If an app launches in the app store, and no one hears it, did it really launch?
Today, digital products like apps and sites require marketing. Luckily, this has become easier to do, even a limited budget.
If an app launches in the app store, and no one hears it, did it really launch?
Today, digital products like apps and sites require marketing. Luckily, this has become easier to do, even a limited budget.
From start-ups to banks, design has never been more central to business. Yet at conference after conference, I meet designers at firms talking about their struggle for influence. Why is that fabled “seat at the table” so hard to find, and how can designers get a chair?
Designers yearn for a world where companies depend on their ideas but usually work in a world where design is just one voice. In-house designers often have to advocate for design priorities versus new features or technical change. Agency designers can create great visions that fail to be executed. Is design just a service, or can designers* lead?
*Meaning anyone who provides the vision for a product, whether it be in code, wireframes, comps, prototypes, or cocktail napkins.
Bold claims have been made about applying “big data” to solve the world’s problems, from health (Fitbit) to saving energy (Nest). Data is all around us, appearing in slick devices and colorful dashboards, yet focusing on the technology can cause us to miss the people who have to use it.
Our job as designers is to communicate information. A clean design with big numbers and charts looks good, but how can we make sure people actually understand the data?
The most visible way people are now bringing data into their lives is with fitness trackers such as Fitbit, Vivofit, Jawbone Up, and many more. Over the last year, I’ve been talking to people about how they use data in their lives. This article is not about the merits of any one device but a compilation of interviews about how real people use data, merged into personas, with some ideas for design.
“Ironically, devices that initially helped foster engagement in fitness sometimes became too naïve to support increasingly sophisticated fitness priorities.”
Ken is a middle-aged guy who needed to get in shape. He heard about the Fitbit, and being a gadget guy thought it “was a good excuse to play around,” but he is starting to feel the limits. For good health, you need to get your heart rate up occasionally; not all steps count the same. “This might be a tool more for getting started. I’m training for a 10k and ‘steps’ isn’t a useful way of looking at it. I’m checking out GPS running watches now, but don’t tell my Fitbit.”
Ken’s is a common story; a study noted that “ironically, devices that initially helped foster engagement in fitness sometimes became too naïve to support increasingly sophisticated fitness priorities.”
It is important to realize that the tools themselves are not going to get people to exercise, but they can inspire people to try. For new users, extreme simplicity is a requirement, but this very simplicity becomes limiting as the user expectations increase. At each stage, the tool needs to give back more value than it asks, to address more complex questions as users’ experience and competency increases. For example, colorful charts can show progress and sustain interest for a while, but inevitably we all plateau. The experience needs to develop to address day 100 as well as day 1.
Susan came at quantifying herself from a different angle–sleeping. One of the promised benefits of the devices is to track and understand your sleeping patterns and ideally wake you up at the most auspicious time.
In the end, her Jawbone Up didn’t have a big impact on her life. “I didn’t really use the data–I still drink coffee too late in the evening and read the web in bed, but there is a nice feeling from seeing the chart. It has certainly made me think about getting more sleep.”
That’s probably the best benefit she could get. A blog post in Nature cites research that shows “the stage of sleep we’ve been woken from … does not actually have an impact on cognitive performance,” but “sleep is arguably one of the most (if not the most) important health factor that can be altered.”
One insight from Susan’s story is that many users are not looking for deep statistical analysis. Susan was just curious.
If there is no goal, it can be very hard to offer more than tracking and generic advice. Like the Cheshire Cat said “[If you don’t care where you end up], it doesn’t matter which direction you go.” Sleep cycle info isn’t something Susan can use to control her sleep.
In contrast, designing to a specific issue is much easier. If you read enough health stories, the climax is usually an insight like “My heart attack was a wake up call.” One person who suffered from Crohn’s disease wrote: “I know/knew that alcohol can be bad for anything in the bowels, but it was when I compared pain/flare-days to lite alcohol consumption … I saw a pattern… having the data made it plain as day.”
Achieving that insight is still a human process. Each one of us has to recognize there is a problem before trying to fix it.
A good design should communicate the resolution of information, that measurements are imperfect and the person should focus on the high level changes.
