I frequently argue with Matt (the other UsabilityHub co-founder). More often than not it is for recreation, but occasionally we also argue about things that matter, like business decisions. Neither of us are business minded, neither of us have an MBA and this is the first time either of us have run a business. Still, I figure that makes us at least as qualified to have an opinion as anyone else. Our current topic of argument is the merits of free plans. With UsabilityHub we have a very big "free" user base. For every 15 sign ups we get, only 1 or 2 will be for a paid account. On any given day, around two-thirds of our tests are created by "free" users. Our freemium model also means that around half of our support queries are from "free" users.
The current trend in the web product world is to ditch free or hide free . I read a great article on the topic yesterday. Ruben Gamez from Bidsketch makes some great points on why free doesn't work. Some companies are dropping free altogether (like his) and report upwards of 800% increases in revenue. Others demote their free plan to a teeny weeny little link beneath their regular plans in the hope of encouraging more users to select a premium plan. If you're going to have a free plan, be proud of it. Don't hide it. That just comes off as being sneaky. If you believe free is the way to go, have the balls to stand behind it. Free plans can work. Free plans DO work. Ditching your free plan will most definitely NOT guarantee you an increase in sales. In fact, it may well be the worst thing you ever do.
For us, our biggest fans are our free users. Our paid users aren't nearly as vocal about our products as our non-paid ones. I've not once seen a Tweet saying, "hey everyone, look at this great $50 a month product", yet we frequently have blog posts, articles and tweets telling the world to check out our free product. "Free" puts us in direct contrast with every single one of our competitors. But that's not even the best part. Our free users ARE our product. Our usability testing is conducted by the same people that other companies don't want the hassle of dealing with. Our competitors charge hundreds of dollars per usability test, and thousands per year and yet you have to bring your own testers! You get given the tools, but you still have to source your own results. Madness! We leverage the goodwill of our community and in return give them free usability tests. It's a win-win-win situation, and we all know that is the best possible outcome.
It's easy to say that we're in a special position in this regard, but that's because we put ourselves here. We created Fivesecondtest and our other UsabilityHub products from the ground up with this model in mind. Our free product isn't a demo or an artificially restricted version. It operates with a completely different dynamic, with a completely different set of users in mind. We don't expect our free users to upgrade. Think about it; If your project management tool works great for free with one 1 project, and I only have one project, then I will never ever buy your product. EVER. End of story. To me, this is obvious. If you're doing freemium like this, then you're missing the point. Ok, so we know our free people will never upgrade. But rather than seeing them as a burden, we can actually leverage them. We can get value beyond dollars and cents. For example, if you're so concerned about offering support to free users, why are you doing it? Set up a forum for self-support. Let your free users help each other. You never need answer another support email again! You'll find that they're more than capable with very little input from you. Solve a problem once, don't spend every day of your life re-writing the same emails.
Free users are also a great marketing tool. You don't need 100,000 users to get value from word-of-mouth marketing. Make it part of your product. For our free users to get results more quickly, we ask that they tweet their tests, or post them to facebook, or even email them. By sharing their tests to 10 of their friends, not only do we not have to provide results for that test, but those new testers will go on to do even more tests. They may even sign up themselves and become a new customer. If they don't it doesn't matter, they've already performed a valuable service. Nearly half of our traffic comes Twitter and Facebook. That's word-of-mouth working wonders.
Just don't tack on these features as an afterthought, make it part of your product. Make it a tool for your free users. Make them want to advertise your product. Make them sing for their supper. But even if these methods don't work for you. If you've got a business that definitely cannot leverage it's user base, there are still options. The best option is one where the end-user doesn't even get to choose to upgrade or downgrade. They're simply on whatever plan their usage dictates. Chargify is a perfect example of this (we love Chargify btw). You don't pay a cent until you start doing 50 transactions a month. This low entry point has seen them pick up over 2300 signups in the half year that they've been out of beta. The best thing about it is that every one of their customers WANTS to pay them more money. If you're on the $749 a month plan, that means you have over 5000 paying subscribers. Who wouldn't want to pay $749 a month for that?
