IA Summit 2009 – Donna Spencer – Design Games

What are design games?
they are always something that will provide an outcome that can be used in a project

why design games?
because it’s much more fun to have fun at work
by playing a game, we use different parts of our brain. we can stimulate creativity
if you tweak your process to make it more fun you will get much better outcomes

games to play with users

• design the home page — very good for focus groups – just ask people to design a home page that will be perfect for them. it doesn’t have to “look like a home page” (logos, nav bars, etc.) just think about the things that are important and get those down. this will tell you a lot – what is most important to people
• divide the dollar – give the group a set of features (or have them brainstorm a set of features) and then have them divide the money across the features – and justify their decisions.  what if they just divide it equally??
• metadata games — use to illustrate that people call the same thing the very different thing. [ex. what do you call x (image) cozy -- terminology is so damn important... we need to know the terms that people use. ]
• Freelisting — good for terminology – tell me as many of x as you can think of. (e.g. tell me as many dogs you can think of). the ones that she says first are the ones that have meaning for her, dogs they have owned, do own, etc. can get an idea of her mental model. list varieties of beer. very good way to identify cognitive patterns… 
• card sorting — write content ideas on index cards and ask users to sort them in ways that make sense to them. this is a very common technique, but we can make this more “game-like” – prizes, time contstraint, get fun with the content. 

games for design teams

• idea cards — you have some sort of problem that you are working on (design challenge), brainstorming, draw three cards from the deck and come up with a solution that uses those concepts (ex. audience, integrity, travel, difficult, risk, images, personality, impersonal, etc.)
• reversal – rather than tackling the problem head on, reverse the problem and solve that. [ex. going through airport security is awful. how could it be made worse?]

planning design games 

• you definitely have to want an outcome, so you must plan to get that. 
• rules/constraints have to be defined
• have to figure out a way to make sure that everyone is involved. 
• prizes? criteria? how will success be measured – how to communicate that without corrupting the experience
• instructions – make them clear to avoid stress for the players. [ex. identify the things you want to see on the home page, discuss which are most important and why, sketch a home page that represents those ideas this can be pictoral or done in words (show examples to make people realize there is more than one way to do it).] 

More games (suggested by audience):

• design the box - use with clients, product team, lots of folks – we all know how the product works. we all know how boxes work. if your product was a box, design the box. if the agency was a product, design it’s box. now sell that product to the rest of the team.  if i was a product, how would i sell myself?
• brain writing. use sticky notes. every time you have an idea, stick it up to the wall. then sort it. 
• different hats - pretend you are a person from a brand designing your site? (?)
• creativity cards – oblique strategies. posting them on twitter as oblique chirps. 

————————

maadmob.com.au
@maadonna

IA Summit – Naomi Norman – How to Work Within “Government Time”

abstract 

User research  

• must be done early. the earlier te better.
• cuts through deliberation, makes decisions easier
• use large samples

Quantitative methods:

• user surveys
• scored interviews-
• data mining
• card sorting, cluster analysis

Qualitative Methods
-A/B testingdefine your consultation 

• train your client
• user research
• content development
• consultation-your process

learning to love governance

• know your client’s governance
• collaborate to establish it
• know each group’s responsiblity
• target your communication appropriately
• get them to champion your process

documentaiton and presentation

• establish an audit trail – all comments need to be codified, documented and presented. It’s important to recap this, helps to gain consensus

what role do we play?

• faciliatate engagement
• give people a voice
• raise awareness of issues

….

Our goal is to transform government services for the benefit of our users.

ROI metrics for gov’t must tie back into their mission. But to be successful, we must understand their are their desired outcomes. 

• reduce fone calls?
• increase traffic to a particular area of the site?

The Government loves consensus and committee based design. Some tips to help:

• User research will help to illustrate facts of use (not politics of opinions)
• affinity diagramming – aids the group decision making process

Biggest key to getting things done: Find out who needs to sign off on it and how you can get that to happen. 

IA Summit 2009 – A. Hinton – You are (Mostly) Here

[slides are here]

Digital Space & The Context Problemfake news:  They are going to watch you in Vegas.  They will video everything you do. They will record everything you spend money on. They will keep a line-by-line record of everything you do in the city. Then they are going to notify everyone you know….

