Why does Google show a 4.8 rating when I have all 5-star reviews?

Why does Google show a 4.8 rating when I have all 5-star reviews? It is a question that comes up repeatedly on my blog and in the forums.

Google, on its help page says: We calculate an overall rating based on user ratings and a variety of other signals to ensure that the overall score best reflects the quality of the establishment.

Screen Shot 2014-12-20 at 5.25.38 PMObviously they are not calculating the average the way that most people calculate averages. If, like in the screen shot, you have eight reviews that are all 5-star reviews you would be hard pressed to come up with 4.9 average if you did the math.

There has been the suggestion that they may be counting other sentiment around the web which is a possibility. But in many cases I have looked at there are no other bad reviews around the web.

What is going on?

When you use reviews for ranking, like Google and many other companies do, you have a problem. Is a business with eight 5-star reviews actually better than a business with 48 5-star reviews and two 3-star reviews? Should the one with eight actually rank higher?

Probably not. Sooner or later the company with only eight reviews is likely going to get some lower reviews.  Or so Google assumes and they don't want to give them more credit than is their due. What is a company like Google to do?

extrapolatingWhat they always do. Write an algorithm.

Google clearly is weighting 5-star reviews differently if there are less than a certain number and they are probably using an algorithm that is something like a Bayesian average.

What is a Bayesian average you ask? From Wikipedia:

Bayesian average is a method of estimating the mean of a population consistent with Bayesian interpretation, where instead of estimating the mean strictly from the available data set, other existing information related to that data set may also be incorporated into the calculation in order to minimize the impact of large deviations, or to assert a default value when the data set is small.

In other words, when you have a very small sample (i.e. few reviews), Google is using its large dataset of review stars to model and estimate what your average would be IF you had more reviews. All very Googly, as they say.

The problem is that Google doesn't really explain all of this and folks fret a great deal. Often for good reason. The Bayesian logic, while integral to Google's thinking, is very foreign to most of us. Here is a tale from the forums when I explained how Google likely calculated this "average":

Well that is disheartening.  I've already had one person accuse me of having a bad review that Google is "hiding" for some reason because it's so obvious that the math doesn't work out without a low review being factored in somewhere, I can only imagine that other people investigating my business might think the same.  I really wish they would just use actual numbers instead of "assuming" that I'll get a bad review at some point.  I think real life is a lot more useful to customers than predictions!

What can a small business do? Not much, you just need to keep at the project of providing great service AND encouraging your clients to leave reviews. Sooner or later (20, 30, 40 reviews later?) Google will calculate a real average for you.

What should Google do? Be more transparent and tell people how this works. (It's that time of year when I can ask for presents, no matter how unlikely I am to receive it.)

44 Responses to “Why does Google show a 4.8 rating when I have all 5-star reviews?”

  1. Good hypothesis! And yes – it’s unfortunate that whatever way Google is calculating this average isn’t more transparent.

    • Google created Google plus to make money.
      They exist to sell advertising.
      Like Yelp they use ridiculous, useless algorithms that pretend to know something about your business and how it helps your clients, so they can sell you advertising. Google is about advertising and making money and they don’t understand anybody who has a different perspective.

  2. Thanks for bringing this topic to the table, Mike.

    A couple of years ago I worked for a review platform, who also used the Bayesean average to calculate the overall score for each company. The model is much more fair because it represents the voice of the crowd: A few moaners don’t break a company’s score, just like it takes quite a few delighted customers to achieve perfect scores.

    But explaining the logic of this, when lots of people hardly understand how to calculate an average…. Easier said than done.

    • @Joakim telling folks that is what you are doing and why and then sending them to a meaningful FAQ makes sense. That sounds like what you did. Unfortunately Google is NOT that transparent.

    • This sounds like the most sensible way to handle things and given our experience with Google’s reviews it still feels like a very young and far from balanced system.

      The 4.8 stars issue causes a lot of confusion and head scratching so a simple FAQ page detailing how reviews are calculated could save users that pain. The ‘voice of the crowd’ approach seems the fairest way to do things so why they operate this black box approach I just don’t know.

      Much like why they don’t have some kind of system (like Yelp) that helps to show / hide good or bad reviews. At the moment, we have a client that is getting some abuse in reviews from an ex employee they had to fire. This guy is setting up brand new Google accounts with spoof names (Mike Hunt – I kid you not) and then leaving slanderous and borderline racist reviews. And… these reviews stand, at the top of the page.