Dan is a former fitness tracker user, but has fallen off. “The biggest issue for me was that the numbers didn’t add up.” “When I saw I was getting credit for steps when lifting my fork, I had to laugh.” He is not alone. The study Phases of Accuracy Diagnosis (PDF) noted most users “learned to regard the data provided by the Fitbit as highly suspect.”
Bad or inconsistent data is a de-motivator. This is important because, as the Phases article concludes, “users are inclined to abandon their use of the Fitbit because: they don’t trust its accuracy; they have trouble understanding what state it is in; and they are unable to make the Fitbit more accurate as they become expert users.”
Data uncertainty raises an ethical question for designers. We have a responsibility to tell the whole truth. Bold claims about improving health made in marketing need to be honestly built into an app.
Good design with a simple number may hide the messy fact that the number may not mean much at all.
In all the user research I’ve observed, people are always asking for simplicity. Our challenge as designers is that a simple answer is almost always incomplete. The calories burned each mile varies tremendously from person to person, from flat ground to hills, from running to walking. It feels good to design a screen around a single, simple number, but it just isn’t, you know, true. A good design should communicate the resolution of information, that measurements are imperfect and the person should focus on the high level changes.
As Wired Magazine noted, because many of the algorithms are hidden and proprietary, it is hard to know how accurate they are. Comparing them against each other suggests that measuring calories burned via an app is not an exact science.
The Withings scale Health Mate app is designed to coach people to achieve their weight loss goals. Withings has the data. Does the product work, or not? Only they know.
Grice’s Maxim of Quality recommends you don’t say what you believe to be false (don’t lie) but also don’t say things you lack adequate evidence for (be completely truthful). Hiding complexity under bold graphics makes for a beautifully designed chart, but it would be more honest to share the uncertainty behind these numbers.
Once you start to ask questions about health data, you learn that it is often tentative and is revised with new research. This is hard for people to make use of, especially when communicated through fear–and we have to design for this.
At your next family gathering, listen to people talk about their health. Cholesterol levels, blood pressure, diabetes, HDL, LDL? Which one is good now? Can we eat eggs again? Oh, those doctors, why don’t they make up their minds?! Understanding health is a design problem, just as a GUI is easier to use than a command line interface.
At the heart of many of these devices is a promise that design can make the complex understandable. An article on wearable medical devices noted the biggest challenge: They need to actually work before they can be used for medicine.
Life recently reminded me that more data does not always give better answers. I do quite a bit of running and wanted to quantify it by using a heart rate monitor. During my runs, I noticed my heart rate was above normal. Way, way above normal. My doctor had me wear a high end EKG system for a few days. After a few weeks of fear, the answer came back: “Yeah, some people’s hearts just beat faster.”
After all this digital measurement, it was a shock to be reminded that all those numbers are just rough guidelines. I learned that doctors care more at how quickly it descends after exercise.
In fact, my heart showed “J point notching of the ST elevation symptomatic of athletic people.” No average person is going to understand this, nor should they be making their own medical diagnosis. Even EKGs routinely get mis-read by professionals; here’s a cheat sheet to distinguish people at risk of a heart attack from the athletic.
Designers have a responsibility to consider whether regular people can plausibly use a tool, and when the subject matter is best for professionals. Just putting an EKG on a watch does not answer the question “Am I healthy?”
One method is to look at the context of use. Specialists know how to use or to ignore a particular number, but the average person rarely has the training to know if a particular reading is good or bad for them personally. If the information is meant to be understood relatively, over time, perhaps avoid specific numbers; use unlabeled charts or icons. For example, the Fitbit Flex uses 5 LEDs to show progress. A color LCD implies greater precision, whether or not the data is any more precise.
In contrast, using interesting design and high resolution of information risks both overwhelming the user and losing the marathon by focusing on the steps.
Alternatively, capturing and reporting info may not be enough for good design. Knowing that tracking alone does not motivate most people, consider whether the experience could be built around a plan to get from A to B. This lets people select the challenge they are looking for and sorts them into groups that can get better advice.
As an example, the new Apple iPhone Health app has a staggering number of inputs to choose from. What it lacks is a meaningful way for the person to get value from it. The inputs are dutifully displayed on the “dashboard”, sans context, guidance, or help. The dashboard chart even clips the Y axis to make it look prettier, at the cost of making a one pound drop look like a big deal (“… not that there’s anything wrong with that!”)
A design that only works for a new user is not a very good design.
Mark is an older guy who wanted to be more active. He was a passionate step counter for the first few months but has fallen off. He echoed a comment in a recent study:
“I liked it a lot in the beginning and I think now I’ve sort of fallen out of love with it. You know it’s like you’ve been married for a long time and… it’s all right but the excitement has gone.”
There are a lot of people like Mark out there–the drop off rate of these products is said to be as high as 1/3 in the first 6 months. The biggest problem is simply novelty fatigue. People love new experiences but become bored or distracted. Gyms know this–they would never have enough space if all the members actually used their memberships.
Mark noted that he was excited to hit his steps but began to feel judged when he didn’t hit them, especially on rainy days. Taking long walks on vacations set high bars that he could never reach on a work day. Tracking also has a perverse emotional effect. “If I forget my Fitbit, I wonder ‘why bother walking if I don’t get the credit,'” Mark said.
Roger S. Gil says “monitoring progress can elicit more anxiety since it makes the inevitable ‘plateau’ periods … much more noticeable.”
After the initial honeymoon period where someone plays with a device, it needs to quickly become a tool to achieve a personal goal, or it gets put in the closet. No device is going to make someone exercise, but a design that only works for a new user is not a very good design.
A good example of going beyond tracking is RunKeeper’s Goal Coach feature. It lets you set a challenge to yourself, offers a plan, and shows how you measure up. This is not for the faint of heart. A challenge for the Marks of the world is that most people lack a specific quantifiable fitness goal in their lives. ‘Being healthy’ is a lifestyle, not a 10 week goal.
Designing for this situation requires the tool to avoid judgments, celebrating progress at each turn.
No one tool is going to solve everyone’s problem. Different stages, from novice to professional, require radically different interfaces with the appropriate amount of challenge and responses. Taking a comprehensive lifestyle approach can help–for example asking people to plan when they want to exercise–to make it a part of their weekly schedule. For some, sharing with friends helps stay focused, but many feel more anxious about being judged.
Simple measurements have a lurking danger known as Goodhart’s law: “When a measure becomes a target, it ceases to be a good measure.” That is, focusing around a single number causes people to work to improve it, even if the larger picture doesn’t change. For example, measuring “steps” may have the side effect of justifying eating more.
In all of these cases, design is clearly an essential tool for addressing the problem of our age: personalized data. It’s up to us to make it work FOR the people whose lives are being quantified.
The first article discussed the pros and cons of different UX team structures. For companies that depend on user experience for business success, a strong internal team is essential. But how do you get there from here? Having built one UX group from scratch and managed another 230+ person internal UX groups, I’ve learned a few tips, often the hard way, that can help.
The hardest part of building an in-house design agency is answering the basic “Why?”
I’ve been asked “why” by senior executives, database admins, and the mailman. It took a long time to recognize this question in its many forms, each with its own answer. But fundamentally, they all ask why should UX be a part of the conversation and how can it help them.
In any large enterprise, user experience can still be a new concept. I’ve made the mistake many times of assuming too much–assuming that help was wanted, or even needed; assuming that people understood the terms I used, like ‘deliverable’; assuming that everyone bought into the value of design in general, or on this specific project.
Take the time to meet with people across the firm to explain the services you offer and how you can help them. It is just like starting a business. Discover their issues and the language they use to describe them. Don’t be a salesperson–only offer user experience if it solves a problem they actually have. Once I understood that a UX agency is there to help other people succeed, life got a lot easier.
It helps to understand what the people you are working with mean by success. I’ve had the pleasure to work with many entrepreneurial leaders at a number of firms. They can be fantastic partners who drive real change, but their needs are very different from a product team. Senior managers are typically more concerned with defining the overall vision before building the whole project. User research can validate the concept; concept designs can help communicate their vision. Hitting a fledgling project with the style guide is a great way to not help.
