Tag Archives: user testing

interface testing UI design tubik

Make the Right Choice. A/B Testing for UX Improvement.

To be or not to be? That is a question and a constantly repeating situation. People often have to face some choices which are hard to make. Designers’ job is also full of the situations when two solutions seem like right but you need to make a choice and not to stumble. When it happens, a creative team runs tests for both to find out which solution is the best. One of the popular methods is called A/B testing. The article describes the facets of A/B testing methodology and shows how designers apply it for UX improvement.

 

What is A/B testing?

 

A/B testing, also known as split testing, is a method of comparing two versions of a digital product in order to find out which one performs better. A creative team divides users into two groups and each of them is shown different variants. One half sees A version, the other — B. Such an approach helps to determine a more profitable solution.

 

The differences between A and B options can vary from the smallest to the big one. For example, it can be a color of a small element or different placement of all the components. The effectiveness can be measured by different criteria such as page views, clicks, number of subscriptions, or sale-leads. The choice of criteria for measurement depends on the goals a company or a creative team has established. In some cases, it may be good to combine these metrics to achieve deeper insights of all aspects.

 

How can designers use A/B testing?

 

UI and UX design for a product is always created to gain certain goals. Designers thoroughly choose all UI elements as well as plan clear and pleasant interaction and navigation system for effective user experience. However, what once worked well, may not have the same effect after a while. Business goals constantly change during the time a product functions, so the design requires some improvements.

 

A/B testing is an approach helping to make design changes carefully so that users wouldn’t feel inconvenient. Designers can collect the needed data and metrics while people continue to use an app or website.

 

So, what do designers get from A/B testing? First of all, it’s a good way to learn behavior patterns and peculiarities of the target audience. By testing different versions of UI, designers can see what kind of influence each solution has on users and which one performs more effectively. Moreover, A/B testing assists to remove bad design decisions if there were ones and replace them with something more profitable.

 

In addition, A/B testing can reduce conflicts between designers and the clients. It often happens that the opinions of both sides differ and no one wants to step aside. To solve such a kind of conflict, designers can run A/B testing for two variants which a creative team and a customer see as the best solution.

 

As we said above, differences between versions A and B can be either small or big. Here is a list of elements which designers often choose for A/B testing.

 

  • CTA buttons (their placement, size, color or copy);
  • Headings and subheaders;
  • Images on landing pages;
  • Presentation of clients’ form on websites;
  • Copy (length, placement and content);
  • Offer display;
  • Videos (presence or absence).

 

A/B testing is not a one-time operation. Designers can test different elements one by one and then combine gathered information to check which of their solutions is more user-friendly and goal-oriented.

 

usability testing for web and mobile

 

A/B testing process

 

If you never tried A/B testing before, it’s never too late to experiment. Besides, experimenting with new approaches can help to open new opportunities. Moreover, A/B testing is not a complicated method. It can be accomplished easily if you take the following steps.

 

Step 1. Gather data and rates.

The first aim of A/B testing is optimization. It can be revenue optimization, user experience improvements or just a product upgrade as a whole. That is why the first step before you run A/B testing should be data collecting. Analytics results can help to understand what aspect needs the optimization the most. Pages with low conversion rates or CTA buttons with a small amount of clicks can be the first targets.

 

Step 2. Set the goals.

To make the improvements work right, designers need to determine what they expect to gain from them. For example, it can be more subscribers for a blog, more time spent on a website, or more clicks on CTA buttons. Setting the goals you can track the effectiveness of your decisions.

 

Step 3. Build a hypothesis.

This step is needed so that designers could use the gathered information for the future improvements. After the goals are set, you need to think of the reasons why the new solutions will work better. A research on the target audience helps to generate some hypotheses about what can improve the current version. It will guide designers when the testing ends and will give deep insights on the target audience preferences.

 

Step 4. Create A and B versions.

When designers have decided what changes they want to implement, it’s time to bring them to life. There are many free and paid tools helping to divide the users’ stream into two groups. It may be a good idea to QA the testing at the beginning to make sure everything goes as it should be.

 

Step 5. Go testing.

It is the most pleasant part of A/B testing for a creative team because it’s time for users to work. People use an application or visit a website and everything they do is measured and transformed into useful data.

