> If your marketing department wants to know anything about why the user is cancelling, put it in the form. Two pages of boring questions is a great way to reduce conversion.
> Break the form into many pages so it takes longer. Include links to FAQ pages. And avoid using defaults; it maximizes the number of conscious choices for the user.
> Ask them to explain their reasons for cancelling, and require at least 100 letters of text. Explaining is hard when your reasons are emotional.
Really? Obviously I don't want to make it too easy for users to cancel, but to make it too hard just seems petty to me. Asking them to fill out the reasons for cancellation is a great idea (and worked really well for us), but forcing them to write at least 100 letters is just horrible.
Also, there are some users that you really do want to cancel. You know, the ones that suck up twice as much support as anyone else and complain incessantly.
Surely its more important to be focusing on why users are cancelling, not luring them through a maze of "two pages of boring questions"? Does anyone reputable actually do this?
There is a sweet spot somewhere in between, but this seems to be irritating and would just make me want to write complete junk responses.
"Designers, if you call yourself a “UX Designer”, you’d be the dev who called themselves a “Variable Declarer”."[0]
Anyone who thinks they are a UX Designer who wants to make the cancellation process as hard as possible obviously has a very shitty product if the only way to retain your users is to try and make them fill out forms with minimum values. Really, 100 characters? "aksfwejfjwefjwewefj" will be all you get, you evil shit.
I think there is a difference between a designer and a UX designer. I meet both kind regularly. The designer is still print-focused and can't wrap his/her head around the fact it's no longer possible to force how something should look. They think 'the look' is what design is about. Often they work from their ivory tower and their idea of communicating is 'sending the PSD'. This kind of designer (and agency) is going the way of the dodo but they still exist.
> I think there is a difference between a designer and a UX designer.
Then your vision of the design world is quite simplistic. Having studied both CompSci and Design Engineering at college, I can tell you that design can be very complex, and the term “Designer” is as abstract as “Computer Scientist”, where you have distributed systems researchers, back-end devs, Ops, database administrators, release engineers, QA Engineers, and thousands of specialities.
The design field is HUGE (go and visit IDEO), ranging from mere Graphic Designers (what you've reflected as "PSD Designer"), to UX/UI Designers (those building human-computer interfaces ), to Design Engineers and Industrial Designers that have a solid base of engineering (physics, maths, materials, CAD/CAE, ergonomy, etc) and are able to shape tangible as well as non-tangible products.
Leading Designers [0] usually belong to the latest group... Pininfarina (Ferrari's Designer), Philippe Stark (who worked closely with Steve Jobs crafting his Yatch but also does graphic design and a lot of other design specialities [1]), Jony Ive (leading the iDevices industrial design but also the UI), Dieter Rams, etc.
Yep, I feel very lucky to work with a designer who actually cares about UX, has a very good knowledge of how HTML/CSS actually works and everything that goes a long with that.
I call those people "graphic designers" to their faces to be clear they are not UX designers, but make pretty pictures. Behind their backs I call them "web decorators."
Cancelling a service should ALWAYS be easy and painless. I make sure all apps/services I have are just as easy to cancel as they were to sign up.
I actually find that about 50% of people who cancel end up coming back. I also have NEVER had a user contact me about how to cancel their account. They just do it. No problem, no interaction, no bullshit 'let me talk to you first' run around.
If you are trying to keep people from cancelling by making it difficult to do so, YOU ARE BAD AT BUSINESS.
Here's some UX for you - the user owes you nothing.
Sure have an optional form or field for cancellations but 5 minutes? How is this beneficial for anyone?
So, it's a free product - they will just never cancel their account and you will have inactive users sitting there. Who is this good for?
So it's a paid product - they will cancel their account and tell everyone about the ridiculously long winded process they had to go through to cancel it.
... or they'll just dispute the recurring credit card charges for your paid product, and you'll get your merchant account frozen after a few "cancellations".
We can argue over whether or not the original suggestion constitutes fraud, but the credit card companies will definitely treat it as fraud.
The post you had here before you edited was completely nonsensical.
A UX post, on how a site should prevent users from cancelling to benefit the website, at the expense of the user.
That is the opposite of a good UX.
Here is some 'User pseudo-psychology' for you... Having your blog titles take up more than half the vertical space of my browser (when it is full screen) stops me reading your writing.
ha yea i read the first few sections & they were so hollow i had to come check the comments too -- good UX is not putting a bunch of filler in your opening 10 paragraphs. Only thing thats been said so far is that "3-click rule" that my 900 year-old boss spews to management after raving to me about how much he loves the Mac philosophy.
XD
In all seriousness, would love to see a guide from the more scientifcally-researched vein of this type of work.
Joel, this work appears to be heavily plagiarised in places. I've only had a quick browse, but it looks like you've cribbed content from UX textbooks and online sources. Some people would consider this unethical, particularly since you have something to sell on your site (your book on 'Persuasion, manipulation & brainwashing.').
The solution is simple, you need to clearly cite and link to your sources.
As well as making the source authors happy, this will be very helpful for readers who want to do further research.
I think this actually represents a massive gap between what UI or UX designers believe and what users actually prefer.
Consider this: Microsoft Windows (prior to Windows 8) was by all accounts a clunky, poorly conceived UI...that allowed users to modify it to look just about any way they wanted.
Now we have applications like Twitter and Facebook constantly re-arranging their users layout, and the most recent changes have been to increase font size, reduce the information on the screen, all of the things that many users (such as myself) do not want.
I find this perplexing...how did we as a UI culture going from having bad design that I could control to having bad design that I can't control?
