First on the agenda is a HUGE thanks to @justindavis for his time today

I literally solicited him by cold emailing and he was a legend. Took about 2 emails to lock this down. Quickest ever.

So thanks Justin, for being such a star

:boom:

Here’s the formal introduction:
Justin Davis is a senior user experience designer, and co-founder of First Chair Partners, a design and research collective working to help companies solve some of their most difficult design problems.

He is based in Tampa, FL, and has worked with companies like Google, Coca-Cola and Metlife to help deliver interesting and engaging experiences to their customers. With nearly 20 years experience in technology and design, Justin brings a wealth of experience from both the design and technical sides. He also loves boats and bourbon.

Read his blog at http://justindavis.co.

@justindavis What are your thoughts on conversational UI?

Yay, bourbon!

Woohoo! Thanks Hawk, appreciate the intro!

let me start with a little context, and we’ll dive into some questions here shortly

I found Justin when I stumbled on this talk he did: https://www.youtube.com/watch?v=pn6ezjkEuWk

And now I’ll let him give you an intro to the topic

Impressive! Look forward to picking your brain Justin

:+1:

The floor is yours Justin!

thanks!

So, we’re going to have a conversation about…conversations. I’ll get that kicked off with a little context.

As web folks, you’re all probably somewhat familiar with the request/response model. A user makes a request to a server, the server responds. Pretty straightforward interaction.

The other place where that happens is when we talk to people in real life – when we have a conversation, it’s the same model. We say something, someone responds, then we respond, and so on.

The great thing about conversations in real life is that they’re usually super engaging, we can often get lost in conversation and time flies by. That’s a state called “flow” that ‎Mihaly Csikszentmihályi wrote about in a book by the same name.

Getting in a flow state is desirable, because it means we’re super focused on the task at hand, engrossed, and enjoyment of the activity tends to increase with that state.

Well, if that’s the case with person-to-person conversations, and we think that interactions online follow the same kind of model, then it stands that there are things we can learn from real life conversation that we can apply to how we design those moments in our products.

So, what are the things we can learn?

First, we know that good conversations have a great beginning. This beginning is the “engagement” piece, where we first start chatting with someone. First impressions count when chatting with people in real life (you know quickly if you don’t want to talk with someone long term), and they similarly count in our product experiences.

When you chat with someone in real life, they don’t ask you for your name, email address, password and agreement to the terms and conditions, do they? No!! You just jump into conversation.

If that’s true, then it stands that our online interactions should follow suit. Instead of the first moment of interaction being a formal authentication into the conversation, how about just starting to talk?

If you have a product where someone writes blog posts, why not just let them get started writing, then once they’re finished, ask them to provide a name and email address so you can notify them of replies? By doing something like that, you’ve let them start “talking” to you immediately, instead of forcing them through some arcane setup, just to see if they want to chat with you.

Also, other ways of helping people get into the conversation easier are by paying close attention to the “zero state”. When you start chatting with some friends, and you walk in mid-conversation, there’s often someone who’ll say “Hey, we’re chatting about X”. That gets you “onboarded” into the conversation so you can participate.

Similarly, we can acknowledge that when people come into our products for the first time, we need to get them up to speed on what we’ll be talking about. We usually have terminology and wayfinding that we need to establish, and building that into a zero state or initial walkthrough is a great way to get people up to speed.

One we’re chatting in real life, there’s a period of time where we’re engaged in the conversation. We’re engaged because the people chatting (usually) follow a set of social rules that keeps the conversation on track. They don’t interrupt (well, most people don’t), they pay attention, and they’re polite.

We can do the same thing as our products are talking to our users.

Remember, when people are using your product, they’re not interacting with an app or a site. They’re interacting with YOU, mediated by an app or a site. If you keep that in mind, you’ll think more carefully about how to handle certain interactions. A few examples:

But what if he’s a Nigerian prince?? :wink:

– Ensure you’re providing really quick and clear feedback. When someone performs an action, make sure you’re telling them what to expect next, or that you’re working on a response. In a real conversation, we don’t just stare blankly as we think about how to respond, we fill that space with “loading indicators” like ums, uhs and other verbal flotsam.

– Use natural language! If you asked someone for a password to a secret club in real life, and they said “Oreo”, you wouldn’t say “Sorry, passwords need to be 6 or more characters with two special characters, but not comma, asterisk or dollar sign. Try again”. You’d say “Oh, that’s a bit short and easy to guess. Try something harder perhaps”. Natural language is a huge step toward better conversations with your users.

has joined #ask-the-uxperts

– Keep on topic. When you’re in a real conversation, you don’t skip around topics recklessly. The same should go for online interactions. Remember where someone left off last time they were using your product, and start back there. Acknowledge the time that’s passed between then and now (“Great to see you again! There’s some stuff to catch up on!”). Remember that in a user’s mind, it’s a single narrative over a period of time. Do what you can to help preserve that from session to session.

Finally – and I’ll stop yammering on so we can get to questions – remember that conversations eventually end, and that’s a good thing. We can’t talk to a group of people forever, and people can’t spend all day on your product. Think carefully about how people reengage – what conversation starters can you deliver to them later, that gets them back in the conversation? These need to relevant and personalized, so that like talking to an old friend, it feels like picking up where you left off.

In short, we can learn a lot from how we talk to people in real life, and the dynamics of those conversations, that can help us think about how we design those moments into our products. For many people, talking to real life people is a far more enjoyable experience than using some dumb product. The closer we can make the conversations between our products and our users like those in real life, the closer we’ll get to bridging that gap.

nowwwwww, questions and …wait for it….conversation :slightly_smiling_face:

Haha – questions are GO!

I’m interested in what happens to trust where you put initial sign-up/auth at the back of an action, e.g. blogging. Do people exhibit trust issues with this kind of interaction IYE?

so, @alex.lee, re: conversational UI, I’m a big fan. Can you tell me more specifically what you’re curious about?

@justindavis I’ve been working on chatbots lately and trying to integrate it into user engagement flow in new visitors to a website to answer FAQs

i think they’re a great tool, but obviously they have to be executed on very well to work. It’s a fine line between a real conversation, and conversation-esque, right? :slightly_smiling_face:

I think you have to work carefully on how to handle responses and the conversation flow, to ensure things feel as natural as possible

If you’re shite at small talk, can you still be good at conversational interaction design? Asking for a friend. :grin:

Yes, I’m interested in UX at the intersection of AI and natural conversations, without coming across fake and insincere

interacting with bots is still a bit of the wild west, but I think it’s something that’s going to be a key part of product interactions moving forward

@alaa pinned a message to this channel.