 

Step 6. Results analysis.

The experiment lasts for certain time and when it’s complete designers work on results. All the data and metrics from two versions are gathered and compared. Based on the results designers decide which variant performed better and is capable of accomplishing the goals that have been set at the start.

 

gamification in UX design

 

Benefits and limitations of A/B testing

 

There is no perfect methodology which would suit any situation. However, each approach is unique and brings certain advantages which others can’t. Let’s see what A/B testing can offer to designers and what are its limitations.

 

Benefits

 

It measures the actual user behavior. People use a product without knowing they are a part of the experiment now. That’s why they act naturally which helps to identify the model of their behavior quite accurately.

 

It is a cheap method. A/B testing doesn’t require applying expensive tools and hiring people who’d like to test a product. All you need is two versions of the experiment and a software which would randomly divide users into groups.

 

Quality content. A/B testing is a good way to find out what content is valuable and works best for a product.

 

User-centered design. A/B testing is absolutely oriented on users’ needs and their behavior patterns. After the experiment designers get the data helping to make a product even more user-friendly.

 

Fast and easy analysis. A/B testing is usually focused on a few certain metrics. So when the results are delivered, it’s not that hard to determine which version performed better.

 

Everything can be tested. Not all the approaches allow testing really small things such as buttons color, fonts or headings placement. However, even such little elements can have a great impact on users’ behavior and their perception of the design.

 

Reduced risks. Some mistakes may cost a lot if they are made at the important stages. A/B testing can reduce the risks of losing money because of the wrong design decisions. By testing several versions of design at the early stages, designers make sure UI elements work well.

 

Limitations

 

A/B testing works only for fully implemented design. You can’t test just CTA buttons in UI without copy content or any other component because the results won’t be the same. Everything should be as a fully accomplished product ready for the usage.

 

It doesn’t tell you why. A/B testing is an effective method of gathering certain data and metrics showing what works well or not. But if you are looking for the answers why it is so, you need to do user research which will go deeper into the peculiarities of the target audience.

 

It can’t tell you what is wrong. Continuing the point above, we need to add that A/B testing isn’t meant to identify the problems in design. It only can show which improvements work effectively.

 

If design goes into real usage without proper testing, there is a high risk of its low effectiveness. That’s why designers are recommended to make testing an essential part of UX process. The choice of approach is a matter of circumstances and designers’ preferences. Stay tuned!

 

Recommended reading

 

A/B Testing, Usability Engineering, Radical Innovation: What Pays Best?

Putting A/B Testing in Its Place

How to A/B test landing pages

Tests Go First. Usability Testing in Design


Welcome to see the designs by Tubik Studio on Dribbble and Behance

usability testing for web and mobile

Tests Go First. Usability Testing in Design.

A tech guru, Steve Jobs, once said: “Design is not just what it looks like and feels like. Design is how it works.” Unfortunately, some people still think that designers’ job ends with aesthetic aspects of a product. Nevertheless, UX design depends on several key factors including usability, utility, desirability, attractiveness. Each facet plays its role for a sufficient product, so a creative team needs to make sure all them are done right by conducting tests. Today’s article is devoted to the peculiarities of usability testing and various methods helping to discern usability problems.

 

Why does usability matter?

 

Before we go to the essence of usability testing, let’s find out what usability means and why it is called a core part of effective UX. Nielsen Norman Group defines usability as a quality attribute that assesses how easy user interfaces are to use. The level of usability depends on five aspects: learnability, satisfaction, efficiency, memorability, and errors. Learnability stands for how easy users can attain tasks during the first usage and satisfaction is how pleasant the usage process is. Efficiency is determined by users’ ability to accomplish a task quickly while memorability depends on time which people need to re-establish their proficiency after a period of not using a product. And finally, designers track the number of errors users make and how easily they can learn from these mistakes.

 

When all these factors are considered, digital products provide a high level of usability which means people can easily and effectively use them. But why expert designers are so concerned about a single aspect of UX? Isn’t it more important what services and solutions apps and websites provide? Obviously, the features make a product valuable for people. However, if users can’t figure out how to use an application, it brings zero profit to them. Even the most beautiful interface can’t guarantee a success without a clean functionality system. That’s why one of the core stages of UX building is usability testing.

