Resume reading, filtering and candidate selection for interviews will be one of the major tasks for any manager running a reasonably sized team.

1. Our recruitment manager, and his team must each be spending 2-3 hours each searching scanning resumes
2. I and the other managers spend some time thinking about our requirements, explaining them to our HR team
3. I and other managers again spend time reading, scanning filtering resumes that are discovered b our recruitment team
4. And this is a activity that never ends in the company.
5. On an average something like 5 man hours are spent everyday across the company scanning and filtering resumes
6. And YASU is a relatively small company with over some 120-140 employees, I think
7. You can now extrapolate this to a large size organization constantly hiring new members.

I was reading some where today, don’t remember where, that in 2007, it is projected that HRMS investments are going to shoot up, and I was wondering if business rules can help?

Well, lets understand the resume filtering process itself. Here are the steps.

1. Search job websites like Monster using basic keyword search
2. Download the identified resumes
3. Scan resumes based on preliminary criteria as specified by the respective manager
4. Send the resumes that pass this level 1 filtering to the managers
5. Managers spend time reading the resumes and making their selections. This is level 2.
6. Then take these resumes for serious processing and start calling the candidates.

Here is what I think a better way (using Business Rules) to improve this whole process and reduce a lot of manual work.

1. I am sure the job sites will be able to provide a resume search Web Service that supports key word search.
2. Most probably these job sites will serve resumes confirming to the Resume and Candidate XSD as specified by the HR-XML schema
3. Build a Web Service Client that will be able to look up the Job Search Web Service and download the HR-XML compliant resumes.
4. In the next step, we will build Business Rule driven filters to do a second level filtering of the resumes.

For example, check out these rules

1. If Candidate’s Roles & Responsibilities for 3 or more projects is repeated verbatim, then Reject Resume
2. If Candidate has worked for less than 3 years, and claims to know more than 10 technologies, then reject
3. If requirement is for DB expertise, and candidate has worked mostly on ASP.NET only, then reject
4. If Candidate has less than 3 years of experience and has worked in more than 3 companies, then Reject
5. If Candidate is older than 33 years, then reject
6. If Candidate’s experience is greater than 5 years and salary less than 60000$, then reject
7. If candidate’s experience is greater than 5 years, and expectations>90000$ reject

So, what we are talking about here is to capture the resume selection criteria used by managers, and capture them in a form that can make automating this process easy. And you need a Business Rules engine to do this for you.

If I have a requirements, I will supply the keywords, and my rule filters to the system.The system will download the resumes using my keywords, and will further narrow the search by applying the rule based filters that I have specified on the selected resumes.

This system will, in my opinion, directly reduce so much manual work that is done by recruitment teams and managers that is best spent doing other important things.