Rather than not offering a free product at all, why not make every customer a free customer and adjust your pricing so that they start paying you only when they start becoming a burden? Free is still a great way to get new customers. But you have to smart about it. There is no point offering your premium product for free and expecting people to want to pay you. Shareware doesn't work, and your product is not Shareware. Take the option away from the user. Either the free version is all they need and all they will ever need, or they absolutely must have your paid version no matter what. You can't have both. If your user can exercise choice between free and paid, you've done it wrong. Users will nearly always choose free, even if it means more hassle. If you're offering your customers these choices, you're doing it wrong:
- Can I live with 5Gb for free instead of paying for 10Gb? WRONG!
- Can I live with managing 5 clients at a time or do I need to pay for 15? WRONG!
- Can I live with only 1 person having access or do I need to pay for 3? WRONG!
If a user is making that decision, they'll more than likely make do with the lesser option. You can't stop that behaviour. The best thing you can do is just accept that your free users will always be free users, and target them appropriately. Make your free product a product designed for free use. Don't use your free plans as a means to get users to upgrade unless your product is something which the user simply has no choice but to upgrade at some point (like Chargify).
Focus on gaining value from your free users and stop trying to get them to pay; they never will. I agree when Ruben says to stop blindly copying others. That's a great rule no matter whose blog you're reading this week. This week's "7 ways to riches" is always going to be different to next week's. Rather than blindly following this week's guru, stop and have a think about what works for YOUR product. I want you to put a little bit of thought into making real use of your free users, rather than seeing them simply as an upgrade path or a burden. Whilst it's true that a free user will most likely never upgrade, there are still plenty of ways free users can actually ADD value to your business, you just need to find them.
- Do you want keyword tagging like the old test?
- Do you want to split results into Fast and Slow clicks, or who clicked where first?
- Do you want a time limit on the test?
- Do you want to restrict the number of times a user can click?
In other news:This update also includes some other important changes. The most important of which is prioritising tests based on subscription. What this means is that users on the Team, Studio and Agency plans will get their results before other users. It's worth noting that Solo plan subscribers will also be prioritised over free users, but not nearly to the degree of the other plans. This is the first of many additions to give more value to our subscription plans. Keep an eye on this page for details on some of our upcoming features.
So we've had a heap of great feedback from the community. I'm thrilled that we have so many passionate users!
We've made a few changes today in response to feedback, and we have a lot more changes coming. Nothing incredibly huge at the moment, but I just want to show that we're listening, and that we WANT your feedback, your praise and your hate mail.
Here is a short list of some of some of things we did today:
- Result sharing - You can now share your test results with others. In your test results page you'll see a new special link which you can safely share with others.
- Print friendly results - We've updated the CSS on the results page (and the new share page) to be printer friendly.
- Test sharing - We noticed that not many users are sharing their test links. This is the fastest way to get results, and also means you can target your own audience. This is especially true if you're conducting a test in your native language. To get people involved in doing this, we've added a prompt to get people sharing their tests.
- Random tests - We've added a new "random test" function in the dashboard to allow users to do ANY random test rather than just doing 5st or Navflow tests. This is heaps of fun!
- We've updated a lot of the links and redirects to stop users being taken back to the home page when they're logged in. Hopefully that alleviates some of the frustrations there!
- We fixed a heap of issues with IE7 that were preventing some users creating tests (sorry folks!)
- We've added "requested responses" to your test results page, so you can see how many responses you have remaining for your tests. e.g 3 of 20 responses.
We're still working hard on a lot of the other things that have been requested, and whilst we can't (or won't) make all the requested changes, it is important that you know that we are listening and we are actively creating solutions to problems you guys are having. I know we're not perfect, and I know we can't please everyone, but a lot of the requests we're getting are actually quite reasonable. Rest assure that we're doing everything we can do make this app what you want it to be. Please give us your feedback, we love hearing from you!
