I had the pleasure to talk with David Bernstein, VP of Big Data from eQuest at the HR Technology show and then had a follow up conversation with him in December. The below is some of that interview.
First of All, Who is David Bernstein?
David is the Vice President that eQuest has brought in to figure out what to do with all of this amazing data they have. From our conversations - there are a couple of key take aways that I had:
1. David, first and foremost, is a HR practitioner! He has a long history of being a HR practitioner and is truly focused on figuring out how to develop something that will benefit the HR community.
2. He is very bright and tasked with trying to convert their data sets into something that is also of commercial value to eQuest.
3. Lastly, he is passionate about what he is doing!
Big Data - What is it?
eQuest has this amazing business where they distribute jobs from companies to job boards, and they click response rates. They do this with about 5 million records a week, and have been doing that for 19 years. Aside from the raw data, here is what they realized they had at a Macro - big data level.
So they have lots of very cool data. For my readers that are numbers folks - they have over 1.1 BILLION records. And you guessed it, they are having to build some pretty cool systems to work the data - no more Microsoft Excel for these folks!
What good is it?
This is where David gets really, really excited when talking - as there are all sorts of applications. However, thankfully for eQuest, David also has a background as a product manager - so he is being measured about his focus and not getting distracted and chasing every potential aspect. Here is what they are currently working on.
Optimized Ad spend:
David's first focus is to help a company hire efficiently and save money. Given the data they have, they can help "predict" where postings should be done by job or job type. So they can say - if you want to get great applicant flow on a Project Manager position - put it on your internal site, twitter, but pass on Career Builder (making that example up).
Employer Brand Analysis:
This area got me equally jazzed - as they can see, and run analysis showing a company's traffic and compare to the other factors to see if you are lagging. There is nothing better to measure the Employer Value Proposition than seeing how applicants vote with their mouse. Not only does this compare company name, but they can identify if other factors are impacting you - such as the job description, the title, etc.
How do they do it?
David pointed out that Data without context is not only not helpful, but can be erroneous. He strongly believes that you have to understand the client and their business to make big data helpful. So as opposed to making the offering a pure self-service offering, eQuest assigns a business analyst who works with the client to really dig into the data. He explained that the client will get the benefit of the analysis with context, at a fraction of what the company would pay if they tried to hire a full time employee to do the analysis. Furthermore, he found that a lot of employers want the data, but rightfully can justify bringing in a full time certified statistics beanie weanie to data mine their employer data.
A bad case of Data Envy:
So as my closing question, I asked David what the coolest thing they have seen in the data. Was there a strange finding? He laughed and said they had just finished some analysis to understand the patterns on how and when people read and respond to job Ads and they found some pretty interesting trends. A year ago, they were seeing lots of job searching from 10 am to 2 pm - monday thru Friday. But now, they are also seeing a second large bump from 7pm to 9pm. So there is definitely a lot of searching going on. If you want to read more on this - here is the link.
The Funniest finding he has seen so far is the response rate by device type and operating systems. He said that even though there are a TON of iOS devices out there, the folks that actually apply for jobs on mobile devices are mainly the Android users. This seems like perfect fodder for jokes about Android users are smarter, etc..
First of All, Who is David Bernstein?
David is the Vice President that eQuest has brought in to figure out what to do with all of this amazing data they have. From our conversations - there are a couple of key take aways that I had:
1. David, first and foremost, is a HR practitioner! He has a long history of being a HR practitioner and is truly focused on figuring out how to develop something that will benefit the HR community.
2. He is very bright and tasked with trying to convert their data sets into something that is also of commercial value to eQuest.
3. Lastly, he is passionate about what he is doing!
Big Data - What is it?
eQuest has this amazing business where they distribute jobs from companies to job boards, and they click response rates. They do this with about 5 million records a week, and have been doing that for 19 years. Aside from the raw data, here is what they realized they had at a Macro - big data level.
- They can see job creation before anyone else! As a result, eQuest can tell by sector who is starting to hire. Compare this to ADP and their payroll numbers - which show who has been hired, and getting paid. As a result eQuest is about 90 days ahead of that.
- They can see job seeker demand since they can see response rates to the ads.
- They also have a ton of other variables that are interesting such as location, the company name, the job title, the job description, etc.
So they have lots of very cool data. For my readers that are numbers folks - they have over 1.1 BILLION records. And you guessed it, they are having to build some pretty cool systems to work the data - no more Microsoft Excel for these folks!
What good is it?
This is where David gets really, really excited when talking - as there are all sorts of applications. However, thankfully for eQuest, David also has a background as a product manager - so he is being measured about his focus and not getting distracted and chasing every potential aspect. Here is what they are currently working on.
Optimized Ad spend:
David's first focus is to help a company hire efficiently and save money. Given the data they have, they can help "predict" where postings should be done by job or job type. So they can say - if you want to get great applicant flow on a Project Manager position - put it on your internal site, twitter, but pass on Career Builder (making that example up).
Employer Brand Analysis:
This area got me equally jazzed - as they can see, and run analysis showing a company's traffic and compare to the other factors to see if you are lagging. There is nothing better to measure the Employer Value Proposition than seeing how applicants vote with their mouse. Not only does this compare company name, but they can identify if other factors are impacting you - such as the job description, the title, etc.
How do they do it?
David pointed out that Data without context is not only not helpful, but can be erroneous. He strongly believes that you have to understand the client and their business to make big data helpful. So as opposed to making the offering a pure self-service offering, eQuest assigns a business analyst who works with the client to really dig into the data. He explained that the client will get the benefit of the analysis with context, at a fraction of what the company would pay if they tried to hire a full time employee to do the analysis. Furthermore, he found that a lot of employers want the data, but rightfully can justify bringing in a full time certified statistics beanie weanie to data mine their employer data.
A bad case of Data Envy:
So as my closing question, I asked David what the coolest thing they have seen in the data. Was there a strange finding? He laughed and said they had just finished some analysis to understand the patterns on how and when people read and respond to job Ads and they found some pretty interesting trends. A year ago, they were seeing lots of job searching from 10 am to 2 pm - monday thru Friday. But now, they are also seeing a second large bump from 7pm to 9pm. So there is definitely a lot of searching going on. If you want to read more on this - here is the link.
The Funniest finding he has seen so far is the response rate by device type and operating systems. He said that even though there are a TON of iOS devices out there, the folks that actually apply for jobs on mobile devices are mainly the Android users. This seems like perfect fodder for jokes about Android users are smarter, etc..
No comments:
Post a Comment