I worry about setting false expectations and eroding trust

@jakkii on the issue of putting signup at the back of a process, in my experience, I haven’t seen many trust issues. I think the key there is setting expectations – if it comes across as a bait and switch (“Start now, no accounts needed”), and you hit them up randomly, it can definitely lead to that

but, I think that if you set up expectations correctly (“Get started, we’ll worry about accounts later”, etc) it works quite well

@alex.lee, yeah, I think like @jakkii’s question, it’s some about setting expectations

“I’m <name of bot>, I can help you with some basic questions, and if I don’t know the answer, I’ll get back with you”

dumping out those things to manual support queues

and @jakkii, the good news is, even introverts can do this well :slightly_smiling_face:

I was thinking about forms and surveys – most don’t follow a conversation format but there’s no reason why they can’t

forms are a PERFECT example where this goes wrong

we just jam a bunch of fields in front of a user, because the database is structured that way, and we don’t think about the bridge between human and computer

unfolding forms or surveys over time, with smart branching logic, can make arduous tasks feel much more conversational

Turbotax proved this really well

by getting people to go through hours of tax prep, without absolutely hating themselves

a small miracle, to say the least :stuck_out_tongue:

@justindavis are there any good / bad examples to show for conversational UI?

there are definitely a bunch of examples @amirasallam. I wish I had some screenshots here handy, I’ll try to dig some up. Turbotax is an example of it done well, and there are some support interfaces that are doing it pretty well. TBH, some phone trees with voice recognition also do it well.

Tho, some also do it terribly :stuck_out_tongue:

Any form services or tools you recommend for creating conversational forms?

I like typeform.

forms are a basic example – errors messages like “Invalid email format” are bad conversational devices, right? But saying something like “Oh, that doesn’t quite look like an email address” is a much better way of speaking

And outgrow’s survey tool

so much of this is in the microinteractions like that

typeform is certainly well done

i’m a big fan of it

It just feels like a conversation

ding ding ding!

Ha!

How do you strike a good balance between natural conversation (which obviously takes a lot of room and adds load) and CTAs

have you worked with global apps where “conversational” can vary?

Sometimes you just want to ‘get through the process’ rather than be walked through a process with a friend

@alex.lee, good question. I think you can have a succinct conversation, and keep away from too much chit chat. Good conversation isn’t always long – sometimes it’s short, friendly bursts

and it can be as simple as simply changing the existing copy to be more conversational, friendly

So be human. Don’t ramble

bingo, nailed it

That’s where I think Chatbots fail.

To chatty

hey now humans are great at being chatty too

@isha, also a good question. I haven’t had a chance to work on many things in cultures where social norms for conversation are very different, but like anything multicultural, it definitely bears paying attention to

that said

at our core, everyone is human

Cultural differences are hard

and speaking to people, and interacting with them like people – not machines – is a universally attractive thing

true…translation makes it a little harder to be conversational but I get what you are saying…

indeed, there are certainly some low-level challenges to deal with there

Yes, I’d imagine if you put too much colloquialism in front of someone for whom English is not their native language, it may be more challenging

one technique we use a lot is to sketch out “wireflows” of interactions, which lay out how the conversation will go, before we get into UI details

Do you have examples of apps that are conversational that aren’t very wizardy (like turbo tax AND typeform are super wizardy)

seeing the back and forth from a high level really helps to map that out

so, I actually think of Medium as quite conversational. It’s a different way of thinking about it – Medium is great at conversation, because when you go to write, it gets out of the way

So in the email password prompt, you might say “Hey, that password seems short. Can you think of something longer?” and have a tip box at the bottom with suggestions on uppercase, lowercase etc?

@justindavis have you had a time when product owners were resistant to using conversational UI? How do we get product owners to approve that :) ?

like a great conversationalist who listens, more than talks, Medium does the same thing

that’s another part of being good at conversation – is knowing when to let the other person speak

@alex.lee, you’ve got the right idea!

@amirasallam getting stakeholder buy-in can definitely be tough at times, ESPECIALLY in enterprise scenarios

because often, speaking more conversationally – at least in terms of tone and style – can come across as glib or overly informal

that said, good conversation can even happen with formal tone

Or of not getting past ‘legal’…

if you go to the hospital, they’ll still put you through a formal process to cover what they need, but they’ll do it in such a way (hopefully) that feels kind

so, you can strike that balance between being formal in language, but conversational in process

again, when you go to the hospital, instead of jamming a form in your face immediately, they’ll say “Hey Mr. Davis, we’re sorry you don’t feel well, we’ll take care of you. Please fill this out so we can get you on your way to a doctor”

so, formal – to a point – covering the legal issues that need to be covered, but still human

there are definitely “levels” to designing things with a more conversational interface. The low level, copy, and the higher level narrative that unfolds

We’re at the end of the queued questions, so if you’ve been sitting on one, now is the time!

When and how often might you go about iterating conversational flows for your UX when you’re designing such conversations?

Curious about your thoughts on text on buttons…

@alex.lee, do you mean in the design process itself?

@justindavis do you recommend books for this topic?

Yes

Me too!

so, in our design process, before we get to screen level details, we try to write out the narrative of the user, including the back and forth with the system

that’s in the initial design work

@justindavis, I get your point about being conversational, and heartily agree, and I think there needs to be a balance. We don’t want conversants guessing what the other one knows. In the case of the password rejection, isn’t it in the end more civil to state complexity requirements instead of saying the proposed password was simply and easily guessed?

@justindavis that’s a great suggestion. Like a storyboard but for conversational narratives only

only once we get those high level flows on paper feeling good, do we start to fill in the lower level details. So we work out the high level conversation details, and move into the lower level details

@alex.lee yup! It’s exactly a storyboard

Very interested in learning more and would appreciate it you elaborated on the initial design process where you write out the narrative of the user…and how you do it, and what you call it and who all are involved.

@isha, text on buttons – what are you curious about specifically?

