I saw my reviews on the interview process for IBM and couple of other companies disappear without trace. So my guess is their business model is trying to get companies to pay to remove bad reviews.

Glassdoor is hopelessly spammed by HR departments, “Advice to management: none, they are wonderful!” and so on.

My old company did that. PR sent an email to all staff saying "please leave us a good review on Glassdoor". I checked and the week before four new reviews had come up, all using the same writing style, all incredibly positive. Stuff like "negatives: you have to do lots of hours, but you're paid well so I don't mind" (spoiler alert: only half of that sentence was truthful).

I reported all of them to Glassdoor and sent them the PR email, the reviews are still up months later.

edit: just checked again and everyone left fairly truthful 1* reviews in a short period of time a few weeks after I left, I'm guessing they did another round of "please leave positive reviews" and it backfired.

And for some reason these are the one that are displayed by the default "relevant" filter and the upvoted (more nuanced or critical) one are hidden.

Just checked and yup, all the planted five star reviews are at the top despite having no more than one "helpful" reports. Negative reviews with up to ten "helpful" reports are on at least page two. This is while sorting by "popular".

Seems to have taken the opposite strategy of Yelp, which bumps one star reviews to the top and makes companies pay to hide them. Or my ex-company is already paying, it's hard to tell.

Are they an "engaged employer" in the top right corner? If they are, my theory is that translates to "paying customer" hence the weighting given to the positive reviews.

Yes, they are. That theory makes a lot of sense tbh

This exactly.

I worked for one company where everyone who left started posting bad reviews. The next thing you know people who were still there were posting wonderful reviews and getting an "official response" from the company.

That's my suspicion as well. Years ago I consulted for an agency that was full of negative reviews on GlassDoor, I noticed each time I checked back every few months that a huge number of the previous negative reviews would be gone and their rating would be higher.

In my experience, Glassdoor is highly motivated to investigate and potentially remove reviews due to complaints from companies paying them, and puts little to no effort into removing reviews due to complaints from everyone else.

I won't go as far as to say that they remove reviews purely on request, but I've seen removals due to requests that were based on what I felt was a pretty shaky justification.

Isn’t that just because they only want reviews from current or former employees, not prospective ones?

No, they solicit reviews from candidates and interns too.

> How they ever made any money?? What was their business model?

Advertisement [1] but also [2][3][4].

[1] http://resources.glassdoor.com/advertising-on-glassdoor.html

[2] http://vator.tv/news/2015-06-20-how-does-glassdoor-make-mone...

[3] https://www.quora.com/How-does-Glassdoor-make-money

[4] https://www.quora.com/How-does-Glassdoors-business-model-wor...

At some point in the past I was wondering how hard it would be to do sentiment analysis based on Glassdoor reviews of public companies to try and trade on the results.

I could swear when I dug into this I found that they had actually licensed out their dataset to a couple of hedge funds already that were doing exactly that (and paying for the privilege).

I can’t find this anymore, so maybe I am imagining things? At any rate this FT article implies there were people out there likely doing this though unclear if they paid Glassdoor for the privilege: https://www.ft.com/content/d86ad460-8802-11e7-bf50-e1c239b45...

Yes, they offer their dataset to companies as a paid service.

Source: my prior company paid for their entire data set and I personally used in a data science project.

> my prior company paid for their entire data se

Are you aware of how expensive that kind of data set would be? Also curious on what any company would do with that kind of data. Predict which company employees are the happiest?

The most valuable insights from this would be to get trickled insider information. Once in a while a comment gives you an insight that is not currently available elsewhere. Like unannounced layoff rounds. Statements about employee fluctuation? Reading reviews by software engineers gives you and insight about how this co is executing its IT projects.. etc.

Would not surprise me given the current obsession by HF to find value in alternative data. Would be interested if you find details.

Advertising jobs - it's quite an expensive option, but a lot of the bigger companies do post there as part of their rec ads mix.

I assume they make money from either advertising jobs or commission on jobs they place someone in. Pretty much the same as LinkedIn, except more targeted to people looking for the jobs and without also providing a platform for recruiters to cut them out.