Glassdoor Ratings and Reviews Revisited – Part 1

There are Benefits on Both Sides of Glassdoor

TL;DR: Glassdoor is the go-to resource for job seekers, but employers should use it as well because a clean Glass Door can double as a mirror. There are benefits on both sides of Glassdoor: It is an information resource, a corporate self-check tool, and a communication channel.

Glassdoor Reviews are Not Biased – and May Cost a Lot

I have checked it myself: GD is a safe space. Confidentiality is what employees worry about and what employers are not comfortable with. As a consultant, I have the same concern and I assure you:  Just one accusation of breaching confidentiality would mean closing the door, for Glassdoor as well. But there has never been a single proven accusation yet.

Likewise, GD reviews and ratings are NOT negatively biased. Two-thirds of employees posting on GD say they are “satisfied” with their job and company. The total histogram of the company ratings is very close to Gaussian distribution, with the mean and mode close to the official average rating of 3.3-3.4. If you are friends with stats and math, you will have no further doubts.

More skeptical are the employers – or rather, they pretend to be. My short article posted last year was a good conversation starter with many business owners and executives. Notable observation: On several occasions, the “deniers” – those business owners who “don’t care about their GD reviews” – would call me back and ask triumphantly: “Have you seen our latest GD rating?!” – Obviously, they were happy to see their rating going up – positively impacted by our joined effort.

That confidentiality is the cornerstone of GD value. Some of you may know these statistics: Having a one-star-higher overall company rating on Glassdoor (on a 1-5 scale) had six times as large an impact on the odds of attracting job applicants from outside metros as did paying $10,000US more in salary. In simple words, you may offer an extra 10K – but there’s a high probability that the applicant will go to your competitor who pays less but has a better GD rating. How many people do you hire annually? How many years do you want them to stay?

The Three Benefits of Glassdoor

There are benefits for all, on either side of the GlassDoor

Glassdoor is often called a career site. It was launched as a company ratings site and although it has added job posting and job search functionality, it remains the go-to site for various information on the target company.

Business owners and executives, take note: The Glassdoor reviews are used not only by job applicants – but by your potential partners, clients, and competitors as well.

I analyzed data for over 30 companies that I am either familiar with or that are so well known that almost any data is publicly available. The three major benefits of the Glassdoor platform are:
30 Glassdoor Ratings

  1. GD provides candid anonymous reviews of various businesses (currently over 1 mln companies reviewed in 50 mln reviews), including salary information and even photos submitted by the insiders.
  2. GD gives a good opportunity to the businesses to see themselves through the eyes of the outside world and update the company image whenever a gap appears between the company desired and perceived image.
  3. It is an open communication channel for the employer and the employees.

How to Spot Fake Glassdoor Reviews and Ratings

Glassdoor provides a sneak peek of the company to the potential new hires. Understanding the impact of the reviews, some companies are trying to game the system. But this is easy to spot: look for irregularities.

In general, ignore outliers, try to see the trend. Read the reviews, look at the graphs, and compare your target company with other companies within the same industry or niche.

The distribution of genuine reviews follows the normal distribution curve. Of course, the graph will look more granular for a small company with just a few reviews. Nevertheless, when the reviews are genuine, the histogram (a) resembles a bell curve, and (b) the reviews are evenly spread along the timeline.

Knowing the dollar value of the GD rating, some employers try to coax their employees into writing positive reviews. Usually, that results in a spike of 5-star reviews, often appearing during a certain period of the year.

For example, one of the companies I analyzed, with a respectful 3.6 rating, shows this graph:

Sample Glassdoor rating #26

The distribution could have been perceived as normal, skewed towards the higher end – quite normal for a good company. But why then there’s an influx of 1-star ratings? This is not normal in either sense. Besides, as seen from the graph and confirmed by the date stamps on the reviews, there seem to be year-end 5-star spikes… Could be a coincidence…

Another suspicious pattern is when the company has no “mean” rating, or the rating distribution is an inverted bell curve.

Look at this company, with a better than average rating of 3.6:

Sample Glassdoor rating #15

Its entire score was built over the last year, with three consecutive 5-star ratings, after a stable 1-star history. New administration as of early 2020?

Likewise, another company, with a moderate 2.8 rating, has one 2-star rating, but otherwise, it is a battle of fives and ones. Could it be another coincidence?

Sample Glassdoor rating #3

After all, with small and midsized companies and a limited number of reviews, the rating may be very volatile indeed. Of course…

But as the story goes, if early in the morning someone knocks on your door – it could well be the Queen of England. But more likely it’s the milkman.

This is how an ideal profile should look like:

Sample Glassdoor rating #0

Check some companies and you will see that an exemplary distribution like that is quite normal.

Here’s another one: Maybe it is not the highest rating – but it looks like a steady and honest continuous improvement – which actually is what this company has been working on for several years:

Sample Glassdoor rating #5

The following two companies on my list surprised me a lot. They are definitely outliers – but are worth some further analysis. (If you guys recognize yourself, I would like to hear from you).

The No.1 on the list is a serious engineering company with a global presence and over 100 people working locally. It has a one-star rating based on just one(!) review. There’s not much to see there, but this is how the company looks like on GD, for consistency:

Sample Glassdoor rating #1

Any potential candidate with a university degree (and that’s over 25% of their employees) – will surely say that the available data is NOT statistically significant. Of course. But the data is sufficient to drive away many qualified candidates – the ones with a degree. However, I am sure that this is a great example where a very small effort can change a lot for the business as soon as they take the matter seriously and work on it.

Another miracle is company #30. It is a small software company – but with huge potential. Their GD rating is 4.5 Most probably, they are lucky to have a very cohesive team. Compare them to the one above …. Which of the two would you consider first?

Sample Glassdoor rating #30

I am looking forward to offering my Q7 test (Collectiver Culture Compass) to their team. Most probably it will show a very high level of the team’s alignment. And if they maintain a cohesive team, the sky is the limit for this business! With a modest number of reviews, the company demonstrates a steady growth over the last 2 years. If you check the dates, you will see that there are no “spikes” of the employees’ affection. It is what it is.

There’s another cool observation about these two outliers. On the “total” graph, you can see the GD rating (blue) and yet another Engagement Ratio (orange) which I came up with myself. The GD rating is a good indicator of employee satisfaction. It correlates with employee engagement, but employee satisfaction and employee engagement is not the same thing.

Employee satisfaction and employee engagement is not the same thing.

30 Glassdoor Ratings

The GD rating to E.Ratio correlation proves moderately significant within our limited sample. Thus, without being too scientific, we could use it for the initial assessment of employee engagement – because this KPI is readily available.

Our two outliers indicate that Company #1 has room for improvement in the engagement department, while company #30 must be on engagement steroids!..

(To be continued.  Subscribe to the newsletter or bookmark this page for more on GD rating and E.Ratio. Part Two: Beyond the Glassdoor Stars)