As far as suggestions on conversational interactions that aren’t wizards, you can give Alexa a spin at https://echosim.io/. It’s entirely transactional; it seems hardly any context is maintained from question to question. Their API is here: https://developer.amazon.com/public/solutions/alexa/alexa-skills-kit/docs/alexa-skills-kit-voice-design-best-practices

@amirasallam re: books, I really like Designing the Obvious by Robert Hoekman, Jr for helping with this kind of thinking

@justindavis the PMs in my company tend to focus on “save” vs “ok” vs “apply”….but then we are also a very global company and these translate differently…

@uxcourt certainly. Having good conversations is about being clear too :slightly_smiling_face:

of course, it brings up another question as to whether those strict requirements are a good idea to begin with, but that’s a whole other conversation… :stuck_out_tongue:

Would like some tips.. conversational UI seems great! I am thinking to use this narrative to design an app for patients with low literacy.. the app will help them fill out some health forms.

@isha, we use a combination of a few things. One, we write out actual narratives – paragraphs that describe how a user completes a task. We then take that and make storyboards of the interaction, iterating on it until it feels as conversational as possible

SHAMELESS plug, I wrote a blog post today talking a bit about that: http://justindavis.co/2017/04/11/how-to-onboard-a-product-designer/

@alaa low literacy is a FANTASTIC place to use this technique

shameless :slightly_smiling_face:

for a couple reasons

1. you can unfold an interaction over time, instead of all at once, which is easier for folks to digest who might have a harder time with density

2. you can use language that blurs the interface, so to speak, so it doesn’t require that they have prerequisite knowledge about technology to interact with the product

it’s amazing how easy it is for us to assume people know what we mean when we use certain terms

like “special characters”

which ones are special?

the “J” because my name starts with it?

:stuck_out_tongue:

J is certainly special

How would you evaluate programmatically translated copy when your team and available testers are only english speakers?

and, as always, in UX, it’s about knowing your audience

@lmacneil the bad news is, if you’re evaluating copy in a language other than your own (or one you have deep experience in), you probably have to find that resource somehow. I’ve had success in reaching out to folks – friends, other colleagues, etc. – to have them look at things for me to gut check how it sounds

so, you might have to get creative on that, but no way to really substitute someone who really groks that culture

:slightly_smiling_face:

@lmacneil can you outsource to a localization company and develop a relationship with one of their native speakers? Can you go to a local community college and lurk in the ESL classes and maybe entice an instructor to help? Also know that dialects are important; a Portuguese speaker will sound kind of foreign in Brazil, and a person from Barcelona might have no idea what a person from Merida MX is saying.

yup, those are great ideas @uxcourt

guerrilla techniques for translation :stuck_out_tongue:

@uxcourt awesome :smile: & @justindavis TY both for the advice

We have about 10 mins left… last chance to get your questions in :slightly_smiling_face:

We’ve tried doing this in a form for reporting cyber security incidents. Many people will be stressed or won’t know what the thing is called that has happened so we made a narrative component to it (second screen). There’s still lots to tweak / improve but it’s a great start. https://www.cert.govt.nz/businesses-and-individuals/report-an-issue/

ah, excellent, I love that example!

there are so many places – like government services – where this kind of approach is sorely needed

and TBH, so many mistakes can be prevented by this level of clarity

and mistakes cost money, so clarity, and approaching things from a more conversational stance, isn’t just a branding move, it’s an actual cost-savings move as well

in general, what if we’re supremely successful designing conversational interfaces? There’s an episode of Black Mirror (Be Right Back, Season 2), or the film Her that comes to mind as a potential negative outcome. I guess there’s Star Trek as a potential positive outcome. What can we as designers do to ensure our tools build towards that positive outcome instead of the dystopic one? Feel free to pass, because this is really a philosophy question.

@uxcourt ha! Thanks for the tough one. I haven’t seen Black Mirror or Her, but I’d say that dystopian outcomes are borne from dystopian intent

haha! intent vs impact

I think designers have an ethical responsibility to guide the process and not just be tools of the trade

at the end of the day, we still control the machines (so far, let’s talk again in a decade), so we can guide those interactions towards positive means

Black Mirror the whole show is a bit jarring, but gets me all fired up to do better work

gonna put that on my list to watch!

in both of those scripts, the humans developed emotional bonds with the machines, and the machines were incapable of fulfilling the expectations, leading to greater depression and isolation.

@danielle loving the form on this site you shared, btw. Really clever work.

but that’s still fiction and we’re a long way off from anything that believable.

@uxcourt wow, yeah, it’s probably coming at _some_ point, but hopefully we’re a bit far off :stuck_out_tongue:

This session has been amazing!! Any last questions before I cut Justin free to enjoy his evening?

thank you!

Any conversational tips for different platforms?

Also, feel free to connect on Twitter (@jwd2a) and we can keep chatting there!

Yes, this has been super helpful!

Thank you!

@alex.lee I think at a high level, platform differences come down to technical execution issues. If you start with a high-level narrative, and work that out, the platform differences end up being technical accommodations more than anything else

Thanks so much everyone

Big thanks to @justindavis for some really interesting insights

And thanks to @hawk once again

really appreciate y’all for coming out, this has been a blast!

Thanks SO much for your time, your energy and your wisdom today @justindavis

absolutely, honored to be a part of it!

You totally nailed it!

has joined #ask-the-uxperts

Now go – enjoy a gin and tonic with your wife.

Thank you!

Thanks @justindavis

Ha! You read my blog :stuck_out_tongue:

I did. :slightly_smiling_face:

I shared that one too because I love it

thanks y’all, hope to see y’all around again real soon! I’ll try to bang around this Slack team more often as well!

First up, a huge thanks for @feather for his time today. He’s a crazily busy man and I’m honoured to have him join us.

He’s a man that needs little introduction, but here goes anyway:

Derek Featherstone has been working as a web professional since 1999 and is an internationally known speaker and authority on accessibility and web design. He leads the team at Simply Accessible, based in Ottawa, Canada.

Derek always puts the user first and strives to make the web a better place by designing experiences that are easy to use for everyone, including people with disabilities. Derek’s ideal accessible experience combines engaging and rich content with brilliant design and technical development excellence. That user-centred focus sets the course for Simply Accessible, and—more importantly—their clients.

I’ve been lucky enough to see Derek speak, and I’m looking forward to learning from him today

@feather The floor is yours

Fantastic, thank you @hawk :slightly_smiling_face: Very happy to be here, and very happy to use this medium, as it has a lot of advantages

One of the advantages is it allows us to engage in a very different way than we might at a live event.