This is exactly what facebook did (beacon). what was the outcome? giant user revolt!

issues:
-What does the word “friend” mean? In the same way that ”ready-mades” (dada) disrupted notions of what “Art” is. changed the frame of reference for the “object” – changed the way people think about high art, low art and culture, mediated environments have changed our conceptions of relationships and information sharing.the issue is: labels – what we call what we see/do
[Colin Powel's presentation picture of "WMD's" in Iraq relabled as an IHOP] 

the issue is: situational ethics – how we justify our actions within different contexts

—————————————————————————————————————

Everybody Loves Brain Scans:
frontal lobe (logic, reason, most recently evolved area of the brain – spock) vs limbic system (id – capt. kirk). basically we perform cross-benefit analyses between the two on a constant basis. (Capt Kirk vs. Spock) 

• cheap wine vs expensive wine: the brain scan showed different neural reactions to the same wine with different labels. something about it costing more makes it taste better (literally). 

“you all are IA’s, you label things for a living – i’m simplyfing things, but… “It’s all about labels. It’s labels that are important. Context and language are highly symbiotic. They affect each other in many different ways that we don’t entirely understand. 

A map is to help you navigate streets. This neighborhood is made of streets with names and other stuff without names – Boylan heights (research by denis wood, nc state) – can listen to this at This American Life, episode 110. 

What do the maps mean?

Jack-o-lanterns map correlates to newsletter mentions map. what does this mean? No matter who lives there, the houses are mentioned in the newsletter. Do certain houses attract certain people? 

“the territory was there first, and the map came later but the map has a lot of power over how we understand the territory.”

Denis Wood’s  book is about this. 

There are fuzzy boundaries between real and virtual. 

Vertigo - describes the effect of the realization that we are actually in more than one place in the same at the same time.

Context collapse – radical readjustments of context. They’re really screwing up when we think about what it means to be “here”

There is all this “fuzzy human stuff” that we are trying to make into data

“love” 

• go to facebook and say which of the 6 mutually exclusive labels about that love relationship – single, in a  relationship, married, engaged, open relationship,  etc. 

Eats Shoots & Leaves – something as small as a comma can change what we mean in a significant way. 

nook vs stadium

• d vs @ 
• private vs public
• re: vs re: all
• these distinctions are not so obvious in the machine context. 

context shapes identity:

• linkedin – professional me
• facebook – school me
• chemistry – dating me
• etsy – crafty me
• livejournal – private-thoughts me
• myspace – party & music me

crossover of contexts:

• are you friends with your kids on facebook? your boss?

 People are uploading pics from high school and carefully tagging them….”i deserve to have that in the past” 

Our identities are tied to the spaces that we are in.

We are in multiple spaces.

What does that mean about our identies? The self as a multiple, distributed system… a decentered…  – sherry turkle

tweetdeck is not playing by the rules
Changes the rules of how people behave the in the space -  filtered updates, putting people in groups, etc. 

People/Applications/Both are constantly changing the rules. 

Implications are everywhere – news, money, learning, entertainment, family. The context problem exists anywhere or with anything that can be online. Even if people themselves cannot  be online, information about them can be. What are the human limits to comprehending context?

you can look at this, but you can’t do anything about it “in this context” – ex. google earth – holocaust memorial museum crisis in darfur 
 
Language and context shape one another – especially online where everything is language based (literally must be coded to be there)

Information archecture is great at findability, but findability is only part of the value proposision of IA. The act of defining content via findability is designing context. We lack a suitable language for this…

———————————————————–

Q. Issues of changing the rules (e.g. post once, ping many)
-twitter updates on facebook – these are being presented as value-adds, but they are actually taking the value away from context.  They’re not going away, it has always happened (example of using newspapers for not what they were intended for). We are all sort of learning a new literacy. Learning new ways to ingest/filter things out.

Q. Are you positive or negative on the fact that people can see you?
Neither. things are just changing. in some ways it can be positive, in some ways it can be negative. we are going to have to come up with a language about privacy – this kind of privacy or that kind of privacy?

—————————–

recommended/related presentation: luca rosati, andrea resmini - towards a cross-context IA (slideshare)

IA Summit – Dan Brown – Designing (business) Rules

[slides are here]

What is a rule?

We are not talking about “what you should and should not do,” we’re talking about rules that that lead amazon to suggest particular thing to us (of all possible things). If a wireframe tells us what goes on the screen, then a rule tells us how a screen changes in different circumstances.

