CV analysis can be one of the most time-consuming parts of a recruiter’s job, requiring considerable skill to perform accurately yet quickly. Thorough curriculum vitae analysis is necessary to ensure recruiters don’t miss good candidates or put forward incorrect candidates; but it is also important to process all applications quickly to ensure a fast response for clients and move to the interview and placement stage.

As the number of applicants per position rises and recruiting resources stay the same, or even reduce, many recruiters are turning to resume analysis software to help. But can automated curriculum vitae analysis really offer the accuracy and intelligence required by today’s recruitment environment?

The Challenge of Curriculum Vitae Analysis
When a human recruiter is performing resume/CV analysis, they are actually doing two things at the same time. First they are processing the data on the CV to understand what it says (this is known as parsing). Second, they will have a specific vacancy or set of vacancies in mind and they will be assessing whether the CV they are looking at could be a good match (whether this is a candidate that should be offered to a client for consideration).

Successful CV analysis software therefore has to perform both accurate CV parsing and CV matching, both of which are challenging tasks for a computer. A human learns through experience that ‘July 31 2013′ means the same as ’31st July 2013′, ’31/7/13’, ‘31.07.2013’ or even ‘7/31/13’ if the applicant is from the US; a computer similarly has to be trained to understand all these different formats. Even more tricky is the change in meaning that can occur due to the context and the surrounding language. For instance, if a particular job requires familiarity with Excel, you want to include CVs that say “Expert in Excel” or “Excel 2007”, but exclude CVs that say “I excel at customer communication”. Simple keyword spotting, for instance, therefore cannot achieve the level of intelligence required for reliable curriculum vitae analysis, struggling to get past 70% accuracy.

Most parsers use either a statistical approach, which requires training using a large number of example resumes in order to generate accurate, context sensitive models, or a grammar approach, which requires a large number of linguistic rules to be encoded and made available to the parser. These approaches by themselves can achieve around 85% or 90% respectively – both falling short of human accuracy levels of around 96%. DaXtra Parser, by contrast, can achieve close to 95% accuracy, by using a sophisticated algorithm combining both statistical and grammar-based parsing.

The Operational Importance of Accurate CV Analysis

When comparing accuracy rates of 85%, 90% and 95%, the differences between them may not appear very great, but on a day to day level these differences have a big impact on operational performance of a recruiting operation.

Looking at it a different way, a 95% accurate resume parser produces a 5% error rate, while a 90% accuracy means doubling the error rate to 10%, and an 85% accuracy means a trebling of the error rate to 15%. As error rates increase, recruiters are less inclined to trust the output of the parser, and the temptation rises to “check” by manually reading CVs – which of course undermines the time savings of using the parser.

An accurately parsed resume is of no operational benefit if it cannot then be easily and correctly matched to job vacancies. DaXtra Search is designed to allow quick and, if desired, automated searching and matching against current vacancies. It allows recruiters to use natural language expressions to search for matching CVs, and to set these searches to run automatically against new applications that come in, notifying consultants of any new matches. DaXtra Search also ranks the results of searches, showing which CVs are the closest match to a job vacancy – greatly aiding the process of compiling a shortlist of candidates. It also allows resume matching in the opposite direction – taking a new resume and checking its fit against current job vacancies. This is a great advantage in larger recruitment operations where no one consultant can be expected to keep all current vacancies in mind when reviewing a new CV.

DaXtra Parser & DaXtra Search – A Complete Resume Analysis Solution

Between them, DaXtra’s Parser and Search modules offer all the curriculum vitae analysis functionality required to take a recruiter from raw CV’s to a shortlist of candidates, quickly and accurately.

But DaXtra’s software offers recruiters a great deal more than simply accurate CV analysis. It is designed to integrate with your existing systems and databases, and uses intuitive methods for interacting with the user interface, all minimising the time and effort required to start using and benefiting from DaXtra. It also supports your preferred workflow, rather than requiring consultants to adopt a new way of working.

We offer all our clients the chance to test DaXtra against their own data, so you can see for yourself how DaXtra software can benefit your operation. Curriculum Vitae Analysis – Contact us today to request a demonstration.