Hacker News new | past | comments | ask | show | jobs | submit login
Ask HN: Glassdoor is unbelievably bad, why no one disrupting it?
48 points by cheikhcheikh 10 months ago | hide | past | favorite | 45 comments
How the hell does a company that make 300m$ in revenue a year has it's main application be this garbage.

I'm not even talking about the "review-wall" requirement. The whole thing is garbage, the cities filter for example, it lists all the cities in the world!! you can't filter by country easily, and the filter don't stick from tab to tab, and you can barely search for sh*t, it's so clunky and bad it looks like it's a side project done by an intern.

I'm just so confused as to why it's the still by far the most popular website of its genre, I mean look at Twitter, it's a way more sophisticated enginereing product and with so much higher UX investments and it's getting destroyed by Threads soo fast the moment they started slipping.

Excuse the rant, I just don't get how there are no real alternatives to Glassdoor, I know levelsfyi exist and stuff similar to that, but nothing at the scale and breadth that Glassdoor covers




Glassdoor, Yelp, Goodreads... all of these user review services are terrible, but they are so entrenched in their respective markets they're difficult to disrupt. The value of these services is not the companies themselves, it's the users. They're essentially specialized social networks. Social networks take an enormous amount of time and effort to build. It isn't the tech, the tech is easy now. It's the people.

The companies that run these services contribute very little to their success. In fact many times they're popular in spite of the company. Think about how Yelp or Glassdoor will remove negative reviews if the business pays them, etc.

Just more landlords extracting value from things people create and enshittifying everything.


Let's try and see if the problem can be reframed. It isn't really about reviews. It's about effective hiring. If you wanted to optimize hiring how would you do it?

Reviews might be unnecessary if candidates can collectively gather up the data that shows if a company is actually hiring, and actually reviewing resumes, and actually responding to candidates with above-generic feedback.

The reason this data isn't collected and presented is that job sites and application tracking software make their money from companies. There isn't a site that makes money from candidates to counteract this force. The power imbalance between candidate and employer is tilted in favor of the employer.

If a job post showed there are 286 applicants for a job would you bother writing a cover letter? I wouldn't. I would move on to the job posting that responds promptly and has fewer candidates.

If a job post showed there are 100 applicants, the job posting has been open for 30 days, and there are zero resume reviews, would you bother applying to this job and others like it in this company? I wouldn't. It means the employer isn't really hiring or is disfunctional enough to announce a position without giving it attention.

I'd even be willing to pay the recruiter for their time giving me somewhat helpful not legally binding feedback like we think "you're overqualified", "you're missing an important skill", "we're not sure yet". If the recruiter starts generating revenue for the company, the company might hire more of them.

I'd also be willing to recommend my friends to good companies.


IIRC LinkedIn actually does tell you how many applicants have applied via their platform, but since it doesn't tell you 90% don't have the right to work in the country never mind meet the job spec in other requirements, it's not information that's particularly useful.

"We're the job board that forces you to jump through hoops to provide evidence of how active your hiring process is" isn't a sales pitch particularly likely to attract employers (in which case there's not much incentive for candidates to pay them either). And even if it did, it wouldn't replace Glassdoor' reviews - with all their inadequacies - for people that care more about whether the employer sounds better or worse than their current employer than whether they will get a job offer by the end of tomorrow.


> but since it doesn't tell you...it's not information that's particularly useful.

If it did it could be - if it could show how many applicants are actually qualified. Then a job board that listed more prominently jobs from responsive employers could get attention.

I'm not sure it's "jumping through hoops". In a well functioning company, providing evidence for a hiring decision is something that must happen anyway. How can the recruiter justify why they said yay or nay to a candidate? They might not remember the details but they're probably scoring candidates in broad categories. These categories tell you "90% don't have the right to work in the country never mind meet the job spec in other requirements".

If this data is available and isn't communicated to candidates there could be some other reason, likely that job boards make most of their money only from companies.