We use Slack for pretty much everything at work, and one of the things that we’ve been doing a bit more of recently is inviting clients in as single channel guests to participate with us there.

AND one of the benefits of that? We have been able to have real time conversations with clients that are Deaf.

Awesome

It was one of those moments where we as a team smacked ourselves in the head and kind of said “Oh, of course!”

And it helped us ask a really important question — related to today’s topic

When we say we work with everyone, what does “everyone” actually mean?

One of the foundations of the work we do is inclusion, and we need to ask regularly, what are we doing that is exclusive?

For many many years, we’ve seen solutions to accessibility problems that are focused on “how things work in a screen reader”

That’s okay — its pretty natural to want to work on those types of accessibility issues

But we’ve seen situations before where a focus on “how this thing works with a screen reader” really turned into “lets make this work with JAWS 14″

and we’ve seen other situations where it turned into “Lets make this work with a screen reader, regardless of what happened with other assistive technolgies”

And these were VERY well-meaning people that had a tremendous history of inclusion and accessibility

They had created an interface that worked extremely well for a screen reader

BUT

when someone using voice recognition (Dragon Naturally Speaking, for example) tried to use the exact same interface, it didn’t work properly

Another forms based interface worked perfectly with a screenreader. It performed exceptionally well in the usability testing we did.

For people that used a screenreader.

When we tested the same interface with people that had low vision — they were using a magnifier — it didn’t perform well at all. And the client said “But… but… it’s accessible! It passes all the checkpoints!”

And yet, it couldn’t be used.

When we talk about “Designing for Everyone” we very often mean “We are designing for the people that we can envision right now”

or “We are designing for people that are just like us”

or “We are designing for a fictitious ideal user”

That’s not everyone, but we’ve tricked ourselves into believing it is.

So we have to open the net a little bit wider… when we talk about diversity and inclusion, what is that conversation about?

When we’re talking about making sites and apps accessible, we can cast a lot wider than we have been.

We had an intern that was working with us, and their comment after about 3 hours was “You don’t have anyone working here that is under 30, or over 50”

We were mad. Because they pointed out what had already been staring us in the face.

Diversity and inclusion means a lot of things

Age, gender, nationality, culture, ethnicity, philosophy, experience…

and ability

and… and… and… likely a lot more than any of us can easily consider in a single sentence.

But all of that is where the question comes from. Which everyone do you mean?

If we truly mean everyone, we *need* to go wider, deeper, farther.

So, from a practical perspective, we need to make sure that things “work” for people that use any type of assistive technology.

Ok – who has questions?

Screen readers, voice recognition, magnifier, braile displays, switches…. and more

What are the top 3 mistakes that are made when it comes to accessibility?

@jacqui_dow5 Great question :slightly_smiling_face:

I’m going to answer it in two ways… first philosophically:

Are you suggesting that we test our sites against EVERY assistive technology? That can prove to be cost prohibitive for the smaller developer.

1. Thinking that you’re done. You never are. You’re always iterating and always getting better. Or at least should be :slightly_smiling_face:

As a lover of data-grounded personas I’m wondering how design tensions between different assistive technologies are tackled?

And now practically…

2. Forms. Difficult designs that don’t get translated well to code.
3. Focus. Making things that can’t be focused naturally is a big issue.

Can you expand on 3 please? Not sure I understand?

2. Cont’d: 1:1 relationships between a label and form field are easy. But when you get one label that maps to 3 form fields it gets tricky.

If we work in a11y, how can we know about all abilities and assistive tech types, to know that we’ve made the most inclusive experience? User research is great, if you’ve got oodles of users and cash and time, but some clients just want a compliance audit.

2. Cont’d: think of a phone number with 3 fields, but only one label

3. Cont’d: I say focus often now, because focus is something that impacts both traditional desktop AND touch screen devices

Do you feel like reducing the goals to WCAG 2.0, level AA or AAA accomplishes the concept of “everyone” or do you feel like there is room for improvement there?

3. cont’d: instead of saying “Keyboard” — it’s really more than that

its about how things work on a touch devices as well… and those are both often problems of Focus. In other words, actionable things need to be Focusable

What do you mean by focusable?

@r_saviano:

> Are you suggesting that we test our sites against EVERY assistive technology? That can prove to be cost prohibitive for the smaller developer.

No, definitely not against EVERY assistive technology. But, widening the net is important. We often don’t recommend you test with a screenreader anyway and work towards the standards and best practices. As an example, you don’t need a screen reader to tell you about heading structure.

Thank you :slightly_smiling_face:

@r_saviano I DO however, recommend that at some point you try something else outside of a screen reader. You then catch principles that you need to understand quickly and incorporate that into your work

@r_saviano So NOT overkill on testing everything to the max. But thinking about other scenarios, yes.

I work on a web application, and want to do an accessibility review, but am unsure where to start. I’ve ran it through WAVE, but don’t really know what to do next. Do you have any tips for this?

Hold fire on your questions for a few minutes all. We have quite a queue building up.

@dean:

> When we talk about Accessibility, it is primarily developers that are talking about the technical challenges, automatic testing, ARIA, and inclusive code snippets… aside from the basics (eg. color contrast / font size / responsiveness), where do you see designers / UXers role fitting into this?

That’s a BIG question. To me the designers most import role is making sure that accessibility and inclusion is incorporated in the vision of the [thing that you’re creating] from the beginning, and then making sure that lives on through the process.

@dean: there’s a LOT that designers/UXers can do to ensure that accessibility lives in the process, not just in code.

A :question: for you @feather – we had a WCAG content audit recently and are working on the fixes. How might teams better go about inclusive design up-front (and throughout) project life-cycle rather than have the bigger job of repair work in hindsight?

Sorry, just saw this

@irith:

> As a lover of data-grounded personas I’m wondering how design tensions between different assistive technologies are tackled?

I’d love for you to tell me more about that… I’ll start with this though, on the assumption (perhaps faulty) that you mean designing something that is good for one group, but not another. We tend to ensure two things that help:

1. We include different accessibility needs in personas from the start so that the needs are expressed from the beginning.
2. We do usability testing with people that have different types of disabilities. That helps us provide the data we need to be able to make the best decisions. We let performance decide what our next iteration is, and which direction we ultimately take next.