• Rules drive the user experience, but they are not a single cohesive system.
• Rules make use of language, but they don’t define that language.
• They are not linear, they apply only in specific situations (which meet very specific criteria). they are not a recipe. They change depending upon context.
• Rules help us make choices about what is seen and how we see it. 
• Rules craft one aspect of an experience (mediation between director and actor).

What DB does:

• templates – basic element of a web site
• Navigation/classification – how people find the content
• Content types (structural) – giving us a meaningful language for talking about the range of information that is available 

Rules tie these three things together.  

IA integrates real-world content into the abstract framework they are dealing with. 
rules are what help us (what we use to) do this. 

**Content Rules**
what are the mechanisms that we need establish to put the content into the templates that we design. 

cooksillustrated.com - 
what goes on home page? do i pick from all categories of recipes? do i pick the newest? do i pick the most popular? 

what are the set of guidelines that we use to make these decisions?

**Navigation Rules**
Imagine a room, depending upon the room, it has THESE type of exits.

cnet.com > mp3 players
3 categories – price, manufacturer or Other
these choices allow us to change what we see.
interesting differences – mp3 player 20-50 dollars is the lowest; digital camera less than 100 dollars is the lowest. mp3 lets you see All prices. Digital camera does not. 

ecommerce product page 
available in stock – this is visible when certain criteria is met (rules).

————–
Rules are a criteria for selecting responses to behavior

Rules are not patterns - rules require context.
Patterns are things that require rules to be managed. 

Components
Navigation blocks are components. Single components can have many different rules applied to them.

Sun’s accordian component:  content is not defined, but you can write rules about how the component is to be used in different area of a site.  

How do we structure our rules?
Things to think about:

1. Action – what is the action that is performed – (show the content? hide the content?)
2. Scope – from which content types are you going to draw your content? what is the range of available content
3. Filter – this component will use x scope (e.g. recipes), which ones (e.g. fish)?
4. Quantity – how many
5. Format 
6. Default – What happens if no criteria is met?

——————————-
SCOPE: 
Facets – what can users filter by
Options – how will users select the value that they’d like to sort by?
Range – upper and lower limits. how will these be determined and/or set. for all areas? per product?
Effect – selections and presentation – does the user’s selection alter the interface in real time?
See All – do i let them do this? is there a see all option?

——————————
Types of (wireframe) Annotations:

• Prose rules - e.g. If, then – if you are in x state, show sales tax
• pseudo-code – If then – IF state=x THEN display tax

Table - 
Wireframe content – Lastest articles (label in the wireframe)

Flows – render the rules visually

what makes rules good?

• good rules are user-centered – provide content that is meaningful to the user. 
• good rules ares are unambiguous
• good rules are feasible (use existing IA parameters) – most popular articles — are we collecting information on determining whether articles are popular or not
• good rules have specified responsibility – we need to specify – does the machine determine the content or does an individual dtermine the content (using the same rule, but different entity is responsible for it’s implementation)
• good rules will degratde gracefully – a.k.a DEFAULT – if all else fails…something relevant shows up

Do the right thing 
We need to take our rules very seriously because they govern how our product interacts with its users.  This is a somewhat ethical enterprise.   

Documentation of Rules -

• Do the people actually pay attention to the rules? are the rules implemented?
• Consider making the documentation fun, entertaining & engaging.  This will be helpful to help the rules get implemented.

Consideration: is it better to get a stakeholder on board with enforcing the rules? or do we be prescriptive about it? Consideration: Feasibility of rules:

Whether they are enforceable? how are they to be enforced?  Hand-offs: the spirit of the rule vs. the letter of the rule
As IAs/Designers, we communicate the spirit – but we hand off implementation. 

When we document a rule, we are documenting what the machine or the editor needs to do, but they don’t communicate what the overall user experience is. Wireframes need to include content to help communicate the end result.  IA was defined as the language of the web site and rules make use of the language (?).

Q. Do we create the rules and then engineer the IA from the rules? 

We are designing user experiences and it’s very easy for us to get caught up in the minutiae of what we do, but we need to ground ourselves in the user experience.

Q. Who “owns” the rules set.  the client? the designer? 
Rules have a ripple effect. If it’s unfeasible in some way, I want to know. We want to make them as feasible, as implementable as possible.  Who is really going to consume this deliverable?  We design for the user, but we also need to design for the editor/for the organization to make sure that the rules acutally get implemented. 