Expecting companies to do their candidate scoring on the job board's platform sounds like "jumping through hoops" to me. If I have a choice of job boards that let me and my team make my own decisions about filtering candidates and when and how to invite candidates to interview and one that expects us to use theirs, I'm probably not using the latter one.

Consistently identifying whether an individual has the right to work in a country from the semi-structured data in a resume is a surprisingly hard problem for software to solve (for n>100 countries, where the right to work is often a requirement not stated in the ad). And actual humans working in recruitment fail pretty spectacularly at "is this individual's programming experience appropriate" often enough

(There probably is a market for software that does that - and plenty of controversy when it turns out the AI process uses heuristics like "is his name Muhammad" to filter out 'unqualified' candidates - but not something you'd be likely to focus your effort on if your business is selling ads rather than HR systems ad buyers probably don't want to use)


I didn't say you can't make your own decisions. I didn't say it has to happen on the job board's platform either.

> Consistently identifying whether an individual has the right to work in a country from the semi-structured data in a resume is a surprisingly hard problem for software to solve

Many job posts provide a field that requires the candidate to enter if they have the right to work in the country the position is located.


then how do these behemoths go down? I'm sure there's a scenario or two where better solutions win eventually, is it all up to luck and time or is there an actual plan for that to happen


It's probably easiest for another company to pivot into this.

Say you have a maps / direction software with a very large userbase and decide to add a rating feature to locations. Or perhaps you have a professional social network and you add a feature to rate your employer.


LinkedIn.


Then they cannibalize the money they extract from recruiters to harass me for roles/services i have absolutely no interest in. Could also disrupt the power balance with companies paying for said subscriptions versus the free users.


Also, a cunning "growth hack" to improve user engagement ends up publishing employee ratings of their company in their colleagues' newsfeeds and sending out an email alert. This feature can be disabled by upgrading to LinkedIn Sales Navigator


I suspect it's because they make money off extorting businesses who get bad reviews and not off users trying to research companies.

Why have a geofilter if the vast majority of people are logging into the site to see if anyone is talking shit about $COMPANY and the only way to make money is if $COMPANY pays you off to take those posts down?


I'd point out that Glassdoor benefits from a huge breadth of search coverage across all their pages. Something on the order of 80%+ of their traffic I'd imagine comes from search. They have a huge leg up building up a ton of rank over the years (similar to companies like Booking.com). People Google this stuff and land on Glassdoor millions of times everyday, and that's what keeps it going.

One of those cases where distribution trumps product veracity. And to be fair, most people who use Glassdoor probably do still find it valuable (some information is better than no information).

Tangential point, but it's pretty crazy that Google can turn scores of companies to dust within seconds simply via search rank. Some sites that are > 70% search traffic: Genius, Yelp, Wikipedia, TripAdvisor, Quora, Urban Dictionary, Investopedia, Expedia, Glassdoor, IMDb


Isn't Blind kind of a Glassdoor disruptor?

I think Glassdoor may have lost trust by giving too much access(?) to employers that paid $$ to control how their companies were reviewed


Blind is even worse than Glassdoor. Last I checked, it had some kind of weird restriction about if you are not in a big tech or a well known startup, you can’t see any reviews.


Really? I can see reviews just fine.


blind is very big tech and US centric, it's not really a replacement for glassdoor


A good product opportunity isn’t always a good business opportunity.

Someone certainly could challenge Glassdoor on the product front, but as other commenters have noted, they’d be pitted against a well-entrenched incumbent, prone to legal challenges, and stuck with a convoluted business model.


Imo the entire service that it provides is not valuable enough to motivate competition, nor to justify whatever funding Glassdoor already raised. Hence why old Glassdoor is actually better than new


I honestly don't think that it's possible to make a review service that is any good at all if you're allowing the general public to submit reviews.


>> I honestly don't think that it's possible to make a review service that is any good at all if you're allowing the general public to submit reviews.

Glassdoor auth doesnt even work regularly. It has nothing to do with content, if a company cant figure out how to do auth on iOS, there is a serious problem.


> if you're allowing the general public to submit reviews.

Why? Do you believe it would accumulate spam and negativity because it's inexpensive for anyone to submit a review? What if there was a cost attached to posting a review?