Stay tuned for more!
Well it took us a while, but we've finally launched. For the most part we actually finished these apps about 2 months ago. But we've spent the last 8-9 weeks messing around with merchant accounts and payment gateways. A lesson for all startups out there, sort your merchant account EARLY.
So we're up and running now, and everything seems to be running ok. A fairly major part of the installation was migrating users, their karma and their credit across to the new system from Fivesecondtest.com. Basically we have an entirely new system, with an entirely new way of managing credit, and so it took a little shoe-horning to get that in to place.
For those who aren't aware, all existing Karma has been migrated across to the new system. On the new site 1 Karma point is worth 1 result. You won't be able to make tests without karma, so
On the old site, when you spent $5 you got an additional 15 results. We've doubled that. Every token (or $5 credit) has been converted to 30 results on the new site. This has been given out in the form of Karma.
Unfortunately, all the Beta data for Navflow was deleted. This includes logins. About a third of our Navflow users are actually 5st users, which meant a whole heap of account duplications if we migrated everyone across. The remaining users signed up and didn't create any tests. So it made sense to just migrate 5st users across. I'm sorry to those who were having fun with Navflow, but unfortunately we had no choice.
The first MAJOR change in the new system is to Fivesecondtest.com. The click test is gone. Navflow is now the place to do click tests. We found that clicks were not an accurate way to gauge first impressions. Without context, and with a time limit, the click test was a little shaky. We think that navflow gives test authors far more flexibility to find out what they need to know without putting undue cognitive load on the end user.
Through a lot of feedback, and a lot of study, we've also changed the fivesecondtest memory to be a much more traditional and more flexible test. Rather than having unfocussed "what do you think?" type testing, we've now included the ability to ask up to 5 questions about the test image. The user is shown the image, and then you can test their recall about the image. The result is a significantly more targeted and quantifiable result. In time we will add additional tools to assist in benchmarking and iteration.
Navflow and Fivesecondtest are the first two apps in our suite of tools called Usability Hub. When you sign up to one or the other, you're actually signing up to UsabilityHub. One subscription fee will give you access to all of our apps now and in the future. Your subscription entitles you to a certain number of results per month, usable across any of the applications for as many tests as you want. Subscriptions start off from as low as $19 a month, and can be upgraded or cancelled at any time. No lock in.
As always, our apps are available for free to anyone who wants to help give feedback on other users' tests. Earn karma to create tests. One major change however, is that we no longer offer anonymous test creation. If you want to create tests, you'll need to sign up, participate in tests and earn the right to create your own tests. To make up for this, we give all new users 20 Karma to get started.
Well that's about it for now. I need a rest!
If you have any queries, shoot us an email at firstname.lastname@example.org
Thanks for all support till now, and welcome to UsabilityHub!
On July 31st we'll be launching UsabilityHub.com. Usability Hub isn't much at the moment, just a pair of web apps really. But it is the start of something bigger and better than anything we've done to date.
The goal of UsabilityHub.com is to provide a place for designers and developer to work together, share ideas and help each other improve websites and applications. It's about providing a focal point for all things UX. We want it to be about more than just our applications, we want it to be a place where all designers can learn from each other, and help each other out. In the coming months we'll talk more about what our plans are, but for now let's just talk about what it means for our current applications.
Navflow.com beta will be finishing next week. <strong>All existing data will be deleted at the end of next week (23rd July).</strong> Where possible we will migrate user accounts across to the new system enabling you to keep your login in. Part of this migration is migrating our existing Fivesecondtest users. Obviously if there is a clash with your existing 5st account, we'll keep your 5st account in preference over the Navflow beta account.
Five Second Test
We have a completely new Five Second Test which you'll get to see very soon. Whilst we think our new site is better in all regards, we understand that a lot of people really like our existing system. Rather than delete it, we're going to let you continue to use it! The old fivesecondtest.com will be renamed to original.fivesecondtest.com. So if you're currently running tests there, don't be scared that it's all going to vapourise, you can keep on keeping on. Please keep in mind that the two systems are completely separate. Tests created in the old version are not available in the new, and vice versa.
