	OnePlus 12R	7	-	-	-	-	9	-	-	-	-
23	New *	Check price	Motorola Razr Plus	-	-	8	-	-	-	-	-	-	8
24	New *	Check price	Samsung Galaxy S24	12	-	-	-	-	6	-	-	-	-

What is the purpose of MetaRankings?

Experts do a better job researching products that customers. But we noticed that even the most prominent product experts often disagree on what the best product is. A printer might be praised by CNET but panned by PCMag. Aggregating opinions of top experts is the best way to get an expert consensus.

Why are MetaRankings awesome?

So glad you asked! The biggest problem with reviews is that they can be biased, fake or low quality. MetaRankings solve all of these problems.

• MetaRankings are genuine. Since we're aggregating reviews by top experts, we don't have a problem of 'someone on Reddit recommended it'. These are opinions of real people whose job is to review products. You can look up their LinkedIn profiles.
• MetaRankings are transparent. We show you the list of experts that we used to create the ranking. You can click on each expert to see their ranking. Our aggregation methodology is public and all the sources are always available. We turn opinions into facts.
• MetaRankings are unbiased. If multiple experts independently recommend a product, it must speak to its quality.
• MetaRankings are result of hard work. Every MetaRanking is backed by hundreds of hours of independent experts doing research and testing. You can always click on individual links to browse those reviews, or ask our ChatBot to summarize a ranking.

How do we choose experts?

We Google them, how else?

Are MetaRankings done by AI or humans?

Both. Some parts are done by robots, some by humans.

What is a 'fuzzy metaranking'?

Our best rankings are presented in a form of a nice table above (mapping expert endorsements to specific products.) But because of a few technical limitations, we can't always determine which product is being endorsed. This is because often experts choose to endorse a category of products (such as a brand or a product line), not a specific product. Aggregating becomes a more difficult task in this case. So we create 'fuzzy metarankings' that rely on generalized experts preferences instead of explicit endorsements. Fuzzy metarankings are a bit less accurate, and because of this they are weighted a little less in the Awesomeness Score. Fuzzy MetaRankings are not as great as concrete ones, but are much better than nothing.

How are the rankings created?

It's actually quite a process.

Step 1: identify experts. Robots determine most respectable sources on the Internet, then the team manually chooses the best ones.

Step 2: import experts' rankings. Robots try to import the rankings that experts independently publish. Since robots inevitably mess up, team manually checks the data.

Step 3: exclude unrelated products.
In their creative wisdom, experts often tend to include accessories or alternatives in their ranking. Our team has to manually exclude products that don't match product category.

Step 4: determine if experts are endorsing the same product. For variations of the same products our team needs to determine whether the products are the same or not (is iPhone 13 64GB and iPhone 13 128GB the same model or not?)

Step 5: assign weights to expert endorsements and aggregate opinion. Currently we use ranking methodology 0.7.2: being included in the expert ranking scores points. Being ranked #1 carries a lot more points. Being included in top 3 also carries bonus points. Then the point value of every endorsement is adjusted by a geometric progression with common ratio slightly below 1. In other words, product ranked #1 carries a lot more weight than product ranked #6; product ranked #6 carries slightly more weight than product ranked #8.

Step 6: add it all up. Computers are quite good at this bit.

Step 7: do it all again in a month.