@allyraven:

> If we work in a11y, how can we know about all abilities and assistive tech types, to know that we’ve made the most inclusive experience? User research is great, if you’ve got oodles of users and cash and time, but some clients just want a compliance audit.

Part of the solution, I think, is expectation setting up front — making it clear that the ultimate decision maker about whether or not something is accessible is whether or not people can use it, not if it is “compliant”

@allyraven: Another part of the solution is finding ways that don’t require oodles of users and cash and time :slightly_smiling_face:

True, but without much usability testing, are you expecting me to be an expert in screen reader, dragon, switch devices, eye tracking software, blow devices, etc?

@allyraven: Of course — oodles is a relative term, so we tend to try and find out what the resources are and then use that to include something more than we would have otherwise. There’s always a way to sneak in a little more that helps you grow such that you’ll have that better understanding

@allyraven: So what I’d expect is that you’re not an expert in all of them. But I’d expect you to challenge yourself a little, and incorporate just a little bit more. If you don’t know much about voice recognition, set a goal that in the next 3 months, you’re going to spend a bit of time using Dragon to get at least a little more familiar. Iterations FTW. Hope that makes sense?

Oodles – my favourite unit of measurement.

Potentially stupid question: where could I find a list of all these standards, and perhaps compare them?

No questions are stupid :slightly_smiling_face:

@feather: I guess I was thinking of how qual data can produce a number of personas, but we often only select a small number of ‘design personas’ to design for. I guess I was assuming there might be strong competing requirements between different assistive needs… and how you prioritise across those. (Just me guessing…)

@aleonardpalmer

> Do you feel like reducing the goals to WCAG 2.0, level AA or AAA accomplishes the concept of “everyone” or do you feel like there is room for improvement there?

It’s a great start – WCAG is built on years of other people’s experiences. They’re a fantastic starting point. But what I would love to happen is a bit more testing with real people. Even lightweight stuff with real people will help… its amazing the insights that you can get from a 1 hour session with someone using assistive technology, all for the cost of somewhere between “cup of coffee” and “a decent meal”

@jacqui_dow5:

> I work on a web application, and want to do an accessibility review, but am unsure where to start. I’ve ran it through WAVE, but don’t really know what to do next. Do you have any tips for this?

Web apps are potentially tricky… is it a single page app with lots of ajaxy things, or is it more traditional server-refreshy? Either way, one great way to start… run through with someone with a disability trying to complete typical tasks on that app. See how the results stack up to what you found through WAVE. Compare. Contrast. Then prioritize the things you found and start to address them.

It’s a huge web app!

@jacqui_dow5: it sounds like it might be more complex than that, but feel free to reach out if you want to talk more about “what next”

@feather I’m about to start a graduate program for HCI – Human Computer Interaction – I want to focus on accessibility- Do you feel there is a big market for accessibility experts? If not, how can we show people how important this is?

we have a few people lined up for usability testing as we have our own lab, they contacted us as they were struggling with some parts and were using assistive technologies

The other side to it is, we use Bootstrap and I wonder how much this is affecting the accessibility of the application?

@vernon:

> we had a WCAG content audit recently and are working on the fixes. How might teams better go about inclusive design up-front (and throughout) project life-cycle rather than have the bigger job of repair work in hindsight?

It’s all about early and often, right? We need to ensure inclusion from the start. That means a lot of things, but the biggest difference maker to me is getting designers to think differently. Getting them to understand what kinds of challenges people with disabilities might face is key. That is best accomplished by them working with people with disabilities from the beginning of their design process.

@vernon Even just talking with people with different disabilities will help. Remember – it’d be more than they were doing before :slightly_smiling_face:

@vernon: so the plan needs to be testing assumptions, building things, learning how they work, and then iterating.

@vernon: let me know if that helps and feel free to ping me… we have some resources that may help!

How/where do you recruit users for testing? Particularly those with less visible or obvious abilities?

@shuvy:

> Potentially stupid question: where could I find a list of all these standards, and perhaps compare them?

WCAG 2.0: https://www.w3.org/TR/WCAG20/

Start there :slightly_smiling_face:

its very dense, but if you’re looking for the standard, that’s where it is

@lizd:

> I’m about to start a graduate program for HCI – Human Computer Interaction – I want to focus on disability – Do you feel there is a big market for accessibility experts? If not, how can we show people how important this is?

YES I do feel there is a big market. Of course, that depends on how we both define big, but, yes, we see no sign of things going away, and there is a continuing growth interest. So I’d say yes, definitely a big market.

@lizd There are very few people that specialize in this area, so I think its a great idea!

Do you have any tips/tricks for implementing accessibility into Angular applications? (question from our dev team)

Great to know! I am physically disabled, and hope to bring a new perspective to it as well.

My question is about the balance between usability and compliance. So as a UX designer my goal is to design the most usable design I can but sometimes compliance means I have to sacrifice usability for our users with disabilities because of WCAG or 508. For example I was forced to add a skip nav but the page was simple enough that the skip nav added more steps than would have been there without it. I work for a government agency so compliance is not a choice. But I felt I could have given our disabled users a better experience than what compliance was forcing me to do.

Sorry that was long.

Thanks guys – no more qs now please. We have enough to fill the last 5 mins!

@allyraven:

> How/where do you recruit users for testing? Particularly those with less visible or obvious abilities?

We work with a lot of different organizations… international, national, and local advocacy groups. We’ve built up a panel of several hundred people that we get involved in our usability testing for client work. Many are recruited through twitter, but also our relationships with those advocacy groups. Local colleges and universities are also GREAT sources.

@jacqui_dow5:

> Do you have any tips/tricks for implementing accessibility into Angular applications? (question from our dev team)

Definitely. We’ve got a big writeup at http://sateach.es/spangular and have some more updates coming soon!

awesome, thanks, I’ll send that over to them!

@josh3cs:

> My question is about the balance between usability and compliance. So as a UX designer my goal is to design the most usable design I can but sometimes compliance means I have to sacrifice usability for our users with disabilities because of WCAG or 508. For example I was forced to add a skip nav but the page was simple enough that the skip nav added more steps than would have been there without it. I work for a government agency so compliance is not a choice. But I felt I could have given our disabled users a better experience than what compliance was forcing me to do.

Oh, that.

I’ll generally choose actual results with real people over “compliance” when it comes to conflict like that.