Users with credit from Fivesecondtest.com will have their credit <strong>copied </strong>across to the new site. This will enable you to continue creating tests with the credit you have purchased on the old site until your credit runs out, at which point you'll need to sign up to one of the new plans (or keep creating free tests using karma!).
As the new site is subscription based, we have to migrated your credit to match the new system. We're using a fairly generous conversion to the new system to ensure everyone gets a good deal. The old 5st had a variable return on credit which moved from a high of around 25 responses per credit down to around 10. This meant that (depending on traffic) $5 credit bought a different amount of results. I thought it was cool, but the consensus was that it was all a bit weird. To keep this fair, all old tokens ($5) will be worth 30 responses in the new system. This means no matter when you bought your credit, you're getting a good value upgrade in the new system. What's more, we're going to let you keep your credit on original.fivesecondtest.com as well! So consider this a warm introduction to our new system!
Accessing old Five Second Tests
One of the big changes we're making is a complete revamp of Fivesecondtest.com. Click tests have been removed as Navflow will now concentrate on our "click tests" portion of testing. Our old Memory test is being shelved in favour of a new form of the Five Second Test which we feel gives better results and is better value for money than our old system. This means that old tests will not be viewable in the new site. Users wishing to get data will continue to be able to get it from original.fivesecondtest.com when we launch. This will be kept available indefinitely. If you really want, you can even continue using the old site.
The future of Usability Hub
What you'll see in the next few weeks is really only the beginning. We've worked long and hard to make fivesecondtest.com more user friendly and more business friendly. We've listened to our users - and to the experts - and we've improved how the test itself works. We've integrated two great products into one set of tools. But we're not finished.
These two new products are really fairly young, and we have a lot of upcoming features post-launch. So if they're not perfect for you right now, we hope in the coming months they will be.
These two applications are just the first in a suite of tools that we call Usability Hub. We have 3 more coming before the end of the year, and each additional tool will be available at no extra cost to subscribers. We hope to make Usability Hub the one stop shop for all things usability testing and design. We've still got a long way to go, but we think today is a big step in the right direction.
Step 1. Installing Linux.You could use Cygwin, or even Windows, but I'd really suggest not. I tried this, and it's just a pain in the ass. Besides, you won't be using either in our production environment, so you best get used to Linux sooner rather than later! I chose Ubuntu, just because that's what came up in Google. If that statement doesn't make it clear, I'm a complete Linux beginner. I used Linux about 10 years ago, and haven't touched it since. So if lack of knowledge of Linux is what's been stopping you, have no fear. If I can manage it, anyone can. You have a couple of options to get up and running with Linux. Setup a new Linux box, or run a Virtual Machine. I didn't have a spare box lying around, so I chose the latter. So, first thing you'll need to do is go off and download Oracle VM VirtualBox and Ubuntu. Then mount the Ubuntu ISO as a virtual CD drive in Windows (there are plenty of apps to do this, google it). Install VirtualBox and create a New Machine, Call it Linux, choose Linux/Ubuntu as your OS Type. You can pretty safely click next repeatedly during the rest of the install. Once it has installed the VM, go to settings > storage. Click the little empty CD controller and from the CD/DVD device on the right, choose the drive which has your Linux ISO. Click ok and start up the VM. If you've done everything right to this point, Ubuntu should start installing. If it crashes because it can't find your ISO, go back and check your settings. Assuming you have a fairly standard machine, it should all install easily enough. Skip the language packs if you don't need them, they take forever to download and install. Once it has installed, you'll need to "power off" the VM, and remove the ISO so that Linux can boot. When you first get into Linux, it'll want to install a heap of updates...just let it. Ok, so If you're a Linux n00b like me, you wont even know where to start. In fact, you're probably wondering why the damned VM window is so small. To fix this, go to the Devices menu in VirtualBox, click the Install Guest Additions. This will mount another ISO in Linux. Go to Applications>Accessories>Terminal. Welcome to Linux's DOS prompt :P
