When most people hear the term A/B testing, they think of sales and marketing. Most of today’s businesses use A/B testing to measure website performance, landing pages, SEO campaigns, emails, and so on. But that’s not the only thing it’s good for.

A/B testing can also be applied to writing SaaS sales job descriptions. When used correctly, this allows you to gain data-driven insights that can help refine every element of your job descriptions to attract better candidates, fill positions quicker, and increase salesperson retention.

Here are the nuts and bolts of the A/B testing approach to help you write the best SaaS sales job descriptions possible.

Figure Out What Needs Improving

Before doing anything else, you first need to pinpoint exactly what you’re looking to accomplish.

For example, maybe you’re not pulling in the volume of candidates you need. Here your goal would be to improve your job descriptions to make your company more appealing to increase the number of applications you receive.

Or maybe you’re currently getting a high volume of SaaS sales candidates with your job description, but the quality level is lacking. In that case, your goal would be to improve the collective quality of candidates.

Whatever the case, figure out precisely what needs improving. Also, it’s usually best to focus on one area at a time when A/B testing SaaS sales job descriptions, as trying to fix more than one area at once can be overwhelming. Once you’ve made your initial improvement, you can move on to other areas. But stick with one element at first to keep things simple.

Choose Testing Variables

After you’ve identified what needs improving, it’s time to determine the specific variables to test. This can be a little tricky, given how many different components make up a SaaS sales job description. Here are the core 10.

The key to effective A/B testing is choosing variables you believe are most instrumental in helping you achieve your goal. If, for instance, you were struggling to bring in enough qualified candidates, you may want to focus on skill sets and education qualifications.

This should help weed out more underqualified candidates so the majority of the ones who apply have the qualifications you’re seeking.

Develop Two Test Versions

Now it’s time to create two test versions — a control and a variation — to see which performs the best.

Let’s go back to our example about trying to improve candidate quality. In an attempt to drive more qualified candidates, you could try two different test versions of the skill set section of your job description.

Version A could be a shorter section, which covers the essentials but doesn’t go into much detail.

Version B could be a longer section, which is more specific with the skills a SaaS salesperson needs to thrive at your company and goes into greater detail.

A potential hypothesis would be that version B would result in more qualified candidates, as it’s more exhaustive in terms of the skills your ideal SaaS salesperson should possess.

Generate Measurable Data

Once you have your two test versions, it’s time to put them to use in a real-world setting and start generating data. For our example, you could split the two job description versions 50/50 so half of the the candidates see the shorter job description version and the other half see the longer job description version.

This would be a simple way to objectively compare the two versions to see which yields the best results.

The more data you have, the better. Therefore, you’ll want to run testing in as many places as possible. So, for example, instead of just doing testing on job descriptions on your website’s career section, you may want to branch out to job boards, email, or wherever else you recruit.

Analyze the Results

This is the fun part. After you’ve accumulated enough data, it’s time to see what the results are. If you were trying to determine whether a shorter or longer job description was best for reeling in quality candidates, you could analyze application quality, specifically looking at experience, skills, and cultural fit.

Say 75% of the candidates that applied to version A — the shorter list of necessary skills in the job description — were highly qualified and 85% of the candidates that applied to version B — the longer list of required skills in the job description — were highly qualified. Based on those findings, you could surmise that version B had the bigger impact and helped bring in more highly qualified candidates.

That would mean that being more detailed with the required skill set should reliably produce higher-quality candidates.

This example is arbitrary, but you get the idea. Other key metrics you could analyze include the number of applications you receive after A/B testing, the average time candidates spend on a job description page, and the click-through-rate.

Nearly any element you can imagine can be tested with this technique, and a side-by-side comparison should give you objective insight into what works best.

Use Your Findings to Refine Future SaaS Sales Job Descriptions

Knowing for certain what it takes to optimize your SaaS sales job descriptions can be incredibly exciting. Rather than guessing what works, you know for a fact based on concrete data.

While finding the “low-hanging fruit” and improving a critically flawed area of your job description is good, that’s only the start. The key to being successful is continually performing the process until you’re able to write job descriptions like a well-oiled machine.

I suggest making a list of what needs your attention the most with the biggest issues at the top and the lesser issues toward the bottom. That way, you’ll know what to prioritize, and you can move through the list systematically until you’re firing on all cylinders.

If you’re looking to find the cream of the crop sales reps based on critical capabilities like the will to sell and trainability, check out The Original Sales Assessment. It can be fully customized to contour to your unique sales role to quickly filter through the candidate pool and find elite talent.