Choosing metrics that matter: the Top Two KPIs for Job Hackers

I’ve mentioned before that I’ve been slowly absorbing ‘The Book’ on modern web analytics, Web Analytics 2.0 by Avinash Kaushik. It’s taking a while due to the depth and detail which the author goes into on the subject, while covering everything from how to choose a web analytics platform (perhaps slightly dated now – the book was published in 2010) to the intricacies of how platforms define and track individual metrics. Don’t be put off by the exhaustive detail; the major strength of the book is that it outlines a universal process for understanding and using web analytics instead of presenting a one size fits all formula to copy. In that spirit, lets use the process to identify and understand the best metrics to use for job hacking, and share some tips on avoiding common mistakes Job Hackers may encounter along the way.

To combat the overwhelming amount of data gathered in the old “track everything” mentality, modern analytics best practices has shifted to finding one or two metrics that best represent the overall performance of your business. These Key Performance Indicators (KPIs) will be different for each website and each business, depending on the business model you are operating on or the specific outcomes and goals you are hoping to achieve.

So which KPIs should Job Hackers use to measure the success of their portfolio. I have identified two…

1. Bounce Rate

According to Kaushik, this is possibly the sexiest metric* we have in our toolkit. Bounce Rate measures the percentage of visitors who leave your site immediately without interacting with the page, and is a prime indicator of whether your website is good or bad. A high bounce rate (typically over 60%) indicates that people are either not seeing what they expected to see when they clicked a link to your website, or find it so terrible that they leave immediately. Kind of like accidentally clicking on a banner ad that leads you to a page selling herbal medications and free trips to Bangkok – you can’t close the window fast enough.

Obviously this is a bad thing for your portfolio. A high bounce rate implies that people are not staying on the page long enough to see any of your work, which is the whole point of having a portfolio in the first place.

Your website is bad meme

You should. You really should.

A healthy Bounce Rate is around 30%, if yours is significantly higher you may want to look at two things:

  • Check your SEO and meta descriptions for bad referrals. If you get a lot of search and referral traffic, does the meta data in your <head> give an accurate description of what’s on your site? Make sure you’re using keywords that are relevant to the content on your page, and use fluid and descriptive sentences that people can easily understand. For example, my <meta> tag looks like this:

<meta name=”description” content=”Creative portfolio of Ian Barnard, Digital Communications professional, located in Toronto, Ontario. I specialize in social media management and strategies, online marketing, web analytics and responsive web design.”>

Short, sweet, readable and to the point.

Protip: Google Analytics no longer considers meta keywords when indexing your page, so don’t waste your time coming up with an exhaustive list of key words or trying to embed them into the copy. Just write clean and descriptive content, make sure your keywords are in the title and body text, and let the Google overlords take care of the rest.

For more information on how Google handles SEO and page rankings, check out the 2013 Search Engine Ranking Guide from Moz.

  • Add event trackers to compensate for one-page websites or blogs. I explained this in detail previously, but basically if you have a one page portfolio with no internal links, then Google Analytics has no way of measuring how long a visitor stays on your page and will consider every session to have bounced.

If your bounce rate is still high after looking into both of the above steps, your website probably sucks. Sorry. At least now you know this, and once the pain subsides to a dull throbbing anger you can work on improving your portfolio so it’s less terrible. I have previously suggested some ways to create a good looking portfolio.

2. Time on Site

This may make experienced web analytics practitioners scoff, but hear me out. There are many fancy ways to measure a visitors engagement (and by proxy, their interest) in a website: Depth of visit, conversion rates, in-site surveys, click density analysis and so on. However, I argue that none of these are relevant or applicable to a simple one-page website.

Lets think about this for a minute. Depth of visit can only ever be one page maximum, as there are no other pages for a visitor to click to. In-site surveys seem a little over the top for a simple portfolio, and would probably annoy visitors; they are there to check out your work (ideally because you sent them there via a resume or cover letter), not to fill out forms suggesting how to improve your page’s experience. You could get fancy and set up conversion goals for contact “mailto:” links clicked, or create a contact form embedded in your page, but in my experience most employers prefer emailing or calling you directly from your application documents.

With that in mind, I believe that Time on Site is the best indicator of someone taking the time to really look at your work, which is a fairly reasonable indicator of how likely they are to request an interview for the job you’ve applied for. While bounce rates measure the visitors who left immediately, time on page will tell you who stuck around. For my site, I’ve created a goal in Google Analytics that reports every session that lasts longer than 30 seconds, which I figure is a long enough time for someone to have checked out my page, clicked a few portfolio modal links, and thought “Hmmm, this guy seems competent, lets give him a call…”

The best thing about this metric is that it’s simple and easily understood. My personal rule of thumb: if a metric takes more than 30 seconds to explain or justify, it’s probably not the best metric you could be using.

Warning: Time on Site is not always a good indicator of engagement! If you are analyzing a page with many internal links, Time on Site could also indicate that you have a terrible page layout or navigational issues. Time on Site can only measure how long someone spent visiting pages on your website – this doesn’t mean they’re necessarily having a good experience! A high average session time on site could also indicate that people are furiously trying to find specific content and failing, before giving up and vowing never to return again.

lost in a maze

Is this representative of your figurative journey through a website? Lets hope not.

Always be careful of reading too much into metrics, especially when trying to extract qualitative data from a quantitative metric. In the case of a one-page website with the goal of showcasing your work to prospective employers, average session time is a reasonable indicator of how positively engaged each session is, because most visitors were sent there directly and have nowhere else to go. For a different website, it could mean something else entirely.

So there we have it, the Top Two KPIs for your job hacking portfolio. A low bounce rate indicates that people are finding what they’re expecting for when they click a link to your portfolio, and a time on site longer than 30 seconds indicates that they like what they see. Both KPIs combined imply that you should be getting an interview request call fairly soon, which of course is the ultimate goal of all this work.

Do you think I’m nuts? Are there better metrics out there for job hacking portfolios? Let me know in the comments below.


* As an aside, I never expected to find such a large number of sexual and romantic innuendos in a book on data analysis. Not complaining, but after the sixth cheeky reference to kissing someone you begin to question the author’s grip on relationships.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s