Ranking Methodologies

Category rankings

PeerSpot’s ranking algorithm, represented by a bar graph length, is based on a weighted aggregate score. Our rankings are based on 5 ranking metrics:

• Average rating
• Reviews
• Words/Review
• Views
• Comparisons

Category rankings are recalculated at the beginning of each month based on the new reviews published, older reviews that have ‘expired’ (published > 24 months ago), and other recent metrics such as impressions.

If we suspect a vendor has used artificial means intended to increase the ranking data, our team will thoroughly investigate the suspect foul play and take corrective action as needed (ex: use of a click farm by a vendor in order to increase views or comparisons of their product).

Vendors cannot delete reviews, but if they see a post that they feel is inconsistent with our Community Guidelines or Terms of Use and should be removed, they can explain why they feel the review should be removed, and our verification team will look into the issue and take appropriate action if needed.

Product Rating Ranking

Product rating stands for the average rating provided by users for the product.

Reviews ranking calculation:

• The product with the highest rating gets a maximum of 25 points
• The 25 points are awarded linearly between 6-10
  e.g. 6 or below=0 points; 7.5=7.5 points; 9.0=18 points; 10=25 points.
• If a product has fewer than ten reviews, the point contribution is reduced: 1/3 reduction in points for products with 5-9 reviews, 2/3 reduction for products with fewer than five reviews.

Product Reviews Ranking

Product reviews are the number of reviews published. A larger number of reviews suggests that a product is popular and widely used. This might be due to its high quality, strong marketing, or a combination of factors.

Reviews ranking calculation:

• The product with the highest number of reviews from the last 24 months gets a maximum of 15 points.
• Every other product gets assigned points based on its total in proportion to the #1 product in that ranking factor.

For example, if product X has 80 reviews, and the product with the most reviews has 100 reviews, then product X has 80% of the points.

15 * 80% = 12 Points

*Reviews that are more than 24 months old , as well as those written by resellers , are completely excluded from the ranking algorithm.

Words per Review Ranking

The Words per review ranking stands for the average number of words per review of a specific product. A review with more info and insights indicates that user is actively engaged and find it valuable enough to share their experiences in detail.

Words per Review ranking calculation:

• The product with the highest average number of words per review gets a maximum of 10 points.
• The 10 points are awarded linearly between 0-900 words
  e.g. 600 words = 4 points; 750 words = 7 points; 900 or more words = 10 points.
• If a product has fewer than ten reviews, the point contribution is reduced: 1/3 reduction in points for products with 5-9 reviews, 2/3 reduction for products with fewer than five reviews.

Product Comparisons Ranking

Product comparisons stand for the number of times a product in a category was compared with all other products in the same category.

Comparison ranking calculation:

• The product with the highest number of comparisons gets a maximum of 25 points.
• Every other product gets assigned points based on its total in proportion to the #1 product in that ranking factor.

For example, if product X has 80% of the comparisons the category comparisons leader has, then product X will receive 20 points.

25 * 80% = 20 Points

Product Views Ranking

Product views stands for the number of times a product was viewed by real users. This indicates potential interest or awareness around a specific solution, and can be valuable for gauging popularity and reach. Since a product can be listed in more than one category, the ranking calculation is relative and based on the percentage of the specific category comparisons out of the total comparisons of the product. This helps to assign only the category-relevant views.

To summarize:

• High product views indicate a popular product that users are actively exploring.
• Low product views may suggest a product with less visibility or user interest.
• Trending product views can reveal rising interest and potential opportunities.

Views ranking calculation:

• The product with the highest number of views gets a maximum of 25 points.
• Every other product gets assigned points based on its total in proportion to the #1 product in that ranking factor.

For example, if product X has 10,000 views in total, 200 Comparisons with other products in the category and a total of 1,000 Comparisons, the product will be assigned 20% of the total number of Views.

25 * 20% = 5 Points

Category Leaders

All products with a minimum of fifty points are designated that month with a Leader badge. There can be multiple, one, or zero leaders in a category.