      Google has so much data on users and user accounts that surely I could write an algorithm to say ‘we don’t really have any info on this user to trust what they are saying here’ so lets hide this or at least push it down but… nope.

      It’s hard not to feel that Google still has a lot to learn on the reviews front from keeping business owners informed so they invest in the platform through to managing the quality and accuracy of reviews that they show.

      Cheers
      Marcus

  3. Our five star reviews finally appeared a few days after getting the fifth review. Like others have experienced, our average is also 4.8. The disheartening thing though is that we actually dropped out of the three-pak for our main keyword (but show up for some long-tailed keywords). We were previously positioned at number 1. I am guessing a reason could be that one reviewer decided to “help us out” by getting a few other people to review us (total of 3 reviews within a week). I believe that those reviews coming in at the same time hurt our rankings just as getting a bunch of links within a short span of time would hurt a website. I am at a loss at how to correct this other than trying to get more reviews, but maybe having them spaced out a little. Unrelated, but worth mentioning is that one of our competitors previously had exactly one review. One day they suddenly jumped up to 8 reviews. The strange thing though was that all the reviews were several years old? So, my question is why did these older reviews stay dormant for years and then suddenly appear? If anyone has an explanation for why this happens, I would love to hear it.

    • Google’s index of reviews is often a touch on the flakey side and they periodically lose and then find them. Also indiviudals can choose to hide them if they so desire. So they came back for one of those reasons.

      Google periodically changes the search results and even though it is related time wise, the change in rank is unlikely being caused by these reviews.

  4. Im guessing a lot of their review star metrics has also a lot to do with “Reviews From Around the web”

  5. “Reviews from around the web”, according to Google, are factored into the star metrics

    https://support.google.com/business/answer/3474120?hl=en&ref_topic=4596755

    Whether or not that is the case everytime, I think depends upon many other business factors

  6. @Jerious

    That help page is just in reference to the “reviews from around the web” section of the Knowledge Graph. It doesn’t speak one way or the other to the calculation of the score.

  7. All of this is great.. now lets fix the yelp reviews issue.. we have over 10 reviews that are true customer reviews that they ” put through their algorithm” and put out as do no recommend review. Its hard enough to get customers to review on your business but this… ?? ideas anyone?

  8. I think it’s a very good thing that google slightly down-ranks items with a low number of ratings. When this is not done, new items with one or very few reviews can unjustly get to the top of the ratings. They should be at the top of the “what’s new” list, not the top of the “best rated” list.

    If I were making the algorithm myself, I would give each item a single additional “average rating” of 3/5, and include that in the average. To get back to 4/5, they need only a single 5/5 rating. To get up to 4.9/5 however, they need to get nineteen 5/5 ratings. It seems Google is more generous than me; for the instance I am looking at it’s consistent with an initial “base rating” of 4/5. So to get back to 4.9/5 they would need to get nine 5/5 ratings.

  9. Hi there,
    Is there a way to show the reviews from google/ Yelp/ Yellowpages on my website.
    I have seen some website showing the cust review as flash & always wondered if there’s an easy way of doing that.
    I am non technical person so any help would be mush appreciated.

    Regards,
    Zainab Rangwala
    http://www.HennaServices.com

  10. Is there anyway to get Google to display the most recent reviews as the first three that show up on Google Maps? It is listing excerpts from only the negative reviews, which are now two years old. We have more recent reviews which are positive. A little unfair, it seems.

    Thanks,
    Laurie

  11. How does google decide which reviews they show. We have a number of reviews but it always hows the same reviews one of them twice.

    • Steve. There are many reasons that Google won’t show reviews; same IP block as the business, known relationship with the business, URLs in the review, duplicate review content. And there are many others we don’t know about.

  12. I would just like to say thanks very much for the explanation, this had been annoying me for a while now.
    As a one man outfit with very few customers in the first place, I was a little shocked and felt hard done by, when my fifth 5 star review went on and my rating dropped to 4.8.

  13. It’s really unfortunate that Google wants to use algorithms to calculate future negative reviews. I have (5) 5-star reviews and they give a 4.8 rating. Very unfair of them to alter review tallies to make it “fair” for others with more reviews and to account for future lower reviews.