 

information architecture for designers tubik

 

What’s usability testing?

 

Usability testing is a technique helping to evaluate a product by testing it on potential users. This way designers are able to see if a website or an app is usable enough as well as distinguish possible problems in UX. Usability testing is usually conducted at the UX building stage before a project goes to the development team. It allows changing inefficient solutions easier rather than in a complete product.

 

Moreover, usability testing is a good way to dive deeper into users needs and preferences by watching their reactions while they use a product. Designers can collect the essential information needed for the creation of the effective user-centered design. Also, the gathered information can be used not only for UX design but at the other stages of product development.

 

Types of usability testing methods

 

To reach the high effectiveness of usability testing and gain the expected profit, designers apply various methods which can be categorized into three main groups: explorative, assessment, and comparative.

 

Explorative methods are used at the earliest stages of UX. Designers present wireframes of a product to users and watch their reactions. Explorative methods aim at discovering how people comprehend a concept of a product and how their mental processes work while they use a certain product.

 

Assessment methods are used at the later stages when a user can test a usability of goods via prototypes. Such methods help to evaluate the effectiveness of an app or a website as well as determine the level of users’ satisfaction with a product. Real-time experiments assist to track users’ reactions and errors they make. The collected information is used for eliminating usability problems which have been determined during tests.

 

Finally, comparative usability testing methods are involved when designers need to choose between several solutions. Users compare two or more variants of UX elements then share the opinion with a creative team. Experts consider all the pros and cons of each variant and select the most user-friendly one.

 

There is a great number of usability testing methods. All of them vary and can be used under different circumstances. The choice often depends on goals which a team sets and the budget of a project. We’ve distinguished some common methods helping to test usability. Let’s see what they are.

 

home budget app case study

Home Budget App

 

Hallway Testing

 

This method foresees testing a website or an app on random people, meaning designers can ask someone passing in a hallway or elsewhere. Hallway testing may seem a bit difficult because it requires finding people who would love to take part in such an experiment and then would be able to give a constitutive feedback. The method is often used when the biggest part of work is done so that users could feel it like a complete product. Hallway Testing is one of the most common methods since it’s cheaper and faster than many others.

 

Remote Usability Testing

 

The name of the method speaks for itself. Such approach allows involving people from different countries so that a creative team could see how their product is perceived in various cultures. There are two types of remote usability testing: synchronous or asynchronous.

 

Synchronous remote testing is handled via video conferencing as well as various sharing tools such as WebEx or GoToMeeting. The synchronic approach allows designers to watch how different people use a product in a real-time. Also, users can receive some support from a creator if needed. Still, synchronous remote testing has its drawbacks. Users should be asked to gather in a certain time and if they live in different time zone it may be problematic. That’s why a day and time should be arranged upfront and it may take some time to find the best option for everyone.

 

Asynchronous remote testing is a bit easier to handle. Designers usually set particular tasks which users need to accomplish and then all the information including clicks streams and users’ errors that occur while interacting with a product are gathered automatically. Also, designers may ask users to give a small feedback about their experience. This way a creative team can track how people from different cultures react to the product using it in their own environment (e.g. at home).

 

usability testing

 

Expert Reviews

 

This common method involves evaluation from the experts with experience in usability testing. The approach gives the opportunity to receive wide and constituent feedback and spend less time for it. Experts usually apply testing tools but also they can test it like users. Also, there is another approach called automated expert reviews. It is conducted programs which define usability problems by the use of common patterns. Automated expert reviews are one of the fastest methods still it cannot test all the peculiarities connected with human factors.

 

Paper Prototype Testing

 

One of the essential stages of product creation is wireframing. It is a simplified and schematic visual representation of a layout for website pages or screens of an application interface.They are usually black and white illustrations, sometimes with bright marks or spots to outline specific areas or points, that give a clear vision of the project structure and connections between different parts.

 

The wireframes can be presented to users where they can perform certain tasks like in a real product by interacting with paper models. The method may seem unsophisticated still it’s a cheap and fast way to find out usability problems at the early stages.

 

Thinking Aloud

 

Thinking aloud is a usability testing method which aims at collecting feedback from users right in the usage process. Designers ask users to verbalize their thoughts and feelings which they have while using an app or website. A creative team can comprehend the attitude patterns of the potential users and apply the gathered info for future UX improvements.

 

design for users copy in UI

Design4Users Website

 

Is usability testing profitable?

 

We’ve described only several methods of usability testing but there are much more of them. Methods can be changed or combined depending on the goals. If you still doubt whether you need to conduct usability testing, let’s look at the list of tasks which it helps to accomplish and benefits it brings.

  • Collecting the info helping to improve UX;
  • Identifying problems and bugs which a creative team can’t notice;
  • Ensuring user satisfaction with a product;
  • Increased efficiency of products;
  • Reduced cost of development process;
  • Fast problem elimination;
  • A product with a high level of usability will definitely stand out of crowd.

They say better twice measured than once wrong. Just remember that it’s easier to fix problems at the early stages rather than rebuild a complete product.

human memory tips on ux design

 

Recommended reading

 

Usability Testing

 

Precious Errors: Testing iOS Mobile Applications.

 

Usability 101: Introduction to Usability 

 

The Brief Guide to Testing Mobile Interfaces


Welcome to see the designs by Tubik Studio on Dribbble and Behance

mobile application testing

The Brief Guide to Testing Mobile Interfaces

Mobile technology and smart devices have already brought massive and significant changes in human behavior, communication, way of thinking and lifestyle. In fact, the boom of apps popularity can be seen in plenty of industries, among which we can easily sort out some noticeable ones such as Media (magazines and newspapers), Travel (maps), Medicine (patient records), Finance (apps for real-time trading), Education (translators and dictionaries) and Social (games and social media platforms). With steady and obvious progress of the sphere, mobile and web apps are definitely going to become more critical for personal, professional and business success and efficiency. So, creating and launching a new application, all sides of its production should test the solutions and environment as well as practical quality of design and development outcome. Extensive practice we have got here in Tubik Studio proves that this is the well-lit way to providing users with positive experience and helpful products.

 

Testing is vital for product quality

 

In our previous article, devoted to the full cycle of creating a mobile application, we have already mentioned the great part of testing in providing high quality products, able to solve problems and improve lives. As it was outlined in general terms, testing is one of the crucial phases of the entire app design and development process: it helps to reveal bugs before the app is brought out to actual users. Practice shows that in the majority of cases, users abandon the app if it has functionality problems, no matter how promising and engaging it seems. Vice versa, even the simplest apps can be successful for commerce, business, advertising and other aims if they work correctly and efficiently, according to target audience’s expectations and solving their problems.

 

Testing doesn’t mean that designers or developers do not provide the upper quality. To set the analogy, the fact that every book, magazine or newspaper issue goes through editor’s eye doesn’t say that journalists or writers are not talented and qualified. Different specialists have different goals and skills in the process to increase general productivity and efficiency, and it works the same way for design and development of mobile apps. Supposing that developers have done a great job and made no mistakes, the task of testing is not only finding errors. Quite the opposite, it helps to understand the quality of the app and find the way to improve via real interactions.

 

Peter Drucker once said “Quality in a service or product is not what you put into it. It is what the client or customer gets out of it.”  In the majority of cases, if users see that anything in the app is broken or it constantly takes large loading time preventing them from quick accomplishment of the task at hand, the risk is really high that they will bounce it and never come back. That’s why testing phase is the hidden hero of mobile app development.

 

Today we are going to concentrate more on perspective of testing mobile apps, as websites testing has its own specific features worth another article. So, let’s review the basic steps and aspects important for effective digital product testing.

 

mobile app testing

 

Start with the type of the app

 

Mobile app testing has plenty of challenges based on multiple factors: there are some compromise decisions testers or developers need to consider and choices they need to make on combinations of different techniques and methods to be used. One of the aspects to think over is the nature of the app itself directly influencing the process and limits of testing. In this perspective, the following basic categories of the apps should be mentioned.

 

Native apps. The apps of this sort are created specifically for one platform, aka operating system of device. They can take full advantage of all the device features such as the camera, the list of contacts, the flashlight, the accelerometer, the compass or others, and are installed via an app store (Google Play, Apple’s App Store, Microsoft store, etc.).

 

Mobile web apps. They are mobile-optimized websites that look and feel like native apps, still they are run by a web browser.

 

Hybrid apps. This type representatives provide the mix of native apps and web apps. Like native apps, they are installed through app stores and can take advantage of the numerous features available on device. As well as web apps, they rely on HTML being rendered in a web browser.

 

Toonie Alarm UI design

 

The type of the app under construction directly influences all creative stages from UX wireframing to final check of all the code. Different operational systems set their own guidelines and recommendations on design solutions and performance, development tools and restrictions. Neglecting those limitations and rules, creators could fail submitting the app to open use and turn all the creative process into wasted time and effort.

 

Plan the testing levels

 

Starting actual work over the app, creative and production team needs to consider a testing plan combining different testing levels. Testing is not the process taken as one single stage or action: it is spread via different stages. Schematically, this process can develop in vertical and horizontal directions. Vertical dimension is presented with testing levels and usually deals with going deeper to the code and functionality from separate units to whole app, its connection with the server if needed, and its technical correspondence to the requirements. Horizontal dimension deals with different aspects of app performance and user experience.

 

Taking the testing levels to account, their basis is variety of tests grouped according to the stages when they are used in the development process, or by the level of specificity of the test. The main task of testing levels is to identify missing areas and prevent overlap and repetition between the development phases. Let’s briefly describe the various levels of testing.

 

Unit testing. This testing is commonly done by the developers to make sure that their code is working correctly: they test the sections of code like classes, functions, interfaces, etc.

 

Integration testing. Integration testing is done when particular units are integrated with each other, with the aim to test the behavior and functionality of each unit after integration. This testing level is usually carried out by a software testing engineer. There are a few types of integration testing such as big bang integration testing, top down and bottom up.

 

System testing. Here the name says everything: at this level all the components of the app are tested as a system in order to ensure that the product at the pre-launch stage meets the specified requirements. As the app is almost ready, it can be tested in the environments really close to the actual in which users will get their experience once it is deployed.

 

Acceptance testing. The main objective of this testing level is to evaluate whether the app system complies with the user requirements and whether everything is ready for release.

 

ios-developers-tubik-studio

 

Define the types of testing

 

All aspects of the app usage should be tested. Designers, developers and testers will need to check it across different devices as the points like screen resolution, processors, battery life and memory are different and can significantly affect on the efficiency and performance of the app, functionality, handling and loading time, as well as UX and UI solutions.

 

Among the variety of types, applied to test the mobile app at different stages of its creation, we should definitely mention the following list.

 

Usability testing. It is carried out from the early stages of app creation to verify if the app fulfills the established objectives and tasks getting a good response from users. The primary focus of this testing is on easy and quick use of an app, simple on-boarding and user’s satisfaction with the entire experience. For higher efficiency and productivity of general creative flow, this type of testing should start much earlier than any single line of code will be written, from the first schemes and transitions put into UX wireframes.

 

Installation testing. At the initial stage on installation, the app should add required software to the device automatically. And uninstalling, it should remove all the available content and databases from the device which are used by the app. So, this type of testing checks that the install/uninstall flow goes properly.

 

Functional testing. It is the most basic test for any app to ensure that it is working according to the defined requirements and there are no functions missed in the process of interaction. Functional testing mainly includes finding possible specific bugs of the device or navigation issues of the app. This type of testing should be done at the primary stages of development. In enables developers or testers to check and measure database or network queries with response times, crashes and memory issues.

 

Performance testing. Rather stressful part of any app testing is performance test revealing the omissions which left unnoticed during functional and user interface testing. This testing is required to be done on actual device only so it means the whole app is coded t this stage. This type covers client-side, server and network performance. For example, it checks the performance specifications and behavior of the app under certain conditions such as low battery, bad network coverage, low available memory and etc.

 

Interruption testing. An app may face various interruptions being in active mode, such as incoming calls or network coverage outage and recovery. This kind of testing shows what the app will do in this variety of cases. The common types of interruptions are:

 

  • Incoming and Outgoing calls, SMS or MMS and different notifications
  • Low memory warning
  • Cable insertion or removal
  • Network outage or recovery
  • Media Player on/off
  • The device power cycle, such as low battery notification.

 

Memory testing. This type checks that each app maintains optimized memory usage throughout surfing. As mobile devices have definite limits of available memory, this testing is crucial for the appropriate product functioning.