Comment from the audience: Bill Scott presented the interesting moments grid as a format for annotating complex interactions – perhaps could be used for interaction rules as well. 

IA Summit 2009 – Dave Cooksey – Taxonomy Validation

About dave:  saturdave ux consulting

Testing and taxonomy – why testing is important 

Taxonomy is an “expert” activity that comes from a priveliged view. Taxonomy is more closely aligned with business/technology goals than user goals.

The facts about taxonomies and metadata

• relationships are implied (as opposed to an ontology which defines relationships).  Users must interpret the relationships.
• A user’s individual experience informs their interpretation. Their experiennce very much differs from ours. This is why we need to test.

Why taxonomy validation is necessary

• justifies the cost of creation
• helps resolve organizational conflict
• vets ideas with real users
• reassures the business that the project focus is on success

Taxonomy validation techniques:

1. delphi card sorting – allows for direct interaction with participants (can ask follow ups and ask people what they mean, to describe their experience, justify their choices)
2. remote card sorting
3. usability testing - focuses on interaction (step-by-step vs. bird’s eye view)
4. search analysis - indirect (no follow-ups, no “what did you mean?”)

Delphi Card Sorting: About

• Qualitative method
• Small sample
• Based on the delphi method
• developed at UB
• Article by Celeste Lyn Paul, A Modified Delphi Approach to a New Card Sorting Methodology

Delphi Card Sorting: How it works

• Heirarchy is laid out in cards (seeded or not)
• Inform the user what the vocab is for 
• Users are instructed to turn over cards that they don’t think are necessary, move cards around, make new cards (add ones for missing concepts). 
• While they work, ask the participants about what they are doing (e.g. Why are you collapsing x and y categories? Why did you turn that one over? etc.)
• Users work in a cumulative fashion (don’t restart the test everytime, let them build on each other’s changes)
• The process generally runs for about 15-20 participants until the heirarchy stabilizes. 

Delphi Card Sorting: Tips

• If possible, get a video recording of the sorting. Video recording is essential
• Take pictures of the table before they start and after they finish (each participant)
• Remember, this exercise is to inform YOU - the end result is the final heirarchy, with notes. (which ones people had problems with, what flip flopped a lot).
• Considerations: What kinds of users are you testing: do they have domain expertise? do you have multiple user types?
• Plan for lots of space. Allow time between participants
• Create a starter sheet to make notes

Remote Card Sorting

• quantitative, large sample
• open or closed (closed – tests structure, open – creates categories, gathers many perspectives)

Remote Card Sorting: How to do it

• recruit via intercept or email (keep these separate, don’t combine the analysis/results)
• determine sample sized needed
• choose software
• focus each sort to 15-20. NOT TOO MANY ITEMS
• put items/categories on cards
• run study
• analyze the data

Online tools for card sorting:

• optimal sort
• mind canvas
• web sort
• cardsort
• socratic card sort
• web cat

Remote Card Sorting: Tips

• run a few qualitative to test, if possible
• pick the most important categories and items (stereotypical problems)
• read the comments
• deliverables will depend on the software used. 

Regular card sorting

• -open or closed: closed – write down your terms and let them organize them. open – let them come up with their terms and then organize them

Usability Testing

• piggy back on system testing
• task based – Where would you go to do this?

Usability Testing: Tips 

• have a clear goal
• keep it simple
• ask follow-ups, probe a little bit.
• illustrate how data is driving the experience.

Search Analysis – Tips

• Try to get as much info as possible about the data. 

IA Summit – Stephen Anderson – The Art & Science of Seductive Design

[slides are here]

example of good design:
iLike

why does it work?

• piques curiousity – will this work?
• visual imagery
• leverages pattern recognition – we like to find patterns.
• relies on recognition over recall – with an empty form field, I have to remember the artists i like. In the iLike system, it’s much lazier this way (for the user), I just have to recognize my likes not remember them.
• uses motivators well - confirmation message includes: play the iLike challenge (trivia questions with a time constraint)

It’s a convergence of user goals and business goals - user has fun, business gets info.

what made this work?

Not IA alone, but psychology, game mechanics, etc. 

Increasing motivation and reducing friction.

Focus on natural human impulses.

