The Wall Street Journal enlisted two professionals to test the capabilities and accuracy of four job sites that use matching technology to connect job seekers: Jobfox, CareerBuilder, Bintro, and Trovix.
After completing the initial set-up that each site required, the testers compared the resulting job suggestions and concluded that only a handful of the jobs were relevant.
For Bintro, a matching site in beta, the job seekers spent 30-45 minutes filling out a questionnaire. One of the seekers, an IT pro, wasn’t matched to any jobs. The other seeker, a marketing professional, was only given two referrals.
With Jobfox, the IT pro was matched with 35 ‘pretty accurate’ jobs. The marketing professional was given two jobs, neither of which were a good match.
Trovix dealt a variety of jobs to the testers, but both complained that none were relevant to their experience levels.
With CareerBuilder, the professionals liked the in-depth search parameters, but the results were also disappointing. The jobs they were matched to failed to fully represent their credentials.
So why are these sites falling short? Is it kinks in those proprietary algorithms embedded within each site, or is simply a lack of relevant jobs? What matching sites have you found to be the most accurate?
Popularity: 7% [?]
July 9th, 2009 at 9:05 am
I’m shocked! Shocked, I tell you! To think that a computer algorithm can’t properly match candidates to jobs better than that job seeker could.. What’s the world coming to when you can’t rely on Jobfox or Trovix to make decisions for you.
July 9th, 2009 at 11:16 am
I am not-at-all surprised with these results. Until now I never understood why VCs are gambling their investors money by putting it in such technology companies. I would guess, founders of those companies must have done a good job by selling their resumes/ideas and promising investors what they wanted to hear – $1b+ market opportunity.
I feel sorry for employees at these companies who are dreaming of cashing out on their stock options.
July 9th, 2009 at 11:36 am
It’s interesting that people that test “Matching” technologies do so from the job seekers POV by creating a job seeker account. Maybe the testers DO NOT realize that the job market is NOT driven by supply (job seekers), it is driven by demand (employers). When there is a mass shortage of jobs all around the country and in every industry and metro, how can you test how well a system matches by creating a candidate account? “I created a profile for a technology professional and didn’t see any jobs therefore the system isn’t matching correctly”. WRONG – there are not technology jobs in the metro area you choose to build the candidates profile in. So if you’re going to test a system, you need to test both sides. If you’re only going to test one side (especially in this market), test the employer’s side. Create an employer account and post a job to see the quality and quantity of candidates you get and how well they suit the job. That’s how all these systems are designed anyway! We designed the Realmatch system to match supply to demand…no one can drive demand with matching…these aren’t magic machines. The Wall Street Journal did report the employers POV with regard to using a matching technology 2 months ago, dont know why they didnt mention that in this article –
http://online.wsj.com/article/SB10001424052970204475004574126832685403014.html
July 9th, 2009 at 12:24 pm
Rafael – If Realmatch deisgned the system for employers instead of job seekers, why would the job seeker even use Realmatch? You have to have both.
July 9th, 2009 at 1:14 pm
Dagwood – I’d agree if you are a destination site then you need to have both supply and demand on your website but we are not a destination site. We do not spend money or resources trying to get job seekers to our site, instead we only drive employers to Realmatch to get their jobs. Then we take the jobs and get job seekers to apply to them on other sites like thejobnetwork.com, phillyjobmatch.com, findkyjobs.com etc etc etc. Forgetting our model for a minute, the entire industry is driven by demand but few recognize this.
I keep reading advice for job seekers about how they should call recruiters and headhunters to try to find a job and I think REALLY? Is this really what people think Headhunters do is find people jobs? To quote our chairman Alan Schoenberg who basically pioneered the entire recruitment industry, “Headhunters do not find jobs for candidates; they find qualified candidates for their employer clients”. My point is that you cannot test a system which is designed to drive candidates to employers by looking at it from the candidates POV because the entire industry is driven by Employers.
July 9th, 2009 at 6:10 pm
Seriously Rafael – enough with your self promotion. Why do you find it necessary to throw in your 0.02 and a link from probably the only article ever written that included your company? Note to you: They already did an article on the employer POV – this article is about he seeker POV. You only make money by pay per hire, right? The cost effectiveness wasn’t an issue to the seekers in this article. You missed the point completely. I am certain if your site were also tested the seekers would have the same issues.
I have found NO matching site to be at all accurate or worth the time it takes to fill out a profile. On Jobfox I finally gave up creating a profile after 30 minutes and I still get tons of spam from them and no matter how many times I request to unsubscribe I still get spam. The same with Careerbuilder and Trovix. It’s like a battle to see who can get the most traffic. Based on those facts I have no desire to try to use any of the other matching job sites.
Recruiters are always complaining about job seekers and how they send them resumes when they aren’t even qualified for the job. Who writes up the job description? What is posted is usually extremely narrow and generic at best. So now it’s the seekers fault that the job match results are weak? What about the seeker who does get an accurate match but still never hears from the employer? Matching is not the answer for the job seeker and if it’s all about the employer why should a seeker waste their time?