Historical Ranking

Rankings for June 2023 and earlier used our previous ranking methodology:

For Average Rating , the maximum score is 28 points awarded linearly between 6-10 (e.g. 6 or below=0 points; 7.5=10.5 points; 9.0=21 points; 10=28 points).

For Reviews, Words/Review, Views and Comparisons , the maximum score is 18 points each.

• The product with the highest count in each area gets the highest available score. (18 points for Reviews, Words/Review, Views and Comparisons)
• Every other product gets assigned points based on its total, in proportion to the #1 product in that area.

For example, if product X has 80% of the number of reviews compared to the product with the most reviews, then the product’s points for reviews would be:

18 (weighting factor) * 80% = 14.4

Other Info

In each category, every product that has at least 40 points is:

Rankings by company size

When viewing the rankings filtered based on company size, the company sizes are:

• Small businesses (1-100 employees)
• Medium (100-1K employees)
• Large enterprises (1K+ employees)

Views and comparisons are based on the company size of people reading reviews. Reviews, rankings, and words/reviews are based on the company size of reviewers.

All published reviews that influence rankings

• Must come from real users that have used the product/service, or have evaluated it for the purpose of using it for their company, in the past 12 months. Reviewers must disclose if they’ve received a financial incentive for writing their review. Posts suspected to be bogus in any way will not be approved.
• Must include a balanced perspective. Everything’s got its plusses and minuses. We require that each review includes a balanced perspective. Posts that lack a balance of content (offering pros & cons) will not be approved.
• Includes the relationship of the reviewer with the vendor (customer, partner, reseller, etc.) Each reviewer is asked to self-identify the relationship they have with the vendor which we verify before publishing their review. We accomplish this additional verification using LinkedIn and publicly available information about the reviewer and their company. We follow up with the reviewer by phone and email if necessary.
• Have been verified and approved before it’s published on our site and included in ranking data. We take investigating suspect or inappropriate submissions extremely seriously as we believe it’s important the community can trust the data in order to make important decisions for their company.

How do we validate our reviews?

We require new users to register with their LinkedIn profile or company email address. This requirement enables us to prevent vendors or other non-users from posting fake reviews. This verification process allows us to put measures in place to identify suspicious users and/or fraudulent posts. The validation process combined with active community moderation and our commitment to review every post, gives us confidence in our information.

We take pride in being an objective site offering authentic reviews from real users. We ensure our reviews are authentic with several steps to ensure authenticity:

• User Validation: We validate the registration information against our own database which includes cross checking with LinkedIn profiles, to properly categorize whether the person is a real user or works for a vendor, reseller, consultant, integrator, or media. If we cannot clearly determine from someone’s LinkedIn profile that he/she is a real user, then the person’s profile is flagged for additional follow-up, including direct contact.
• Community Policing: We enable the community to self-police itself and to flag reviews that seem suspicious. At the bottom of every review is an “Inappropriate?” link that enables the community member to flag a potentially suspect review.
• Erring on the Side of Quality: If we cannot clearly determine that the review is from a real user, then we err to the side of quality and the review is not published. We’d rather have one fewer review in our system and sleep well at night knowing that all of the reviews published on PeerSpot are authentic and relevant for our user community.
• Combating Fraud: In the event that a vendor attempts to circumvent or abuse the system, we will put a red badge on the vendor’s page saying that the vendor has been suspected of planting false reviews. We believe the risk of this happening outweighs the potential gain from a fake review.

Order of reviews on product pages:

The default ordering of reviews on a product page is according to the date the review was published or last updated. Visitors can change the ordering of reviews to be based on review length, highest ranking, and lowest ranking.

By default, all reviews appear on product pages however reviews that are 24 months or older no longer count towards the category rankings. Visitors can choose to exclude older reviews using the filter on the product page.

Top comparisons and alternative products

Data displayed regarding the top comparisons for each product and category are based on live data of products compared by enterprise tech buyers. In order for a comparison to be displayed, it must hit a minimum threshold to ensure accuracy.

Have questions about our ranking methodologies? Reach out to methodology@peerspot.com