Recruiting targeted beta testers is one of the most difficult aspects of beta testing. If your beta testers aren’t a true reflection of your target market, then you risk ending up with irrelevant and distracting feedback that could veer your product off course. This becomes even more difficult if you have a diverse market, or if your target market is defined by a number of different demographic and technical factors.
How can you make sure that your beta testers not only reflect your target market, but also cover the breadth of that market? There are a couple of basic methods: personas and vectors. These two methods both help you to strategically pick beta testers to ensure that your beta tester team represents all the different factions of your market. Each has pros and cons that you’ll need to consider as you build your recruitment strategy.
Note: The information in this post is from our Beta Tester Recruitment Kit. Personas and vectors are just part of the puzzle of building a solid recruitment strategy. If you’re looking to build a complete plan for finding and selecting great beta testers for your next beta test, this kit will give you everything you need.
Method 1: Personas
Personas are a technique borrowed from product marketing in which you group users into named segments (e.g. John, Jenny) based on demographic traits, each representing a portion of your target market. They’re a great way to funnel the right applicants into your test, if you’re able to express your recruitment criteria as 3-5 static buckets.
The first table below shows how you might break down how many beta testers you need for each persona. The second table outlines the different geographic, demographic, and technical requirements for each persona.
As you review and select applicants for your beta test, each will fill a spot in the appropriate persona. Once a persona is full you’ll focus on the remaining personas, thus ensuring that your beta tester team accurately reflects your market breakdown.
The primary pitfall of this method is that it doesn’t always allow for the diversity needed in your tester team. Fitting your beta testers into a handful of predefined buckets may not provide the coverage of testers and environments that you’re looking for. In the situation above, for example, you could end up with only testers using Windows 7, since all of your personas allow for members to have any of the three operating systems. To avoid this you could assign a single OS to each persona, but this would make finding testers that fit all your persona’s qualifications more difficult. You’ll have to decide if ensuring the diversity is worth making your recruitment more challenging.
Method 2: Vectors
The other option is the vector method. In the vector method, each criteria is its own vector, which can be filled independently of other criteria. As we mentioned, the problem with a strict personas approach is that — in many cases — it fails to cover enough ground to fulfill your objectives and accurately represent the breadth of your target market. If this is the case, a vector-based recruitment strategy may be the ticket.
In the example below, we’re specifically looking for education level and operating system — both entirely disconnected from each other. Each tester fills one column for each requirement independently. So you could have a tester who’s a college graduate using Windows 7 and a tester with a high school diploma using Windows 8.1. Both are qualified.
Sometimes, however, two of your vectors are related to each other. These are called tiered vectors. In the example below, we’re trying to gather an even mix of both male and female testers in each of the three geographic markets where our product is sold. So, the gender and geography requirements will consequently make up one tiered vector.
If you have an objective where two requirements are interrelated, then you’ll use tiered vectors to ensure that you have adequate coverage and can meet that objective. If no relationship exists, you can keep them separate as individual vectors (which are typically much easier to fulfill).
After you’ve laid out the vectors for each of your geographic, demographic, and technical requirements, you can start selecting candidates that fit your requirements. You’ll likely find that the easy buckets fill up more quickly than the tough ones. One challenge associated with using vectors is that as slots are filled, the remaining requirements become increasingly more specific, thus more difficult to fill. This inevitably leads to the final slots being the most difficult to fulfill (each one more “specific” than the last).
For example, if we were recruiting for a test using the vector tables above, we might end up needing two French women with Master’s degrees using Windows 8.1 to fill the last two spots in the test, which may not be abundant in the candidate pool. This gets progressively more complicated with each tiered vector you add. For this reason you should minimize the use of multi-dimensional criteria as much as possible.
Based on your tester criteria you’ll want to discern which one of these methods makes the most sense for your recruitment strategy. Putting the time into planning your beta tester team breakdown will go a long way toward making sure you get the very best beta testers for your project.
Check out our free Beta Tester Recruitment Kit for a much more detailed look at the recruitment and selection process, as well as ready-to-use templates (including a worksheet on personas and vectors) to keep you on track while selecting the best possible testers for your product.