  14. This system is the dumbest thing and another reason Google is last in any field they compete in except search …. they over think things in the beginning and then fail at updating/improving their products and move quickly onto the next thing without improving the products they have … gmail needs improvements .. yah we know … oh but look we have created google + … google + needs improving and alot more work to be more organic … oh but look we have goole glasses …

    My ratings are 5 star on facebook 5 star on Yelp 5 star on wedding wire 5 star on trip advisor and google … 4.7 with 5 reviews just stupid

    And the rest of you that justify this you OBVIOUSLY do not own a business that is impacted by yet again another half baked Google product that never gets improved (hence this article is over a year old and still going on)

    They are the only company doing this and that is why no one looks at this system except when using maps which is when it hurts us!

    • @Anthony
      I am sorry that you see this article as a justification for Google. I see my job as explaining how Google works to those folks that don’t see the world the same way.

      While I understand you displeasure at Google using a Bayesian average as opposed to a basic average, who is to say that it is wrong? It is different, it is confusing and it is frustrating. But without seeing the data that Google has, I would be hard pressed to say its wrong.

      As to whether Google updates the functionality of their software in a timely fashion, I would agree with you that they often do not and that they often change it or discontinue before anyone really gets the hang of it… but that has nothing to do with reviews.

      My suggestion to you- get more reviews. Once you hit a few more, Google will switch from a Bayesian average to a regular average and you will once again be a 5.0.

      Mike

      PS I have owned small businesses all of my adult life. I am now on #5.

  15. Very useful Mike. Many thanks for putting this article together.

  16. 13 @ 5 star reviews then you get 5 star rating after that you need 4.9999 average to get back

  17. Thanks for your article! We have 47 Google reviews, all of them 5 star, but have been given a 4.7 average. In the distribution table they show some 3-star and 4-star reviews, but we can’t find them! By the way, we have 1,235 5-star reviews on TripAdvisor.
    Also, Google list us a 1-star Hotel (we are not a hotel) but they list our main competitor as a 3-star Hotel (they are not a hotel) but they have the same amenities as we do. I find it confusing!

    • @Brian
      Ratings are reviews that don’t have any content. They are usually listed last. Are they possibly the cause of the average?

      As to Hotel Stars those are provided by a 3rd party data source, probably NorthStar.

  18. Hi Mike, thanks for the insight on Google reviews.

  19. Found your response when sheer curiosity had me googling the exact same scenario. Great explanation. Maybe Googke could also note a raw score when the variance is obvious. I initially assumed like others a negative but non public review was the culprit.

  20. Hi Mike, great post, thank you!
    And apparently the question is still of high interest: I got here trying to understand how is Google calculating my company seller rating: at present we have 3 out of 5 on
    http://www.google.com/shopping/seller?q=
    and a separate topic, we’ve just reached the 30 review limit to get reviews from independent sources
    Could you please advise on:
    1. how much time would take Google to make a rating adjustment and
    2. what type of calculation method they’d be using (assuming the third party review source will have a different weight than the current rating we have through google surveys)
    Thank you

  21. Well that answers a flipping question! Why oh Why oh Why can not Google just be honest and upfront about stuff like this?

    Thank you for writing this up – much appreciated.

  22. A competitor of ours has a horrible reputation and does bad work yet they have endless 5 star google reviews. they HAVE to be cooking the books?!

  23. Nice explanation. However, wouldn’t this mainly tend to lower the rating of an entity with smaller numbers of reviews? We wouldn’t expect the opposite.

    I only ended up here when there was no way to flag the strange 4.3 stars Google assigned a business with 7 reviews with a *total* of 25 stars. That *should* work out to 3.4…

    How on Earth can we explain a business getting a 4.3 instead of 3.4 when 28% of reviewers only gave *one star*?

    I find this highly suspicious.

    • *Sorry, that would be 3.6 stars (I rounded down instead of up) but it hardly changes things.

    • I would assume that a Baysian algo would work in both directions to smooth out the differences between small samples and a projected larger reality.

  24. This is screwed.

  25. Thanks for the explanation Mike.

    I disagree with Google assuming a business will get a lower rating soon.

    I get Google trying to accommodate businesses who have been there for a longer period of time and might have 50 reviews and a 4.9 rating over a business that only had 5 reviews and a 5.0 average.