I don’t think I can do that

@josh3cs: it always depends on context, so there’s lots I can’t necessarily answer here without seeing the situation. However… things like conforming with WCAG and doing things the way the organization says it has to be are different

And I’m fairly sure you could argue the “no skip link” version of it is still compliant.

Feel free to ping me if you want to dig in a little more… you have to remember to ask though. If you’re going to get in a debate about accessibility, do you want to spend your political currency on that particular issue?

And on that note, it’s the end of the hour…!

Massive thanks again to Derek for both his time, and his knowledge.

Nice work on keeping up with the questions too!

And of course, thanks to the rest of you for joining us today

Welcome everyone – thanks for joining me and @andyvitale today.

And an even bigger thanks to Andy, for giving us your time

I’ll start by introducing @andyvitale.

Andy Vitale is a UX Design Principal at 3M, where he is focused on translating human insights into actionable experiences that improve the healthcare industry.

Andy is responsible for leading a team that creates immersive, emotional experiences for 3M Health Care’s enterprise solutions that seamlessly integrate digital and physical workflows for patients, payers and providers.

Aside from his primary role at 3M, Andy is an adjunct professor for Kent State University’s User Experience Design graduate degree program and often speaks at conferences and events.

Morning all :slightly_smiling_face:

I’m really pumped about today’s session because it fits in perfectly with our theme for the month, which is culture and stakeholders. So Andy, over to you for an intro to the topic today.

:wave: Hi everyone.

Thanks for joining us today

One of the most important things I have been seeing throughout design, is the importance on solving things that may not be traditional design problems with design

One of the biggest things that impact us as designers is culture. Particularly company culture.

I work for 3M’s Health Care Business Group

3M is a 100+ year old company with a historied tradition in innovation through science

When I started on the team I was UX designer #2

Now we’re at about 15

When we think about design culture within an organization, is it where we expect it to be?

Or are decisions still being made without design?

Is design being brought in at the last minute?

Do the majority of projects get buy-in from the organization when they are started in non-design functions?

Are outside vendors and agencies often championed as experts more than the internal team who is just as capable?

Does it seem like the UX team is the only advocate for the user, and the business thinks that they can still just put out a product and people will just use/buy it?

Do any of these sound familiar to you folks?

YES

This is all hitting close to home for me.

yeh 100%

Oh yes :slightly_smiling_face:
> Are outside vendors and agencies often championed as experts more than the internal team who is just as capable?

So for my team, how do we come into a company that is well regarded, well respected & stands on a century plus foundation of innovation?

How do we break into a matrix of established technical, scientific, mktg & business teams with robust agendas to make impact and money?

At 3M less than 1/100th of 1% of employees are UX designers

How can someone like us make a difference in enterprise, whether in house like me or for a client? Being the cool designer is not enough to influence people.

[You’re welcome to jump in with questions at any stage ]

My questions are the ones @andyvitale has already posed! …. eagerly awaiting the answers… :grimacing:

Good UX and candy!

I see a few people typing so I am going to give them a chance before I talk about some of the ways to solve this

Can you tell us how you addressed all of those wonderful questions posed in your intro?

@cystinosis Candy helps.

Good UX for us are pretty much table stakes. It is expected for us to do a great job.

‘Taking them on the journey ‘ …..

how do we show that we are here to help and not to take over their jobs – some stakeholders are territorial :stuck_out_tongue_closed_eyes:

So the best way to focus on design culture is to do what we do best, treat it like a design problem.

How do you transform your product team to being design-led rather than PM-led or marketing/sales-led?

We’re all familiar with the design process in one variation or another but to align on terminology. There are a few steps

^ that

How do you imbed a culture of UX in and organisation which arguably hasn’t jumped on board with UX being a ‘thing’?

I like to think of UX design as a team sport

How did you gain access to people in position to make the culture shift and how did you communicate the value of UX?

(yeah – in 2017 !!)

@andyvitale What kind of early wins can you get with UX without executive support? I’m curious how you got things going to the point where you could even show wins.

So the best way to tackle stakeholders who are territorial, and what answers all of the questions so far is to start with the discovery phase

What is the problem – in this case it’s culture

What are the goals of the organization? The customer? Each stakeholder has a different agenda. How do we align them?

How do you handle stakeholders who keep saying they will get you answers to questions like – who is the decision maker? And never do?

How do you convince key stakeholders that the research phase is important – important enough to get budget to bring designers to users, important enough to give it time, and important enough to require it before something goes from a “feature request” to the backlog?

Peter Merholz said “The experience IS the product, and
the only thing users care about.”

that is a good foundation to gain alignment on

I see the same problem in my organisation, the marketing, or steerco board take control and hear on their on feelings and not the facts, the tests, the designs. Its kinda hard to teach them, that there a “proofen” ways of optimisation…

So @laurelwn that is tricky

You have to start doing things without them, let them be the holdup

We can’t wait for people to email us information, we have to make our most informed guess and use that to bring to the table as a talking point

But doesn’t that mess up the design sprint process?

Sometimes you have to start somewhere before the design sprints kick off

So let’s get back on track with culture and evolving design culture and then I will get to some of the other individual questions, it’s a process to get there.

Still in the discover phase – Look to companies you admire. See who else has a firm grasp on design culture & maturity. How do they do things?

This is about treating culture like a design problem, this will help you get the seat at the table and will lead to everything else we are wanting to solve

Listen to people within your company. Understand their role & perspective on users. Identify their needs for design

Observe how users & design play a role in products, and where & when they are engaged. Are there opportunities to better involve both in the process?

@andyvitale One challenge we have with this is really understanding how other companies are doing these things – unless they have it publicized, we never know, and it’s hard to find that info. Do you have suggestions or resources for this?

Empathize with others who work with design. Although the business may not understand design decisions, designers don’t always take the time to understand the business constraints and strategies before starting to problem solve. Try to understand how design goals can better integrate with marketing, technical, lab, business goals.

Although UX design is key to both strategy and success, we have to realize that we are not the sole provider of either of those

Here’s something to really understand. UX is not the center of the corporate universe.

As we start to define our ideal state, share your strategies with other members of your organization to gain diverse perspectives from cross-functional colleagues. Embracing transparency and inclusion will strengthen your strategy and help deliver a stronger, more aligned vision.

Create a need that focuses more on solving user problems and improving outcomes rather than providing features.