For a ton of reasons. Fake or purchased reviews is a big part, but also the fact that honest reviews are often misleading or incorrect. As one example, I've often read reviews where the people submitted them before or immediately after they've received the product, and so can't possibly have enough information to write a useful review. Or reviews that are often skewed by price point, encouraging people to rave about crappy products simply because they were cheap.

Amazon reviews are a great example of all of these problems, I think.

If the reviews are about an employer, then there's a whole other level of problems with them. Perhaps the reviewer didn't get along with their manager and so have an honestly terrible opinion of the place, but really, the place is generally great. Or perhaps the opposite -- the reviewer got along wonderfully with their manager and so they're ignoring all sorts of real issues with that employer.

> What if there was a cost attached to posting a review?

I don't think that would help. It might even make things worse, because it would filter out more ordinary people and tilt the balance more towards paid reviewers.

I mean, I could be wrong here. But I've not yet seen a place that allows the general public to submit reviews where the reviews were helpful or accurate in terms of figuring out if the product or service is worth purchasing.


If you think of reviews not as score but subjective opinions and do not try to fool yourself by trying to come up with some magic function that extracts all the valuable data from them and nullify subjective judgement, they are valuable. The hard task is to read them and extract valuable information, so that sometimes the effort does not worth the result, but this one is also a bonus, since non-structured data is hard to game.


> If you think of reviews not as score but subjective opinions

I do think of them that way. But I have no context in which to interpret those opinions, so they are valueless to me. If I don't know what the reviewer values and doesn't value, I can't know how to interpret the review.

It's like with movie reviews -- a movie review from someone I don't know is without value, because I don't know what their tastes are. Did they dislike the movie because it's bad, or because they don't like that particular style? There's no way of knowing.

This all would be more workable if all reviews were honest, but they aren't. When you toss in the need to try to discern what reviews are real and what reviews aren't, the entire task becomes impossible.

This is why I've learned to ignore all reviews that aren't from someone that I am already familiar with.


I don't know the answer and can't speak for anyone else but I personally would not want to take on that risk. Every company that feels harmed by comments would be and endless flood of baseless lawsuits I think. Winning or losing the court battles would not even matter, it would just mean keeping a massive legal staff employed full time.

I could envision someone setting up forums on the darkweb to discuss the pros and cons of companies to mitigate some attribution but I do not see how anyone could make money unless they play ball with the big companies which renders the site pointless much like glassdoor as become. The money made would be from bribes from shady companies and then bigger companies might take the money away. Prior to GD was f'd company and I don't remember why they went away.


Well, FuckedCompany.com was mostly created to make fun of the stupid dotcoms and their implosion, much like Molly White's web3isgoinggreat.com


mostly created to make fun of the stupid dotcoms

It did start that way but they had forums where we could talk about all the shenanigans going on without being censored. AFAIK this is not really the same on GlassDoor as they censor comments or at least I've seen multiple people stating their posts were censored.

Perfect username to reply to this btw


Yeah, sure there were forums and back Pud certainly did not censor anybody.

But it was not really a business, that's what I meant. It was done for fun, and so that Pud could make a name for himself.

And later write a book!

My username: thanks. I was a great fan of Pud.

I even wrote the Italian 'version' of F'd Companies https://www.dotcoma.it/book


But it was not really a business, that's what I meant. It was done for fun, and so that Pud could make a name for himself.

That totally makes sense. Being just a forum and just for fun is less of a legal target I think, at least back then. That's cool you wrote a book about it, I had no idea.

Nowadays probably the only way to do this without censorship or taking bribes to promote or censor would be on a .onion site I think and it would probably get DDoS'd into oblivion.


> I do not see how anyone could make money unless they play ball with the big companies

One way is to play ball with the candidates. A big casualty of the job search is the candidate because they deplete their savings while looking for a job. Companies on the other hand can afford to not spend the money on a hire, or delay hiring, or put job postings that aren't really hiring to keep up pretenses or as marketing or feel they're not losing anything. The candidate doesn't have a concentrated force to counteract this.