    Perhaps Google not display the overall star rating until at least 20 ratings. But they allow the individual ratings to be shown as well as explain that businesses with less reviews will not be given a rating until they hit a certain number.

    I just don’t like the artificial lowering of the rating. I’ve been a Power Seller on eBay for about 10 years and I can only imagine if EBay played this game. If you don’t have 100% positive feedback people look at you like you have a third eye. Even 99.5% raises people’s suspicions about if you’re a trustworthy seller.

    I get some of the very best results as a chiropractor in my area. I work hard… and will tell you much harder than most because I’m not an assembly line. So when I see some of the chiropractors who I know aren’t nearly as good with 5.0 I wonder how many times I lost out on people going to them simply because they have a 5.0 and I have a 4.9.

    Anyway… looks like I’m just going to have to ask patients to write a review… that’s the only way around it.

    Thanks again for explaining it.

    Eric

    • @Eric
      I hear you.

      But for nuance I would add that Google doesn’t assume that any one business will get a lower rating soon just that in aggregate enough of the businesses in a large group will get a lower rating that it is more accurate to use a Baysian average than a more common average method.

      I would also point out that a common average is an arbitrary way of looking at reality as well because it assumes that each number is of equal importance. Sometimes a calculation like Net Promoter Score which examines the extremes provide a useful understanding of the aggregate as well.

      In addition the world of consumer reviews is dramatically different than thousands of reseller ratings. Most consumers actually prefer seeing a score less than a 5 as a condition of purchase. See: Study Shows Perfect 5 Star Review Ratings Aren’t Best. If that research is accurate then perhaps Google was giving you a bonus point towards conversion. 🙂

      It is not a simple question. My opinion on this, for what it is worth, is that the calculation is not bad, Google’s lack of transparency is!

  26. To add another complexity into this discussion, the review data which appears on google maps and then comes through to the google search, is now drawn from TrustYou, an independent organisation which scours the web for reviews and amalgamates them. Despite having a name which implies “trust” the data for our business was incorrect, misleading, and potentially harmful to the business.

    We are a new business operating in the hospitality sector. Our name is similar to another organisation based nearby, although there is no link between us. TrustYou initially showed data related to the other company as though it was ours. This resulted in incorrect review comments and gradings being shown for our business. I managed to prove to TrustYou that the data was incorrect and, eventually, they corrected their base meta data which now shows no reviews or comments for our business.

    HOWEVER, the google feed which draws this data has not been updated and is still incorrect. After much discussion with Google support (or lack of) they have concluded that it is not possible to change the data as they insist it is drawn from a range of review sources across the web. As a new business, and not listed on any other review sites except TripAdvisor – on which we have 30 5* reviews and no other reviews – it is simply not possible that review data exists elsewhere on the web related to our business which could reduce the grading.

    I am at a loss to know how to proceed as I have drawn a blank with Google. Any suggestions?

  27. If you provide your business name, I will elevate the issue with Google.

  28. Thanks Mike and yes, that is my business. The other business is called Cuil-an-Daraich.

    The google search for Cuil-an-Daraich is https://www.google.co.uk/#q=cuil-an-daraich+guest+house and their web address is http://www.cuil-an-daraich.co.uk/

    I first identified this problem as the google page said that the data was drawn from TrustYou. I contacted TY and they admitted that they had the wrong data and corrected and deleted it from their site and reset my business to zero reviews or data. It was hoped that when google updated, it would pick up the new zero data but this did not happen. The TY dashboard now shows nothing for my business. There were also some adverse comments which came through onto google but these have now been removed.

    The only review we have are TripAdvisor reviews – all 5* – these are not picked up by TY as they are not treated as “verified” reviews. Google say that the reviews must come from elsewhere on the web but this is not possible as we are not listed on any other booking sites and are a new business.

    I had also proved at one point that the erroneous reviews categorically belonged to the other business, as they were identical to the ones on their TA review and Bookings.com site, and related to guests who had not stayed with us.

    I’m probably rambling now! Let me know if you need any more info, and many thanks again for your help.
    David

    • @David
      As I suspected there is something amiss with your listing. It appears that there is a hidden duplicate where the actual review content for your listing might reside. Google is looking at it and diagnosing. It is a strange problem but hopefully I will have more for you shortly.

      Once that is sorted we can look to see if there is review confusion.

Leave a Reply