UX isn’t done in a vacuum. You have to have access to people of all skill sets (developers, SME’s, marketers, scientists, etc.). You will spending a lot of time together solving problems and sharing insights. Build trust with your colleagues and inspire them to focus on providing the optimal experience for your users.

But at some point people are tired of hearing designers talk about design.

Leave the UX/design lingo behind – clearly communicate solutions to the rest of the team in words they understand. The business mgr is worried about profits, the marketing mgr about brand and experience – we can’t confuse them with design speak.

There is a trick to influencing those around you to think like you do.

It’s called DOING THE WORK.

You have to do the work. Getting a jump start on the work and showing progress can influence so many conversations as well as clearly communicate what is still unknown. By driving with design you are demonstrating accountability and this is a great example to set when trying to influence culture.

Things like improving ease of use and reducing errors should be a given. That’s what they hired us to do. When we do that, it’s nothing they didn’t expect.

So here’s where we start to move the needle

In the beginning you have to work on projects where you bring the most value. You can’t work on every project or create solutions that are everything to everyone. Understand where you can make an impact and start there.

Designers are natural storytellers, whether with words or designs. Offer to help stakeholders visualize some of their objectives – this is a great way to gain their trust and build a relationship with them. Before you know it they will be sharing your story, which you influenced, at all of their meetings. Word of mouth is a powerful thing.

You put in a lot of effort and the project was successful. Invite others to celebrate with you – everybody loves a winner. Share your case study with as many people as possible – your wins will work their way up the ladder. Let executives communicate your wins, people will line up to work with you.

But sometimes all of that may not even work

What do we do?

Anyone?

I’m totally guilty of beating everyone over the head with design speak without having much actual work to show for it.

We all are @jenniferlong but we have to be better

When all else fails and we need to figure out what to do next. Sometimes you have to bring your own seat to the table.

Often times we would just show up to meetings without an invite.

But you have to be careful

You have to make sure you add value to the meetings or not to push too hard before you’ve established you place otherwise you won’t be invited back. Don’t give off the impression that you want to tell them how to run their business but you don’t want to understand how the business runs.

Now this goes to some of the earlier questions you asked

How can we clearly demonstrate that this way of thinking is beneficial to the organization?

How can we measure the impact of UX to the user and the business?

It’s not that we want to justify our value, but more that as designers we are curious about our outcomes.

Before we do that, there is one thing we really need to understand.

STAKEHOLDERS UNDERSTAND DOLLARS

Outcomes that are clearly easy to measure are black and white.

Figures like sales, revenue and ROI are really easy to measure because they have a specific dollar amount tied to them. Compare that to a previous number and call it a day.

I remember a time when analytics weren’t readily available, now we can measure shit we never even knew existed.

Things you never even knew existed are being tracked & because of this, we can measure and tie a dollar amount to them.

@andyvitale What technique do you use to help them visualize their objectives?

What about productivity

There are a many factors that affect productivity. From time to complete a task, to reducing the steps in the workflow to complete a task, or just reducing the amount of errors users have.

We worked on a project where people were doing medical coding for billing.

Earlier @laurelwn posed a good question about no one really wanting to make the decisions which need to be made. Can you expand on that?
> How do you handle stakeholders who keep saying they will get you answers to questions like – who is the decision maker? And never do?

We increased their productivity by 15 cases per hour.

That’s 100 more a day.

Now I’m not great at math off the top of my head but I know that equates to a lot of money.

Sure @chels and @laurelwn

Meet with them indiviually, figure out what their goals and expected outcomes are

All of them

Then put something together – we’re designers – we are way better at telling stories than they are

Then share it with them all in the same room, and they will agree with what you are showing.

That will start to build alignment and allow you to influence the decisions for them

Someone has to take charge.

Why not us?

Here’s another question to you guys

for showing how to measure our impact

Are there training costs associated with your product?

Anyone?

Yup

Simple answer to that – If you make it easier to use, it will require less training.

What about development costs?

Think about Excel.

I’m a designer, if I am lucky I use 5% of Excel

That means 95% of the features aren’t important to me

Let’s eliminate these unnecessary or marginal features.

Find your 5%, or 25% – whatever that % is and focus on that. This will free up the time & money of resources and allow them to work on improving the core features and finding new features that add value.

Why do we constantly put in time to update areas of our site, or pieces of our software that nobody uses? It’s stupid.

But what about things that aren’t so black and white?

We may have to do a little designer math.

User feedback is a nice little gray area.

Surveys, SUS Scores, Loyalty & Customer Satisfaction are all extremely difficult to measure

but not impossible

If people love your product, they are likely to recommend it to colleagues.

This can lead to new customers which can be tracked & tied to a $ amount.

But getting new customers is easier than retaining existing ones. So figure out your customer retention rates and compare that with your cost per acquisition on new ones. If you need help, I imagine Eugene in accounting can tie a $ amount to these figures. Take him to lunch and he will help you with all of the numbers.

Has anyone here ever killed a project?

Nobody wants to see products die but sometimes when you can stop the bleeding you deserve to be rewarded.

Sometimes the market isn’t ready for your big idea.

Sometimes companies make the wrong decision.

Sometimes your idea wasn’t as good as you thought it was.

@andyvitale I think you intended it’s easier to keep existing customers than getting new ones.

Good catch @chadc

I’m trying to keep up with all of the questions I am getting directly

Sometimes your execution flat out sucked.

Not yours @chadc

haha

Knowing when to kill a project and having the guts to do so allows you to reallocate those resources working on that product to something else that has a chance of actually succeeding.

We all know this but…

Chances are, if you are involving users to validate your solutions before launching them, you have gotten used to modifying your solutions. The earlier you make these changes, the less expensive they are. Changing wireframes is less complex than changing designs, which is less expensive than changing code, which is still considerable less expensive than changing something in production and scrambling to fix it after launch.

it goes even a step deeper with healthcare where we have to try to avoid preventable readmissions, complications, etc – which all have financial impact

So, let’s say this all goes well and everyone loves you and wants to work with you. Let’s say this helps get a seat at the table. Now what???

Don’t celebrate.

Well maybe a few nights of drinking but after that…

You’ve been preaching about iterating and how things can always be made better. And your designer ego isn’t satisfied with just having a seat. We’re trying to change the culture, and in order to do this, you have to keep the seat and continue to provide value.