What do you know about people? make a list. start there. 

Curiousity

• mystery hot wheels. there all of these knowns and they want the unknown.
• example CPK don’t open it
• netflx – rate your return to reveal what you might like. 
• linked in
• lovely charts

play hard to get

• private beta + social proof
• social motivation: the single most important factor in determining whether or not someone will adopt is whether or not their friennds hang out there.
• constraints can be attractive (e.g. 140 character constraint). 
• sabre town – like yahoo answers, but it tries to connect people. personalized. your question will be sent to people likely to answer and you might receive questions related to areas you have knowledge/experience. constraint – you have to have karma points to participate. they don’t tell you all the ways to get karma (it is NOT explicit – system should not be gamed)

friskiness, playfulness

• dopplr – delighters – matt biddulph – example of a site good for travels – based on the idea of serendipity. where are you going to be and when — shows you missed opportunities so that they don’t get missed. who is giong to be there at the same time? 
• different colors is used in each logo, for each profile, what is going on here? colors are based on cities (somehow). 
• mooplr – promotes dopplr offline
• supplied me with something i didn’t know i needed. the result: now i am a loyal user. 

collect & catalog what you like and ask yourself why you like it?

Then use the deck to ask yourself — how can we use X to improve this design?

Things to think about:

• social psychology/social design
• persuasion, choice, influence (6 principles)
• game mechanics – what works what doesn’t
• brain research, cognitive science.  - fun humor visual  curiousity etc.

Stephen’s site: poetpainter.com

IA Summit 2009 – Mathew Milan – Leading with Insight

[slides are here]

people don’t know what they want until you show it to them – Steve Jobs

recognizing need is the primary component of design – Charles Eames

design works if it’s authentic, inspired and has a clear point of view – ron johnson

What is Insight? 

according to Britanica, insight occurs when people recognize relationships

according to Webster’s -  the act or result of apprehending the inner nature of things or of seeing intuitively

insights – recognizable truth, things that help you understand the real 

Rules to being insightful

• you must ask thought 
• look beyond the obvious
• don’t be afraid to reframe the problem – restate the problem, play with scenarios, perspectives
• you must learn to trust your gut. 

Who’s insightful?

Karl Popper – wizard merlin of insight

Richard Feynman – brilliant, he is the wizard merlin

michael posner – another way to be insightful is the prepared mind. 

Posner’s is the method we want to emulate. the prepared mind. The prepared mind is a repeatable process. There is no magic to this:

• perpare
• incubate, 
• illuminate 
• verify

Columbo taught me everthing i need to know to be insightful

columbo is:

• treating inside as a consequense rather than a cause in problem soflveing
• understanding the underlying structure of the problema nd resturcturing it to avoid functional fixedness
• applying procedural similarity to takek advantage of representnative transfer   
• waiting patiently for suddeness of solution. 
• validation of inquiry

Moral: Don’t use research to tell you what you already know.

Use insight to uncover the true nature of the problem. 

Real Insights are memes: Immediately Understood, Self-replicating, Seeds of Transformation. 

Good Insight helps you to know your customer/user (and their needs) before they know themselves.

IA Summit – Erin Malone & Christian Crumlish – Designing Social Interfaces

Principles, Best Practices and Patterns for Designing for the Social Web

[slides are here]

Social Design patterns?

• patterns as in alexander
• design patterns as in gang of four and Tidwell
• Social Design as in eGroups, facebook, flickr, twitter
• Also so high-order principle and emerging practices
• designng these interfaces is a holistic exercise that extends from teh data architecture to the presentation layer.

Frankly, you can’t just put a social interface on the top of something and “get social”

Everyone is/has been looking at the social realm on the web for the past 5 years or so. Lots of folks have a lot of understanding, but they are very scattered. We need to get a way for the different understandings together to get some ideas. 

social pattern language 

• emergent interaction patterns
• components and pieces are the building blockds
• support the entire lifecycle of the social experience
• buildinga  vocabulary and language for social application design…

current landscape

• self is in the center – person, with a sense of identity, who wants to engage. 
• right side, participation, then methods of organizing 
• left side – collaboration and relationships. 

Some Typical Scenarios: 

Boss/client: Help!Ii need something social on my site!