If candidates wised up it could get interesting. A web app that is the other half of LinkedIn could be there waiting to be discovered.


The concentrated force for candidates to counteract this would be an industy-wide union. Which isn't happening.. but if you look at unionized industries this is one of the major benefits for candidates, or at least for candidates who make it into the union..


> The concentrated force for candidates to counteract this would be an industy-wide union. Which isn't happening.

It's one force, not "the" force. As described in the other comment, having enough data about which employers are actually hiring could go a long way.


IMHO levels.fyi, and to an extent Blind, are both semi-disrupting this. (You can see Glassdoor considers Blind a threat bc of their Fishbowl offering).

Both blind and levels are missing somethings though. Levels was just comp for the longest time but their forums are an attempt to move into this space. Blind has other issues with their user demographic and moderation imho - its a little too toxic at the moment to compete at the highest level. Theyre also too focused on tech and software careers alone.


As others have pointed out, growing any sort of two sided marketplace (in this case with companies and applicants) is really, really hard. On top of that, the quality of the reviews itself is very hard to regulate because Glassdoor isn't necessarily incentivized to promote the "truth" - they make money from the companies on their platform, not the applicants.

See related discussion recently: https://news.ycombinator.com/item?id=36634529


it's a feature, not a bug. The lousiness of it is it's main selling point. For instance, sortings by low ratings and ordering from most recent is a trivial SQL query, but if they unleashed that then it would be much harder to hide the negative reviews. A good portion of that $300MM revenue is from the ability to hide the negative reviews. Hence, the lousy interface is a money making feature. You can build the competition with a better UI, but you won't get the revenue.


Maybe the choice is not between one and many, but instead between one and none. If Glassdoor is so bad, then if it disappeared today, would you want it to come back? Bad things don’t necessarily need competition for best worst thing.


I’ve thought about this as I would like to know what GlassDoor is supposed to know.

I think the problem is that there’s no good incentive for people to give real info. The best reviews aren’t written because people are busy doing good work at good companies.

Jobs aren’t common enough to stroke egos like restaurant and travel reviews (that also have review problems). So busy people don’t write many.

Company PR isn’t that valuable other than knowing they have money to spend on PR.

And people are angry for just and unjust reasons.

Compound that job titles aren’t standardized, even within companies so a pay difference may be due to different jobs, different performance, or just fabrication.

I get the info I wish Glassdoor had through networking. It’s hard and requires much labor but is accurate enough to recruit people, and, I expect, if I wanted to apply to a company.

The other thing I tried years ago was an app called blackball back when LinkedIn first opened their api, and has since blocked. Blackball let me and all my contacts enter the names of people we worked with who we hated and would never work with for any reason. The app would then let me search whether any of my contacts had blackballed a person and return something like “John Smith was blackballed by 0 1st degree and 5 2nd degree” and did it anonymously to the people querying and queried.

The app was pretty handy, but didn’t work without the LinkedIn api and I think was probably illegal. But it saved the frequent “hey old coworker, LinkedIn says you worked with John Smith, would you ever want to hire them again” checks that can be tricky because I only give a real answer to people I really trust. Blackball was simpler than recommending because it’s a bit more binary.

Tl;dr; it’s hard to get good info in this space


Glassdoor is completely irrelevant now - especially in the tech industry.


I've found that the employer reviews on Blind have a much better signal/noise ratio. Not a big fan of the rest of the content on that app, but it's one thing they do right.


(how) do they make money ?


By extorting companies to improve their score by removing bad reviews or promoting good ones.


they list a bunch of stuff in thei paid features https://www.glassdoor.com/employers/features-and-pricing/

looks like you pay for advertisements of jobs, some more marketing features and some other crap


It doesn't sound like a great business to me, not to mention the risk someone else rightly mentioned here of being taken to court by companies.


It is extraordinarily difficult to win any kind of defamation lawsuit in the US. The bar is really high.

I certainly wouldn't try to run a business like that anywhere else in the world, but in America they're probably fine, with just some basic screening.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: