Usabilla Blog

Usabilla has been selected as a finalist for the Bully Awards, an award honoring the leading European technology, media, and telecommunications companies. We’ve been selected as finalist in the category of ‘White Bulls’ and are thrilled to get the opportunity to present Usabilla at the White Bull Summit event in Barcelona.

A total of sixty European companies were named as finalists for the 2010 Bully Award. The deserving finalists were selected from a pool of entries that included hundreds of nominated European TMT companies. The finalists fall into three categories:

• Yearlings: firms that seek or have received angel/seed rounds or equivalent
• Young Bulls: firms that seek or have received Series A financing
• Longhorns: post Series A firms

“The Bully Award finalists were selected first and foremost for values of excellence and innovation in the TMT sector,” stated Farley Duvall, Founder and Chairman of White Bull Summits. “Each firm has been recognized as a leader in its field, with a bright business proposition and meaningful market strategies, driven by a rich understanding of customer needs and technological solutions.”

Thirty winners (ten winners per category) will be announced live at the inaugural, invitation-only White Bull Summit 2010, Pathways to Exit event in Barcelona, September 20 – 22.

If you look at the test detail view (http://usabilla.com/tests/view/name+of+your+test) and scroll all the way down, you will notice the export options have changed. We basically used two approaches in this update. The first approach is to offer a simple straight forward export that is usable in every experience level. The second approach is to give more experienced user full access to their data in a very flexible way.

Simple exports with CSV (Comma-Seperated Values)

CSV is a simple text format for database table data and is readable by most common spreadsheet applications like Microsoft Excel, Open Office Calc and Google Docs. Simply click on the data you want to get out of Usabilla and a spreadsheet with the data is generated.

All the CSV export options follow the same format and settings. They all start with the fields “page”, “task”, “participant”, followed by the requested data. We use a unicode (UTF-8) character set, fields are separated by a comma (,) and strings are enclosed in double quotation marks (“…”).

Advanced exports with structured XML

The XML export includes all the data of a test in a structured format. We have two options, one for the content and one for the results. By combining these two exports your are able to generate any report you want.

All data is contained in a root element called usabilla. To make debugging a bit easier, we included a version attribute (current version) and a success attribute to tell you if the request was successfully executed.

Content export

The first child element indicates the type of response, in this case content and then shows you the unique id of the test you requested the content for and the language of the content. Here is an example of what it looks like: <content for="7719893004c6e" language="en">. The data follows the structure of a test, so a test has pages, and page has tasks. Here is a simplified example response:

Exporting results

In this case the first child element looks something like this <results for="7719893004c6e" language="en"> to indicate you are receiving the results of a test. The results a split up in to two sections average, which includes task completion times (calculated with geometric mean), followed by a section participants with the data per participant. Here is a simplified example of what a response might look like:

Accessing your data from other applications

Getting your data in other applications, like Google Docs for example is made pretty easy. Go to your account page and click on “Generate a new token”, you can now use any of the addresses below to get your data:

• XML content:
  http://usabilla.com/tests/export/UNIQUE_ID/content/xml/?token=API_TOKEN
• XML results:
  http://usabilla.com/tests/export/UNIQUE_ID/results/xml/?token=API_TOKEN
• CSV results points:
  http://usabilla.com/tests/export/UNIQUE_ID/results/csv/points/?token=API_TOKEN
• CSV results notes:
  http://usabilla.com/tests/export/UNIQUE_ID/results/csv/notes/?token=API_TOKEN
• CSV results context:
  http://usabilla.com/tests/export/UNIQUE_ID/results/csv/context/?token=API_TOKEN

The easiest way to find out the UNIQUE_ID of a test, is to click on “Invite participants” on the test detail page. It is the last past of the address shown under “Your test is available at the following URL:”, and looks like: http://usabilla.com/rate/UNIQUE_ID

To disable API access to your data, just click “Clear API token” on account page and access is revoked.

Enjoy!

Hope you enjoy these new featured and find them useful in your projects. As always we appreciate any and all feedback so feel free to leave comment below or tweet us any suggestion or questions at @usabilla.

When you set up your Usabilla test you can pick one of our suggested tasks or add your own tasks. Most of the tasks in Usabilla tests are predefined tasks in any of the 20 languages we’re offering. I was curious which tasks are most popular and analyzed the tasks and results in our database. I want to share a selection of my findings about how our users use tasks and how participants respond to these tasks.

Which tasks are the most popular?

When we launched our first beta we didn’t provide any suggestions for tasks. Most of our early users were completely lost without any guidelines. Based on tests with our prototypes and the input of usability pro’s we made a small list of predefined tasks. We guessed that these tasks can inspire our users to add their own research questions. The following chart shows the ‘market share’ of each of the 9 predefined tasks (out of all predefined tasks).

Which predefined task is the most popular?

Roughly 80 percent of all the Usabilla tasks are one of the predefined tasks. About 20% of all tasks are custom tasks added by users. Our users are experimenting more and more with different types of tasks. Some of the custom tasks defined by users to inspire:

• ‘Test’
• ‘Where do you click for more information about …’
• ‘Mark the things that you find confusing and tell us why.’
• ‘Where would you click to continue after reading the content?’
• ‘Mark your three least favorite elements of this page.’
• ‘Take a look at this page and give us some comments about layout and styling.’
• ‘You would like to add a comment to an article you’re reading. Where would you go?’
• ‘Where do you click if you want to know more about …’
• ‘Where do you click first?’
• ‘Click on the search box.’
• ‘Where do you click to select a date.’
• ‘Where would you click to learn about … ?’

How many points do participants add?

We offer two types of tasks in Usabilla. A standard task allows participants to answer a question or tasks with zero or more points. Participants can attach notes to these points to share additional feedback. A one-click task can be answered with only one point and participants can’t add any notes. These tasks are especially interesting to measure first impressions and task performance (incl. time). All our predefined tasks are standard tasks, where participants can add multiple points and are able to comment with notes. The chart below shows how many of these points participants add on average for each of the predefined tasks.

How many points do participants use to answer a task?

The positive predefined tasks seem to make participants more trigger happy. Participants answer the positive tasks on average with 12 points and use only 10.5 points to answer negative questions.

How many points do participants use for negative and postive tasks?

What works for you?

We would love to hear which tasks work the best for you. We’re planning to expand the list of predefined tasks in the near future and add some options to make it easier to set up different types of tests (for example measuring call-to-actions, attention, recognition, credibility, findability, collecting design feedback, information scent, etc). Feel free to share your experiences and wild ideas in the comments, on Twitter (@usabilla), or by mail (support@…).

One of our most important focus points is the usability of our own test interface. Participating in a Usabilla test must be simple, fun, and way more exciting than a standard survey. In the past year we’ve released about 15 iterations on our test interface. Each iteration was based on feedback from participants, users, and experts using Usabilla. We’ve just released a small update and we think this update solves the most important usability issues of our test interface.

Problem: Participants don’t have a clue that they can add notes.

Usabilla is primary a quantitative tool. Notes add a qualitative aspect to the tests and are of in many cases of great help to interpret test results.  If you run a test case with for example 250 participants and all those 250 participants add 20 points with notes, your ‘lean & mean’ test is probably no longer ‘lean & mean’. This is the reason why we focus on points as primary response and notes as secondary. Unfortunately in our previous releases we didn’t manage to find a good balance between these two interactions.

Read the rest of this entry »

It has been a while since we sent out an update about Usabilla. In the past months 4500 users ran hundreds of interesting tests with an incredible number of participants, we fixed many small bugs and tweaks, released new features, made big steps in scalability, and expanded our team with a very skilled developer (Say hi to Sido). In this update we want to share some interesting updates

Read the rest of this entry »

We’re looking for a skilled PHP developer to join our team. Are you an experienced and versatile developer with a soft spot for user experience? Do you dream about MVC, PHP, HTML(5), scalability, user experience, and great web apps?

Do you have….

• Experience with developing in a MVC framework (preferably Zend).
• Experience with JS libraries like jQuery, Prototype, Dojo, etc.
• The skills to develop with the latest web standards.
• The flexibility to work in an international team on a variety of complex tasks.
• At least 30 to 40 hours a week time available.
• The right mindset to enjoy working at a small, fast developing startup.

We offer you an exciting job, great customers with very interesting ideas, and very skilled colleagues. Our offices are located in the city center of Amsterdam and we share an awesome office with other interesting and ambitious Dutch startups.

Contact Paul Veugen (paul@usabilla.com) if you’re interested in this position.

Not so long ago I have been interviewed by the experienced usability expert Craig Tomlin about my background, thoughts on UX, and the future plans of Usabilla. Read the full interview on Craig’s blog: Useful Usability.

Usabilla has been nominated for the Dutch Startup Awards in the category ‘Best Web App’. The Dutch Startup Awards are organized by The Next Web and the Dutch magazine Emerce. We’re very proud to be on the shortlist for the finale. We hope to deserve this award because we kickstarted our company in the past year. In less than a year we launched an easy-to-use usability app which provides value for more than 4000 users from all over the world and are eager to continue our growth in the next years.

Update: We won the Dutch Startup Award for Best Web Application! Thanks a million for your votes.

Read the rest of this entry »

In some cases it can be useful to add a survey to your test. Using a survey allows you to collect quantitative information from your participants. This information can be an interesting addition for the analysis of your test results. For example, you might want to know the average age or the ratio of male and female participating in your test.

Embed a form or use the Redirect URL

You can embed surveys in your test with an iframe in the ‘Thank You’ page or simply redirect your users to the URL of your survey. If you redirect your users to an external form, we’ll add a variable ‘?pid=#######’ to the redirect URL. This variable is the id of your participant, which can also be found in our data exports (’subject’ in the XML, CSV, TXT files).

You can use any external survey tool you like. In this post you’ll find an example of how to embed a Wufoo form and use an external PollDaddy form in combination with your Usabilla test.

Read the rest of this entry »