Some Issues – help people want to be social with you

• talk like a person - use the language of contemporary speech (NOT jargon), ask yourself if that’s really how you talk. read any copy out loud and strike out anything that feels awkward.
• sign in/sign up – user wants to access personalized information. folks need to get a value from this. also, you need to figure out the right time to present the sign in so it’s not an obstruction OR an after thought

Some Issues: Getting folks to interact with each other 

• activity around ojects – share this – How? enable people to share. provide minimal interface needed to facilitate rapid sending or posting such as a ubiquitous tool. 
• contact cards – need to provide identity. 
• add friends – reciprical? assymetric? how? 
• circles of connection - user wants to indicate nuances in their relationships with other people to created contexts for communication. use when attempting to distinguish levels of participation in a person’s network. use to disambiguate “real life” as opposed to online relationships.V
• peer to peer rewards - let people rate other people’s content. 
• nudging - low effort way to interact, relate to people. 
• followers badge – this makes people feel like they have company, that they are not alone

Some Issues – misbehavin’

• Norms – a principle of community is to establish and communicate norms -founders and community mangers can play the role of model citizens. demonstrate the good behavior, recognize those who behave well.
• Reputation – leaderboards – use when? if the community is competitive and the activity users engage in are competitive in nature. enables user-to-user comparison. DON’T USE THIS PATTERN WHEN THE ACTIVITIES THAT USERS ENGAGE IN ARE NOT COMPETITIVE IN NATURE (EX. RECIPE SHARING, FOTO SHARING, ETC.). 
• Reputation - report abuse feature – provide a consistent affordance for reporting abuse, attach it to each piece of granular content, immediately hide the offensive content, etc. 
• Reputation – Labels for users

“Be civil to all; sociable to many; familiar with few; friend to one; enemy to none.” -ben franklin

For more on these topics and this research, get involved at http://www.designingsocialinterfaces.com/patterns.wiki 

IA Summit – Aaron Louie & Rachel Elkington – Darwin Does Design

[slides are here]

example of unguided evolution - The Peppered moth – naturally adapted to changing conditions in order to increase their survival rates. 

example of guided evolution – dogs -we want/wanted them to perform a particular function (hunting, herding, pets), so we breed them accordingly. We have outcomes that we want to acheive and so we design to acheive the outcomes…but different people have different outcomes, so how do we measure success?

Design is not art. Design is art that solves a problem. 

(IA/IxD/UX types) - Why do we design? To make happy users, To make money.  Our clients have problems and we figure out what the problem is and then we create a design that solves the problem. This doesn’t always work – why not?Sometimes clients and users ask for things they don’t need and we give it them. Sometimes we design short-term solutions for long term problems.

metaphor of the cannon shooting at a target. 

discipline-driven design — aiming the cannon is strategy and concept. then you pass it on to the designers and the programmers and then you measure to see if you hit the target. This would be a great strategy except it doesn’t account for unforeseen obstacles. 

What we need is an intelligent cannon that can guide the cannon ball and an intelligent cannonball that can move itself around obstacles. This is called performance-driven design. 

performance-driven design:  guides the design over time.  performance-driven design is development that accounts for variants. 

performance implies an audience:

• users
• clients
• stakeholders
• project managers
• programmers

They all have different measures for successful design – usability, flexibility, cost, etc.

We have way too much date.

“data is cheap, insight is expensive.” – Karen McGrane

UX + data =  insight

We need to collaborate to get insights. We need to collaborate on our measurements. 

Key Principles

• strategize – goals analysis, card-sorting, RITE prototype testing (rapid, iterative, testing and evaluation), multi-variate testing, A/B testing –this happens in the order defined above; the magnitute of change to the interface changes diminishes over time (as it gets closer to launch) slide.  
• monetization – how much is x outcome worth to you. 
• generate design variants
• test
• repeat

Data-enhanced personas + persona-informed segments become behavioral models for your users.  You can then target content. 

——————————

HOW DOES THIS WORK IN THE REAL WORLD?

• look at analytics deep dive – look at click through rates, 
• look into monetization model and site performance
• map personas to behavioral segments
• balance your objectives
• figure out what kind of test you can actually run 
• prioritize your variants. 

The expert review

• what + why = insight
• word doc that presents: these are the variants that performed better and why. 
• to what extent can we generalize site-wide as heuristics.  is this purely contextual? or is this a better design?

case study – online hardware vendor

• need – wanted to bundle their product and support
• analytics found the problem – sales dropped off
• ux designed the variants
• ran an A/B test
• usability study on confusing results.  users were confused by more than just that page (where we originally thought the problem was)

Aaron’s post on this session. 

