Saving that turkey...to (beta) test, or not to test? ?? ?? Answers inside.
Happy Thanksgiving folks! ???Before you dig in: here are some tips that can save you months of wasted effort:?
"At any given point in time, there isn't just one version of Facebook running,?there are probably 10,000. Any engineer at the company can basically decide that they want to test something. There are some rules on sensitive things, but in general, an engineer can test something, and they can launch a version of Facebook not to the whole community, but maybe to 10,000 people or 50,000 people—whatever is necessary to get a good test of an experience."?
If one's to believe what Mark Zuckerberg once famously said in a?podcast?- Meta is a master in?Beta...testing?(and I never waste an opportunity to crack a dad joke.)
Now, considering its recent performance (and the?infamous?layoff of 11,000 employees) you may be wondering if this is even something to look up to (and no, I'm not even suggesting you should try to reach such a crazy number of versions in your product).?
Well, the jury is out on?beta tests. I know people who decided to follow their gut and won big time (following Henry Ford's advice: “If I had asked people what they wanted, they would have said faster horses.”)
On the flipside, I know people who did that, and failed miserably.?
A friend of mine was among them.
His team made the decision to work on a new 'killer feature' for 10 months, only to find out that...it's not something that fits the profile of the bulk of their existing users.?
As a result,?adoption of this feature hovered under 5%.?
So when should you do a Beta Test???
Whenever you want to...
So essentially, every time you make a change that affects your users and don't want to rely on your gut.?
?
How to set up?a good Beta Test?
领英推荐
Well, needless to say you can't just decide to run a test and hope for the best. Here are the steps you need to take to ensure your beta test will be successful:?
You can easily create such slideouts in?Userpilot's trial for free.?
4. The key to making it successful is by narrowing your audience down in the audience targeting:?
5. Think which type of a beta test you should run (more on this below).?
6. Set up validation criteria and analytics.?
7. Analyze your results accordingly.?
8. Collect qualitative feedback after the test.?
Types of Beta Tests?
There are many types of?beta tests?you can run:
Now go read that blog and think what?you?should beta test next - your product adoption rate will thank you???
Wanna share your gory story about the consequences of launching something without doing proper user research or beta testing??
I'm all ears!?
+ our free?Onboarding Audit?offer is still on the table - grab it while it lasts!
See you next week!?
Product Leader ? SaaS | B2C I B2B | Data Enthusiast
2 年Great points about Beta testing, however I think the story about spending 10 months on a feature which ultimately didn't fit the customer profile isn't much of a problem of Beta testing, rather it's due to lack of defining the ideal customer profile for the product. Beta testing would help test the solution, after the solution or some part of the solution has been built usually by collecting feedback from those customers who are not fussed about bugs. However, building for the wrong customer is not a problem Beta testing would solve, as by that time it would be wasted engineering effort and the feedback after building the product would only be from the users, "we didn't have this problem in the first place"