#ab-testing

[reneeroberta]

Merkwaarden met EEG- en MRI-onderzoek meten

Merkwaarden met EEG- en MRI-onderzoek meten. #future #marketing #communicatie #neurologie

Het is inmiddels alweer 2017 en vandaag de dag is heel veel mogelijk. We hoeven bijvoorbeeld geen saaie enquête meer te versturen om merkwaarden te meten. Je kunt ook gewoon Big Data, AB-testing, een EEG- of een MRI-onderzoek inzetten. Wil je hier beknopt meer over weten? Lees dan verder.

(more…)

View On WordPress

[1lvblog]

Advanced Landing Page Optimization With Conversion Centered Design (Oli Gardner)

http://ift.tt/1Duph64

[1lvblog]

A Roadmap To Becoming An A/B Testing Expert

A/B testing, also known as split testing, is the method of pitting two versions of a landing page against each other in a battle of conversion. You test to see which version does a better job of leading visitors to one of your goals, like signing up or subscribing to a newsletter. http://ift.tt/1y3hhIA

[sardwon]

[popdotco]

Why we rolled our own AB testing tool at POP

With all of the options for AB testing out there, you may ask why on earth would we roll our own a/b testing tool? Some of our core reasons are as follows:

Crowd sourcing software benefits everyone.

By building this library and making it publicly available, we've opened the doors to accept bug reports, bug fixes, feature requests, enhancements, and more. By promoting an open source environment, we hope to help future startups get up and running more quickly with battle tested solutions. This is a community effort; we just happen to be championing it.

We like to keep things in house.

If we run into problems, it's much easier to diagnose and fix ourselves than to rely on third party support. This is definitely a trade off as we increase our technical debt in favor of reduced reliance on others. We believe that this approach leads to better stability, security, and uptime as our libraries mature.  The biggest gain in our opinions is that we can debug our own code. There's nothing worse than finding a bug in someone else's software and not being able to patch it.

We control the data.

There's much to be said about controlling your own data. It's easy to manipulate and we can mash it up with the plethora of data we collect on our site to better improve our service to our customers. There's no need to utilize API clients to hit third parties and request datasets; we can just hit our own datastore.

We can build tools around our library.

Let's say we want marketing or a growth hacker to be able to create their own tests and variations. With our own custom tool and the fact we are active practitioners of API first development, we can add that flexibility fairly quickly. For instance, we're looking at building a dashboard to manage Winston tests and variations as well as view overall test performance. 

We like to learn new things.

We built Winston in our free time. You know, that little amount of time you have between trying to get hockey stick growth and building out the next big feature. Having previously used a number of popular third party testing tools, we just weren't satisfied. Some of them had poor load times and blocked our page, others were too busy trying to ruin our angular.js heavy UI with custom generated HTML pages. The end result for us was to do a bit of research and create a tool that was both minimalist and practical to use. This library isn't rocket science or as full featured as alternatives, but it fits a very targeted niche for us: run test variations and track performance over time. As a startup that is rapidly iterating and testing new ideas, minimalist has been our mantra.

How do you know Winston is safe/accurate?

As with all implementations of A/B testing, the number one rule is not to trust your test results at first glance. We here at POP treat our tests like a series of hypothesis and experiments. You need to take results with a grain of salt as any number of factors could affect your outcome. For instance, a surge in traffic from a different locality may affect your results. It's these types of factors that we'd eventually like to build into Winston.

If you're interested in Winston, the source code is openly available on github.  We'd love additional contributors!

Written by Corey Ballou

[sardwon]

A comprehensive list of useful tools for UX designers

[clutchio-blog]

Announcing Native A/B Testing for iOS and Android

It's hard to believe that it's already been a month since we launched the Clutch Framework, but we've been absolutely floored by the feedback we've gotten so far.  Rest assured we're working hard to continue to improve the Clutch Framework, port it to other platforms, and make it even easier to use.  The thing that surprised us the most after launch, however, was how many of you have asked us for A/B testing capabilities.

That's why today we're announcing the private beta of our second product offering: A/B testing for iOS and Android.  You don't need to use the Clutch Framework to take advantage of this new product—it's 100% native and works with all native apps.  If you do use our framework, it's just an added bonus that makes even more rapid testing possible.

Please check out the new APIs, apply to be in the beta program, and let us know what you think!  (We'd love to open it up to all, but alas, this service needs to scale up more cautiously.)