IA Summit 2009 – Russ Unger – Heuristic Analysis for the Pitch Process

[slides are here]

What is heuristic analysis?

Heuristic Analysis is a technique to evaluate design. It’s a design review of site or competing sites using heuristics. Heuristics are techniques that reliabley lead to a desired result – the problem of that is that is always changing. 

“shit that worked in the past that may (or may not) work in the future “- mathew milan

Heuristic (/hjuːˈrɪs.tɪk/) is an adjective for methods that help in problem solving, in turn leading to learning and discovery. These methods in most cases employ experimentation and trial-and-error techniques. A heuristic method is particularly used to rapidly come to a solution that is reasonably close to the best possible answer, or ‘optimal solution’. Heuristics are “rules of thumb“, educated guesses, intuitive judgments or simply common sense. Heuristics as a noun is another name for heuristic methods.   source: Wikipedia

Basically, if you’ve got an opinion, you should be able to defend it intelligently.

Let’s get some heuristics!

Get them from “experts”

• jakob nielson
• Bruce Tognazzini 

or you can “roll your own”

• can you find what you’re supposed to see? why/why not
• when you click, did you get what you expected? why/why not

Which to use? Well, as always, it depends.

Remember: Heuristic analysis doesn’t involve actual users.  It should not replace user testing, but it’s a good way to get started… 

how do i do it? 

choose the heuristics you are going to use, and then walk through the site and identify where the heuristics are used/not used.

• keep it manageable – 8-12 heuristics. (a “digestable” amount)
• identify your key observations - 
• assign an impact factor – e.g. mild, severe
• make recommendations 

know your users: choose your heuristics. 

• get participants – entice your guests, give them food. do it in 2 – 4 hours. get folks from different areas of the agency or community (with different interests/skills if possible).
• identify the key areas for attack (e.g. product pages, help pages, etc.)
• keep it simple. give the participants scenarios to work in. (ex. “Jane wants to buy computer her first laptop…”)
• review the current state of the web site – identify problem areas. (e.g. too much text, not enough visual distinction, etc.)
• compare it to other comprable sites in that industry/sector (ex. for a business (HP) – zappos).  Ask the participants, “what sites/pages do you like?”  etsy, gap, apple, etc. 
• Likes from Zappos: short url, grey for content
• Likes from Apple – pithy txt. huge product image – don’t bore you with the specs and details unless you want them
• Likes from Etsy: large product shot, add to cart in two places, 
• Likes from Gap: not a lot of words, real-time inventory, detail view, 
• compile the feedback… write it all up and create the document — in adobe illustrator
• userglue.com/heuristic.zip — AI files
• will show the screen that you reviewed, talk about the problems, explain why. give recommendations – simplify content, display larger fotos, easy to use urls, provide cart/purchase below the fold. 

“Just an IA” – overcoming the stereotypes of sitemaps and wireframes

• Heuristic analysis gets you out of that. It illustrates that we research, we think, we synthesize a lot of things. 
• Sometimes site owners just need an outsider’s opinion to give them a spark. Give them the outsiders view. 

Q. How much of a report do you do? 

A. Brief. Don’t spend hours onthis just hit the highlights to get basic points across.  If you have a whole set of points that you evaluated, Identify areas you”ve noticed as “areas of concern”, areas we can help with, 

Q. How do you present your analysis without offending them?

A. Do not be attacking, be a general overview, take the easy pickings. This is a loose critique.

Q. Do you always use participants? 

A. YES.

Q. a little bit more on details… do you end up with a slide for each heuristic that you use? do you show the things that they have done right? or only the things that they’ve done wrong? 

A. Yes, create a slide for each heuristic. We did 12, but we showed 4 or 5 before they glassed over. 

Q. Heuristic label? do you label each observation? (didn’t get the answer)

Q. What is the next step after this? what if HP says, “our users like this” — then it’s probably not the job for you

———-

Takeaway:  This is low-hanging fruit.  This is hit and run.  This is as simple as it is. This just shows how smart you are; Imagine if they let you do this!

——–

Note: NEVER present designs in pitching… how can you define your design before you’ve done all your work!?!?

——–