 

Security testing. It checks vulnerability of the app to hacking, authentication and authorization policies, data security, session management and other security standards. The fundamental objective of security testing is to ensure that the app data and networking security requirements are met according to standards.

 

Tubik Studio iOS development

 

Don’t forget to test design

 

One of the huge mistakes that could be done about testing it planning this part not earlier than the development stage starts. Obviously, the product getting live via code presents wider perspective of testing its actual functionality and performance. Still, we should remember that any digital product is not just code — it is the set on user interactions which should be thought-over and designed thoroughly with target audience in mind. Therefore, testing should take its first steps at the stage of wireframing to check if all the elements take their places, communicate to users, provide them with feedback from the system and what is more, achieve the goals and solve users’ pains.

 

Numerous prototyping tools enable to simplify and speed up the process of testing user interface so that developers obtained the verified version of design not taking major design alterations in the process of development. This is the efficient way to optimize the general creative flow and provide maximum efficient outcome at every stage. Prototyping gets testing sides, be it client, designer or even potential users, closer to real interaction with the concept of the future interface. Prototypes should not be seen as the analogue of the final product as they aren’t those. Their main aim is to check the correctness and appropriateness of the design solutions way before they will be transferred to developers.

 

The most effective point to involve prototyping for testing design is the step between UX wireframing and UI design.  The prototypes on UI stage can also be created for presentation of application general looks rather than for testing and improving its functional features. Still, it’s crucial that usability should be thoroughly checked first of all at the UX stage, otherwise it will be much harder to change inefficient solution after having accomplished a lot of work on UI and then development.

 

All the numerous aspects of design testing are definitely worth a separate article which will continue this topic in our blog very soon.

 

InVision for UI prototyping tubikstudio

 

Check the code thoroughly

 

When the design is transformed into code, the developers and testers need to make sure that all the quality and performance nuances are considered and included. Pointers in this area are testing on actual devices and simulators. Testing on devices is proceeded on the actual handset where app is installed, run and tested. Testing in simulator applies the software that can accurately imitate a mobile phone.

 

Simulator testing is useful at the primary stages of development as it allows quick and efficient checking the app, but it doesn’t fully measures performance and usability which should have their healthy doze of actual human testing. Still, automated testing got really popular now because it is effective, cheap and quite reliable. iOS Simulator as well as testing tools such as Appium, Frank, Calabash and others are available to support moving the app through the testing process and point out the issues requiring attention.

 

Continuous testing at all stages lays the strong foundation for keeping small bugs from becoming major issues later on. The final testing is conducted on actual devices so that crashes and hangings could be identified. Testing on device is obligatory for every app as it provides developers with actual data on the app behavior in different environments.

 

Test and measure after launch

 

Another mistake that should be avoided is stopping testing after the product is launched. Contrarily, testing will get even more diverse and comprehensive as it will inform the sides maintaining the app viability and efficiency about behavior, problems and preferences of real users. This information is the direct route to product improvements providing users with the updates they really need and want.

 

testing digital product

 

Mobile technology is transforming the way of using devices. Smartphones and tablets of all kinds are rapidly becoming the valid method of interaction between consumers and businesses. People use mobile apps to connect socially, find information, order and track goods, book places and set appointments and do hundreds of other operations improving their lives regularly. It is important to build an app with all features and functionality that are required. If the app is created without effective testing plan, users are likely to come across unexpected bugs and errors. In the modern world with tough and growing competition in this field, the risk is high that they will quickly lose interest to the app while thoughtful testing and analysis is the solid way to avoid this risk.

 

Recommended reading

 

Diverse issues of applying testing at different stages of creating digital products have been an object of professionals’ attention lately. Here is the collection of recommended sources on further reading for those who would like to read more on the theme.

 

Testing Strategies and Tactics for Mobile Applications

 

Mobile: Native Apps, Web Apps, and Hybrid Apps

 

Beginner’s Guide to Mobile Application Testing

 

Mobile Testing: Complete Guide to Test your Mobile Apps

 

17 Strategies for End to End Mobile Testing on both iOS and Android

 

Levels Of Testing

 

Software Testing Levels

 


Welcome to read The Ultimate Guide to Creating a Mobile Application