July 9th, 2009 at 8:51 pm
Someone will eventually get matching technology right, we’re still at the beginning people. The real challenge isn’t for the jobseekers who are eager to weed out all the crap they usually get – it will be getting the recruiters out of the cut, paste and fill up the inbox mode most (not all) have been in for years. It will take a little bit of chutzpah on both sides to try something new in the attempt to become more efficient.
July 10th, 2009 at 10:13 am
Matching is dead – If you dont like looking at comments or links, try reading a newspaper.
July 10th, 2009 at 11:20 am
the whiner has spoken, I guess we should all go sulk now – if everyone had your attitude no business would get done and no one would be persistent enough to achieve new things
July 11th, 2009 at 5:01 pm
Very interesting discussion. I think matching will eventually be done right and could take away a large chunk of the headhunting business. However, it is nearly impossible to get a 100% success rate. For me a 80% success rate would be amazing. The problem with JobFox etc. is that they are going for quantity right from the start. In my opinion these sites should go for quality first (that’s what the human resource department cares about). When they have established a good matching quality, scaling is easy…
July 13th, 2009 at 9:48 am
good points manuel, I don’t think jobfox will be the one to figure it out as they seem to be going in all different directions away from matching but more pinpointed results in a sea of resumes will be quite valuable, I also agree with the point that hhters will need to be willing to try something new
July 15th, 2009 at 12:44 pm
I was wondering if anyone ever heard of this website Uvisor.com? I heard it is a job matching website that has the potential to be the first big hit. Anyone have any insite on this website?
July 15th, 2009 at 1:58 pm
BIGJIM – I can tell you the traffic for the URL uvisor doesn’t even register on compete so no one is going there at all, no one. Compete usually register if at least 500 people a month are there.
http://www.compete.com/
Uvisor.com has absolutely no inbound links according to Google …so no one is linking to them.
http://www.google.com/search?sourceid=navclient&ie=UTF-8&rlz=1T4GGLL_en&q=link%3awww.Uvisor.com
There are only 4 pages of uviros.com indexed in Google
http://www.google.com/search?hl=en&safe=off&rlz=1T4GGLL_en&q=site%3Awww.compete.com&aq=f&oq=&aqi=
There are no reports of funding, product reviews or any other mention from any media org
http://www.google.com/search?hl=en&safe=off&rlz=1T4GGLL_en&q=uvisor&aq=f&oq=&aqi=g2g-s1
So you ask if we have insight….yes, there is a ton of evidence that there is absolutely nothing happening with uvisor.com. Sorry to be such a cynic but you have to come here with a much more refined pitch then that sweetpea.
July 21st, 2009 at 1:44 pm
Good topic, very thought provoking
August 10th, 2009 at 1:50 pm
I think I’ve solved the problem of “matching” you have described above. No matter what your opinion is I would love to have your feedback!
August 10th, 2009 at 7:43 pm
AK, wow. I got set up in less than 5 minutes. I felt like I went on Craigslist, entered a short list of criteria, and presto, every applicable Craigslist post in the world was put on the screen, with your percentage value.
I think you have provided intriguing SOLUTIONS to the problems listed in the article. GJ
August 11th, 2009 at 5:26 am
First let me thank you for your kind words Merl. I started GIGS at Spartan Yard out of frustration with getting my consulting company noticed to the Craigslist crowd. After posting a profile of my services that I offer I noticed by the end of the day I was buried in over 100 links and sometimes carried over to a second page which in everyway guaranteed that I would never be seen. I then tried to get crafty and started posting “gigs” in CL that asked for affiliate sales reps to join my team to help get the word out about my services. By the end of the day I was getting emails from web development companies stating they could help revamp my website. Right away I knew these communications I was receiving were the result of CL RSS feed extractions.
I’ve specialized in building user interfaces and analyzing user behaviors over the past 12 years so right away I know how cumbersome and frustrating form submission processes can be online. That’s the primary reason why I made sure filling out a form as a poster or as the talent never took any longer than 5-7 minutes plus the form process never carried over to another page. I haven’t looked at the other companies’ processes that are outlined in this article as I really don’t want to. My solution was geared towards a problem the “laymen” faces everyday while using one of the internets most long standing posting environments. I’m not a recruiter. I’m not a human resources department. I’m an individual that wanted my god given talents noticed without spending anything outside a nominal amount of cash to do so.
With all that said Merl, can you please be more specific as to what other solutions you feel GIGS at Spartan Yard has solved that were outlined in this article. I gave 1 year memberships to over a 1,000 Slashdot users last week to help me unit test the environment. I had great feedback from them concerning technical and flow aspects of the architecture but now I’m hoping I can get some real-world feedback. Positive or negative feedback is welcome as in no way do I want you to think I have a perfect solution. Thanks in advance.