Now that everyone’s eyes are on you, you can’t become complacent or stagnant.

People talk a lot about failing fast and I hate that.

Never fail.

It’s OK to make mistakes and have setbacks

but you have to learn from them and improve

Inspire this continuous learning and knowledge sharing, not only as individuals but as an organization. Learn from your mistakes. Make each experience an opportunity to learn.

Someone asked earlier about research

Here’s a good question

How do you know how users feel about your product?

NPS scores, User testing ?

ASK THEM

How do we know what technical limitations you have on a product?

ASK THE TECHNICAL TEAM

They tell you.

Bring together users & stakeholders from all competencies to work towards a common goal. This isn’t going to happen by itself.

Exactly @cystinosis

Expose the disconnect in how the business and users interpret each other’s goals. Identify the gap between the two and build solutions together to bridge that gap.

But how do we really start to make the cultural shift?

It’s another little trick.

Your have your process and the business has theirs. While each process may work well independently, when you integrate your process into theirs, it becomes part of THE process.

Like I mentioned earlier, people get tired of hearing designers preaching about design. You need allies and advocates who aren’t designers.

When you’ve convinced stakeholders in mfg, eng, mktg & the lab about how integral you are to success and innovation, and you have executives sharing your work, you truly have influence.

When things like quality and function become expected results and the organizational lens focuses more on the emotional connections and social impact of your product you are redefining holistic values.

So what does it success look like?

Keep in mind, even for us, this is a process and we are making progress. We know what our end goal is, but we are still striving to get there.

For us, our team has grown.

When I get lunch now, I feel like I have a squad of people.

Our process is more defined, we have gained experience and have been asked to be on more projects.

What else does a cultural win look like?

Leadership advocates for UX design at the highest of levels, while everyone on the tactical level understands the importance of trying to provide the best possible experience for your customers and they all want to work with you.

I think being asked to be on more project is a great measure of success for a UX team.

There is no longer a need to translate design language or business acronyms for everyone.
 There is one common vocabulary within the organization. Everyone is on the same page, speaking the same language.

Sharing the same goals

Hell yeah it is @dan_naumann

Projects are being prioritized by the value they provide to the customers, with an emphasis on the overall experience. You are playing a part in that prioritization based upon the value you helped redefine.

At 3M our Chief Design Officer, Eric Quint, often tells us that sometimes you have to take a step back and look at the big picture to really see how far you have come.

So what really is a Culture of Experience?

You’ve help build a team of world-class talent. A team that is empowered. The best team you’ve ever been a part of.

You’ve got an environment that is conducive to being successful. A state of the art space that enables you to do your best work, and all of the tools you need to do so.

You are working on projects you’ve always dreamed of working on and you are making a difference. You are improving lives every day.

You’ve influenced the organization to focus on creating the best possible experience for all of your customers across all touchpoints.

But culture is funny.

You can’t force it or it won’t work. You have to cultivate it.

If you put in the work, build the relationships, foster collaboration

and fight together for the greater good

the culture piece will just come together organically

Lead by example. Strive to make the tough decisions. Have the uncomfortable conversations.

Influence others by your work ethic

and most importantly by being accountable

Accountability is contagious.

That’s the environment I want to work in.

Help the business solve their customers problems and bring value to those who we want to use our products

We’re almost out of time but I want to get to everyones questions

so if I haven’t answered anything please ask again

and if you want to ask more questions later, I’m happy to answer them in a blog post

It looks like we might have covered everything off for now!

Nice!

What an awesome session Andy

Thanks for sharing Andy

Thanks again for giving us your time and thoughts today – super valuable and much appreciated

Excellent insights, Andy. Thanks very much for sharing with the UX Mastery community. :slightly_smiling_face:

A reminder that if you have follow-ups you can post here and I’ll follow up with Andy http://community.uxmastery.com/t/designing-a-culture-of-experience/3025/1

The serious stuff first:
Dr. Paul Sherman has worked in user experience since the days of dial-up.

He conducts user research and user experience design for mobile, web and desktop in many domains, including accounting; banking; e-commerce; financial planning and portfolio management; healthcare; mobile gaming; mobile device hardware and software; network, server and cloud application security; tax preparation; and travel, among others.

He also creates and teaches graduate courses in user experience research and design at Kent State University, where is he is an Assistant Professor and Program Coordinator for the User Experience Design Master’s program.

@pjsherman Do you want to give us a bit of an intro to the topic?

Sure thing. Basically,

I would define onboarding as “The process of getting people to adopt your application or service.”

It might be fully on-screen, but more likely it’s a blend of channels and modalities.

How is onboarding process different from marketing and sales?

Onboarding is post-conversion. Post-sale. It’s the stuff that happens after the customer/user has committed.

*Or, committed an initial effort at least. Onboarding UX is the process of getting them from initial commitment to adoption.

(I am thanking my 9th grade typing teacher right now for teaching me home row)

:+1:

And questions are go…

I’m currently designing a basic onboarding process – and I’m wondering how to know what is best to include and not include. Obviously you don’t want to have to tell them every feature of the app – but how do you decide?

woo!

Paul, when does onboarding start and end (does it still apply to customers who have been with you for awhile but aren’t necessarily utilising your product well?)

@holliedoar TESTING! And also focusing on the critical few things that solve a user’s problem.

What have you found users respond to the most? re: copy, images, animations

@zheng Actually I think it can continue almost indefinitely, and it definitely applies to new releases and features. If any of you use Trello…have

you seen Taco pop his cute little head up to announce new features?

That’s continuing onboarding.

love taco

By the way that’s a good pattern. Giant release notes belong in the dustbin of UX history.

are we talking onboarding features or process?

Taco is just like a modern day paperclip guy

OK I think I have the rhythm down… go to @hawk DM with me, copy the entire q, then answer.

@ryanhoffman : What have you found users respond to the most?
I can tell you what they DON’T respond to.. something like this:

@pjsherman uploaded a file: Click to see image

That’s “modal stepped tutorial”

Not many advantages.

People won’t remember.

And they just want to try the thing.

Asking people to remember things in advance :-1:

More on @ryanhoffman q: they respond to interaction, and learning while doing.

Ugh, as a user I hate those things

What do you think about the one that “locks out” everything but the feature that the onboarding step is explaining? Do you think of those as some sort of modal as well?