sudo sh ./VBoxLinuxAdditions-x86.run
Step 2. Installing Ruby.Unfortunately, that was the easy part. The next part caused me a lot of grief. I eventually worked it out, but only after a lot of Googling, and a lot of swearing and it turns out I didn't actually need to do most of it. Hopefully I can save you from the same fate. The first thing you need to know is, whenever Googling, resist the temptation to Cut and Paste helpful command line tips and tricks. These often include references to a specific version of a package, and can screw everything up real quick. If you need to install ANYTHING, make double sure that you've got the right version. The problem here is that so many versions of Ruby, Gems and Rails have come and gone, and the installation methods, requirements and problems are all different. This can be a major headache for Linux beginners! Even going through this the second time, I'm still running into issues! So, crack open the terminal again. If you haven't already guessed you'll spend a lot of time here! First thing to note is a command called "sudo". Sudo is the Linux verison of "run as administrator". Most of the commands you run for installation will be prefixed with this. It will prompt you with your admin password and then carry on. Second thing to note is that for the most part, you won't need to download anything from any third party website when doing installs. Linux has a rather nifty package installer that gets and installs everything for you! If you're keen you can download source packages and compile them yourself, but we won't need to do any of that. So theoretically, Ubuntu should have most of what you need already, but we do need to go get a few bits and pieces. Make sure that the updater has finished before trying this, as it won't work if it is still running. The first thing you need is ruby.
sudo apt-get install ruby-full build-essential
sudo apt-get install rubygems
gem install ruby-debug
Step 3. Installing RailsOk, so now we have ruby, and ruby gems. Now we need rails.
sudo apt-get install rails
Step 4. Install your IDEGiven that you're a .NET developer, you're going to need an IDE. None of this command line editing BS. We'll be using Aptana Studio. Go here: http://www.aptana.org/studio/download Download the archive. You're most likely using Firefox, so you should be able to find the zip in the downloads folder. Extract the archive to your home directory. You can then run aptana directly from there. You can also create a shortcut in your Applications menu. In Ubuntu, go to Preferences > Main Menu > Programming. Click the New Item button. Fill in the details, navigate to Aptana executable and select it, and also click the icon to change it to the Aptana one. Click Ok, close. And check it out in your Applications menu! Great, that was easy! Well...not so fast! This is Linux! You can't install something so easily! You need to do some more command line hacking. Ubuntu doesn't come with Sun Java, which appears to be a prerequisite for Aptana. So now we need to install that. First we need to add the java repo:
sudo add-apt-repository "deb http://archive.canonical.com/ lucid partner"
sudo apt-get update
sudo apt-get install sun-java6-jre
Step 5. Watch a videoAs you won't have installed it, you'll need to install curl before you get too far in the video, and mongrel (a webserver) too. Just run:
sudo apt-get install curl sudo apt-get install mongrel
cd ~ mkdir code cd code
Step 6. What next?The good news is, that if you got through the video and everything worked, then you have everything you need to create your first web app. One good thing about Ruby is that there is a plethora of resources out there to get you moving from here. The one major pitfall is that there are HEAPS of tutorials that are hopelessly out of date, and can not only lead you astray but cause you some serious headaches if you mess up your installation! There is a lot still to learn about RoR, especially for me, but I hope this is a good starting point. My goal was to write something that got ME up and running. I figure there are plenty of Windows developers out there (like me) that wouldn't have the foggiest idea where to start, so if that is you, I hope this has helped! Cheers, Alan
As you all know we've been very busy building our new app called Navflow. What you may not know is that we've also been rebuilding Fivesecondtest.com as well! The new Fivesecondtest will sit along side Navflow as part of our new system called UsabilityHub.com. There are lot of changes as a result of this move. I can't talk about most of them just yet, but I DO need to give you a heads up about a few things. In the coming weeks we will be migrating a lot of data from the old site to the new site. All your user accounts, karma, credit and tests will be migrated across to the new system.