Can't agree more, but it could be a cultural thing too - see all the discussion around Western vs East Asian web/UI design, for example. So your psychology might actually be the norm for an Asian user, if by "most users" you mean the West.
Well, I'm not Asian so I don't know. I'm a programmer tho. So for me things should be logical and controllable in orthogonal ways, and if there is lots of information display it in ways designed for maximum efficiency of absorbing information (good controllable aggregations, good use of the visual system and screen real estate, fast scrolling, ...).
Sure. Allowing to set ranges yourself rather than infinite scrolling that makes it impossible to go back to number X. A setting to always show little icons instead of only when the mouse hovers over it. Settings to organize lots of data the way you want. Settings to let a single mouse click on a bookmark star put it in the place you want rather than some place where you don't want it. Settings to always expand all content in a thread rather than showing things collapsed first so that you need to open each individually. Settings to put the refresh button next to the forward and back button. A setting to have multiple rows of taskbars in your favorite Linux desktop env. Settings to fix anything that's annoying and designed for less knowledgeable users or users with small screens basically.
One example that sums up most: Twitter. I just want to see/display the feed, not have the UI decide which ones to deem 'important'. I'd like for the app to allow the power of CSS to let me choose a layout as opposed to just picking one that doesn't really work for me.
The overall problem is that what UI designers think 'a more suitable UI' is, probably won't be for many users. Think of it like the open-office plan. It's horrible for employees, most people absolutely hate it, it lowers productivity, and yet corporations have people on staff who are implementing it as fast as they possibly can because it's more "suitable".
I did end up skimming a few more sections to see if it picked up & it was still all hollow, common sense stuff eventually descending into predatory gamification/addiction tactics. I know there are a lot of burgeoning startup types on here but PLEASE PLEASE PLEASE understand that responsible developers should strive to make a product that serves a real need for users. I want to make software that retains customers through convenience, features, value.
This guide reads more like those old guides for how to use SEO to drive traffic to your dung-heap. I don't know how others feel but I want to create a good experience for my users, not convince the masses that they are my users so I can go to some freemium lowest-common-denominator model.
Is web dev the wrong place for me or are we just in a bubble phase for these people who try to use their "clever interfaces" to regress humans to some animal state? (If you look at the chapter headers in this guide I don't think I'm out of line... its literally about addiction, sex, etc.)
We seem to be in a phase where there are many solutions being developed in search for problems which either are so trivial as not to be worth solving or which don't exist in the first place. The PC and the 'net were the incubators for this phenomenon which really took flight when the 'net became small enough to fit in the average pocket. Now there are 'apps' for everyone and everything, all vying for your limited attention.
Alls you need to do is go back in time a bit, say 150 years, to see some parallels with the way that newfangled electricity thing was seen as the solution for all life's problems. Doctors waving Geissler tubes over their patients to cure, electrical beds to improve women's sex drive, electrical hairbrushes because, well, it's electrical and thus cures rheumatism and constipation, electrical baths - both dry and wet - to improve health and stamina, electric underwear to 'cure any ailment'.
And now? Electricity has certainly revolutionised society, but not by using it to cure constipation or zap innocent bathtubs. The web also revolutionised society, and the mobile 'net is poised to extend that revolution even further - but not by creating dating apps or yet another calendar-planner-todo-list-organiser.
As a developer, the only thing that I have to say is: Why are you using a million canvas on your page? Are you trying to do some kind of benchmark? Please improve this.
Hey Joel- great stuff. Can you post your sources? For example, in the Cognitive Bias Lesson in the Decoy section, where did you get this factoid: "even though nobody chooses [the decoy], if you remove it, about 60% of people will choose the cheapest option instead.". Thanks for assembling these lessons!
I used to think UX meant a wider "systems" approach to the man-machine interface, incorporating findings from CHI, cognitive science, user studies, etc.
Now it seems like it is becoming a name for the design of marketing your product in the experience economy, not helping perform a task more efficiently (unless the "task" is selling to the user).
I'd like to see some discriminating term separating this from the UI-to-the-machine (UIttM ?) where streamlined presentation of the right info just-in-time is important, versus the "create more clicks for A/B testing" sales growth-hacking / I can make a web page with fonts.
I dunno, maybe this is a false dichotomy and reflects some kind of design thinking that UX is a term to apply to everything. Did a UX person design the pattern on my toilet paper?
Back when I was at university (not so long ago), usability was like you described - it helping the user get their task done efficiently. Not the more modern version of "delighting the user in surprising ways".
Thanks. It's a good quick run-through with the fundamentals and lives up to the title. I wish there would be simple and short illustrative examples/case-studies - or links to it - in the lessons to prove each point.
> If your marketing department wants to know anything about why the user is cancelling, put it in the form. Two pages of boring questions is a great way to reduce conversion.
> Break the form into many pages so it takes longer. Include links to FAQ pages. And avoid using defaults; it maximizes the number of conscious choices for the user.
> Ask them to explain their reasons for cancelling, and require at least 100 letters of text. Explaining is hard when your reasons are emotional.
Really? Obviously I don't want to make it too easy for users to cancel, but to make it too hard just seems petty to me. Asking them to fill out the reasons for cancellation is a great idea (and worked really well for us), but forcing them to write at least 100 letters is just horrible.
Also, there are some users that you really do want to cancel. You know, the ones that suck up twice as much support as anyone else and complain incessantly.
Surely its more important to be focusing on why users are cancelling, not luring them through a maze of "two pages of boring questions"? Does anyone reputable actually do this?
There is a sweet spot somewhere in between, but this seems to be irritating and would just make me want to write complete junk responses.