Most small businesses are setup based on an idea. A vision, a belief or a thought. Of course, a bit of research always takes place to ensure there is at least some viability in the idea, but the very essence of a small business begins based on an individual’s idea.
As the business is setup and continually grows, although ideas still play a large part in a business’s development, to ensure its success, more structured planning and implementation needs to take place. Ensuring that a decision isn’t made on a whim and that any changes are going to be as beneficial to the organisation as they can possibly be, it can seem like a long-winded way to make a small change, but if the change isn’t required or wanted by the audience, implementing it could have major negative consequences.
Although there’s a variety of different testing methods to utilise when considering a development point, there are two that most business owners are aware of – AB testing and multivariate testing.
However, while most business owners will be aware of them, not many truly understand their workings and there’s more than a large handful who believe they are one and the same. Without doubt having their similarities and both based on the same structure, their differences are vast and understanding them means you can follow the right testing procedure depending upon your individual requirements.
Arguably the testing method that most people are aware of, AB testing – also known as split testing – is the process of dividing your audience into two and showing them different products, services, webpages, etc. For example, you may have two designs for a new website and not certain which one will work best, you launch them both and drive 50% of your traffic to the first new website and the other 50% to the second.
By doing this – and monitoring the activity on the websites – you should be able to get a good understanding of which design is going to be the most beneficial to your small business. For instance, you may have increased clicks, a lower bounce rate and a greater length of stay on one website design than the other, leading you to realise that that design is going to be the most successful when implemented fully.
But what do you do when the results are varied and one website design has some great positive aspects, but the other does, too?
Well, that’s where multivariate testing comes into play.
With AB testing, you’re testing a whole product, such as a complete website – with multivariate testing, you’re testing specifics.
Let’s imagine you carried out AB testing on two different website designs and while one was the clear ‘winner’, there were aspects from the other design that proved to be successful. With multivariate testing, you would essentially create a few different versions of the favourite website design, but tweak specific aspects. For example, it could be that the ‘Contact Us’ box in the sidebar of the second design proved to have a great click-through rate and so this part is tested with three different colours or words. It’s all about optimising and refining an almost finished product, rather than completely reworking an entire item.
It’s vastly important that as a small business, with money so often very tight, any changes you make are always going to be as beneficial to your organisation as they can be – and using the right testing methods will go a long way to ensuring this is exactly what happens.