Using Wordle to filter CVs

I have been reviewing CVs (resumés) for a client. The vacancy is for a Senior Business Analyst in the Pega domain. During the course of reading the CVs, I noticed that many of them, while they used the phrase “business analysis” throughout, had few or no examples of any tasks or achievements that would fall under that heading.

Having read all of them, I organised them into one of the following three piles (in descending order of interest to me):

  1. Clearly and accurately describes the activities of a senior BA. Worth interviewing.
  2. Describes the activities of a BA but without detail or clarity. May or may not properly understand the role of a BA. Certainly does not know how to express it in writing, so unlikely to be a decent senior. Worth an initial telephone interview to be sure.
  3. Has some core skill other than business analysis (e.g., project management or programming) and has attended some requirements workshops. The details provided in the CV do not support any claims to be a BA, never mind a senior one. Clearly does not understand what BAs really do. Not worth interviewing and wasted my time in getting me to read the CV in the first place.

Considering the waste of time involved in reading the category threes, it dawned on me that Wordle might have given me a strong clue as to whether a CV contained the right words to make it worth my while reading. At the very least, it might give me a sense of the order in which I wanted to read them.

For those of you who don’t know the Wordle website, it is (in their own words):

a toy for generating “word clouds” from text that you provide. The clouds give greater prominence to words that appear more frequently in the source text

Far from being a toy to me, I figured it could be a useful tool. I decided to Wordle the CVs I had already categorised to see if my assessment of each word cloud matched my assessment of the CV itself.

Here are examples of the notes I made after assessing the word clouds:

  • Based on the prominence of the relevant key words and supporting words, I would read this CV. In fact, I did read it and would consider this person a candidate for interview.
  • The keywords are prominent but the supporting words are less so. I would read this CV and having read it I found that it lacks enough detail about the person’s experience as a BA. I would recommend an initial telephone interview to ask some filtering questions.
  • Although the desired keywords are prominent, the supporting words are not. The prominent words indicate a project management profile. As a result, I would read the CV, but not as a priority over others with more robust word clouds. I did read the CV and would not consider this person a candidate for interview because his only actual experience of analyis appears to be “feasibility analysis” and where he mentions “technical analysis” as a skill, he does not mention how he applied that skill.
  • Although the keyword “Business” is quite prominent, the word “Analyst” appears to be much smaller, indicating that little of the CV relates to that role. None of the supporting words we would expect to see for the role are obvious either, also indicating that this person is not primarily an analyst, never mind a senior one. Based on this word cloud, I would not even read the CV. However, I did read it and the CV itself bears out my assessment and I would not interview this person.

I wanted to see whether word clouds could be a useful tool in filtering CVs when you have a lot to get through. This little experiment tells me that they are. I will continue to use them for a while in parallel with reading all CVs to see if they truly are a good initial filter.

After this very successful experiment, I tweeted the following on my @AnalysisFu account:

Screening BA CVs for a client. Used wordle.net to quickly scan. Reveals most CVs can’t back claims. Reading CVs confirms this.

This was retweeted by @Rowan_Manahan, careers consultant and author of Where’s My Oasis? and Ultimate CV – a resounding endorsement of my tweet if ever there was one.

Rowan’s retweet got the following reply to both of us:

I’m sorry but your analysis is lazy and subjective

Reviews of CVs are never objective. They reflect the experiences and biases of the reviewer. For example, if I value the experience of employees of one company over another, then I will favour CVs which reflect that experience. If I favour people who can spell over people who cannot, then I will prefer CVs which demonstrate good orthography. It is naive to think otherwise, so I agree that my analysis was subjective.

However, my analysis was far from lazy. I was quite thorough in reading all the CVs in addition to reviewing the word clouds, as stated at the end of my original tweet: “Reading CVs confirms this”.

Not reading my original tweet in its entirety: now that’s lazy.

Might I be “lazy” in the future and use word clouds to eliminate totally unsuitable CVs from my desk? I might indeed, if continued experiments convince me that the approach is accurate. I have no small experience of running highly successful recruitment programmes and in my experience, many candidates (and their agencies) chance their luck in applying for jobs for which they are not suitable. Moreover, many project managers and programmers think they are automatically BAs because they have attended requirements workshops and have “soft skills”. If they read the Business Analysis Body of Knowledge and the IIBA Competency Model, they would realise this.

Knowing that, I would be some sort of fool if I didn’t want to use a tool that would help me eliminate the time-wasters. If some people want to label me “lazy” as a result, I’ll take that hit.

You can find my own CV word cloud below.

Kind regards,

Declan Chellar

Twitter
Visit Us
LinkedIn
Follow by Email
RSS

Leave a Reply

You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

  

  

  

This site uses Akismet to reduce spam. Learn how your comment data is processed.