Our key principles for when and what to test
1. Start testing as early as possible
Involving your audience from the outset is critical. There’s a strong temptation, with limited budgets, to leave the testing as a validation and fine-tuning step towards the end, at which stage it’s too late to avoid significant mistakes.
2. Test ‘little and often’
There is no single point at which all the key questions can be asked, meaning a ‘little and often’ approach, as opposed to a one-off, is far more cost effective.
3. ‘Fake it’ prototype philosophy
Spending hours, not weeks, putting a couple of concept options together on paper or a couple of mocked-up screenshots can be enough to test the ideas and inform those fundamental early decisions. All you need is a realistic facade, so fake anything you can get away with, get quick feedback from your audience, then move on.
4. Strategically time and adapt it depending on the stage in the process:
- Pre-concept development: Sessions observing and interviewing the target audience are valuable prior to any work kicking off. Observing the user playing similar games or those popular with the target audience provides deep insight into their behaviours, needs and priorities which, when combined with discussion, can build a more detailed picture to inspire the creative team. If budget doesn’t allow testing at this stage, bringing in an audience expert to share best practice insights and analysing any previous, relevant audience research your company may have done also helps keep the audience perspective in focus.
- Early concept/prototype: Early testing is particularly effective to try out innovative ideas, compare options and validate general concepts. At this early stage careful session design and moderation is needed to ensure it is the concept and not the design itself under scrutiny. With the right moderator this works really well, even with children.
- Prototype testing: Once the general concept is agreed and higher fidelity prototypes are available, testing them helps to optimise the user experience, highlight areas for improvement and validate the functionality. This can often be scheduled for the end of an agile sprint and turned around quickly to inform later sprints.
- Pre-launch fine tuning: Whilst by no means the most important time to test, it is valuable to test the complete pre-launch version so long as time is scheduled to respond to the findings. If testing has been carried out ‘little and often’, at this stage feedback will largely be fine tuning the end-to-end experience, aspects of the UI etc. Feedback may also help inform marketing materials and provide input to the future roadmap.
- Post-launch product testing: Testing at this stage can be done for a range of reasons: to fine tune or inform the future roadmap; to address consumer feedback or analytics; to inform ongoing development; to benchmark for future or cross-product comparison; to support marketing activities.
For more information on our Audience Testing, and putting this into practice on your project, get in touch: firstname.lastname@example.org.
* A brief word on quantitative testing…
This article focuses on qualitative Audience Testing as it is the most common and generally the most critical method during development. However, questions certainly do arise that need a larger sample size to answer, for example, confirming the statistically most popular opinion. It is well worth understanding the wide range of quantitative techniques available and when they’re appropriate.