A few days before we migrate, we'll turn off the ability to create new tests. You'll still get results for all your existing tests, and you'll still be able to do everything else. We're just turning off the test creation. During that week we'll be migrating all the data across, and we don't want anyone's test to vaporise. Most importantly, we want everyone to have their full quota of results BEFORE we move. There's a very good reason for this...but I can't tell you just yet!
Now, one other fairly major change is coming. We're moving from pay-per-test, to a subscription service. The main reason we want to do this is because as fivesecondtest users we want you to also have access to Navflow. Buying a subscription to one, will give you full access to the other. Our monthly pricing is pretty damned good value too. The best part is that we will allow you to create many tests each month for just a little more than one Gold test costs right now. Not only that but, you'll be able to mix and match your tests, and have much greater control over the number of results you want. Create a couple of fivesecondtests with 30 responses each, and maybe then go create a Navflow test with 40 results. Whatever you need! The best thing about these plans is that unused "value" from one month will roll over to the next month! I'll go into more detail about what "value" means in the coming weeks.
Now...an important notice to people who currently have credit with us! Your credit will be migrated to one of the new plans based on your current credit. Your plan won't automatically renew however. Given that our plans allow value to roll over, month to month, you will still have full access to the full value of your credit indefinitely. You can still use it up whenever you want! Of course, should you choose to sign up to a plan, the new value will be added to your existing total!
This is all happening in the coming weeks. We'll keep you updated with news and changes! Thanks to everyone for your support, I hope you will enjoy the new Fivesecondtest.com as much as we've enjoyed building it!
I was hoping to announce the release of Beta 2 today, but alas, we can only really call this Beta 1.5! First thing's first, if you have any problems with the Navflow beta test, please contact us at email@example.com
There isn't a whole lot of "fun" stuff in this update. Mostly bug fixes, a few feature requests, and a LOT of behind the scenes work preparing for the migration of Fivesecondtest.com. We could've held off on releasing this until we'd had a lot of other "features" added, but we have some fairly important things to get out in this version. So what's here?
1. Karma - We're releasing the first iteration of what will become the new Karma system. This system is based on the same philosophy as the fivesecondtest.com karma system, but with some important distinctions. First of all, we're unlocking the amount of Karma you can earn in a day, and we're taking the cap off how many results you can get with Karma. As of right now, however, you NEED Karma to make tests. To get new users started, you will get 20 karma when you sign up. So you can go ahead and use that straight away to create a test. What this means in the future however, is that for every result you want for free, you need to help out someone else. At the moment, this is a free for all. But there are some "checks and balances" being put in place to prevent "gaming" the system. So be good for goodness sake!
2. Language support - This isn't full localisation unfortunately, but it does give you the ability to view tests that are in YOUR language. When users create a test, they let us know what language it is in, and then we only show that test to people who speak that language. This is based on your browser language, but for registered users is configurable from the settings page. Registered users will ONLY be shown tests for languages they've selected. Anonymous users, however, will be show tests in their language until there are none left, and then will start seeing international tests. Test owners can also set to only allow users who explicitly state they speak that language. Whilst this should significantly improve the quality of your results, this option could slow down the rate at which you receive results depending on what language you choose. As a result of this change, we'll be adding a lot of our international sign-ups later this week.
3. User dashboard - There have been a few comments about how the Navflow logo not taking users "home". This is actually because we hadn't finished our user dashboard page yet. It's now (mostly) finished. What this means is that if you're a registered user, you will no longer need to go to the front page of the site. You will be taken to your own dashboard showing you how much credit, karma and results you have. There is a lot more coming to this page, so stay tuned!
Ok well, that's the bulk of it. There are a lot of other minor changes, and a lot more to come. We'll be adding some more testers today, so if you are desperate to get into beta, email us at firstname.lastname@example.org and let us know! If you read all of this, you deserve to get in!