On the other hand, product managers and developers are more concerned with execution. They have deadlines and launch windows. It’s helped me to remember that there is always a next release; a timely good design beats a wonderful design that never launches. Wireframes help the team agree on what they are building; usability testing often helps make difficult tradeoffs.
Once you have a shared vision, it is on to executing design. To become that trusted partner, there is no substitute for demonstrated competence. Until you earn the name as an expert, you are seen as just another person with an opinion. A sales pitch can open the door, but a UX group needs clear product successes.
Establishing a good reputation by helping other leaders succeed will lead to natural growth. The goal is not to increase the headcount of the UX department, but to serve the firm; growth is an effect of helping others solve their problems. Success will feed on itself, enabling you to manage user experience professionals across the organization.
Every UX leader has learned the hard way that one of the most critical skills is setting and managing client expectations. Be clear about what a UX professional will do, how long it will take, and what delays could happen mid-project. Assume that clients are not aware of the user centered design process. It helps to explain the standard procedures and deliverables, not unlike a menu. Show examples of previous work. Our team made a template for every deliverable with a few sentences explaining what, for example, a wireframe was. Back when I worked at an agency, we used to joke when a client looked at wireframes and asked why their website was all black. Now I know it was our responsibility to answer that question before it got asked.
The biggest barrier I’ve seen to using UX in a firm is often simple lack of knowledge of what UX can deliver.
Spread the word about user experience horizontally across the firm by offering free UX “favors.” Two hour heuristic review meetings or small design projects are cheap and demonstrate value. Clearly define how much time you or your team can devote to it, so no one expects a full project.
Clients may come with projects that are about to launch. Giving a little help now will encourage them to plan you into a future project. I was once literally asked if UX could “put lipstick on this pig.” No UX person wants to burn out trying to patch fundamentally broken products, but the relationship can be worth the investment. That product manager came back to our group earlier the next time, and we did it right.
Running the in-house agency is like running a small design agency. You have to deliver value for your customers to succeed. Credibility is the most important quality of a successful designer. The team has to do good work, every time. There is often no requirement to use design (“Can’t a developer just do that?”), but a good designer makes people want to work with them, even if it costs more. External agencies can walk away from a client with little risk that anyone will hear about a failed project, but companies are very social. Good (and bad) work will be remembered and passed on.
Not every project is appropriate for the in-house agency, and a smart group should not overload themselves by taking on every project nor risk ruining their reputation by taking on projects poorly suited to their team, like trying to do marketing with a product design team. Big, temporary projects or isolated product areas in which the team has no experience are good cases for bringing in “the Hessians.” There are other ways to help, including sourcing and pre-qualifying external agencies and individual contractors.
Design contracts have details that are not understood by most procurement groups. It helps to know what is expected and standard in a design project, such as whether personas are required, or if the firm already has a set defined. An internal agency can assist with writing the contract, such as negotiating billing rates, or checking that the estimated hourly rate and the project length makes sense. Once contracted, the internal team can get the agency set up and be effective, faster.
Structuring the team as a consultancy can be a natural step as many UX professionals have agency experience; the difficult part is establishing the practice internally. Organizations that recognize the value of user experience typically have an easier time, but even if the company culture supports it, a team’s credibility needs to be built one project at a time.
A good balance is to establish an “agency of record” relationship, where you partner with (ideally two) good agencies. Agreeing on a defined level of resources each month for a year builds a relationship of trust, which gets the best talent and enables lower rates. Maintaining a 70:30 ratio of employees to contractors offers a good blend of lower costs and ability to vary staff in case of a downturn.
Some firms still worry that this whole UX thing will blow over, and they’ll be stuck with a bunch of latte-drinking oddballs on payroll. Being able to grow on demand and shrink if necessary calms this fear and shows organizational maturity in a way firms understand.
My biggest passion at work is helping each person achieve their goals and how this manifests in our team culture. This has helped guide my decisions from the large to the most mundane. You would be surprised at the impact getting a fancy coffee machine has over a plain corporate coffee pot. It is one small way to communicate respect. UX people are like many other craftspeople; they are 10 times more effective when inspired and engaged. Typically, UX groups work best physically sitting together while spending a lot of time with their clients, but the team should be organized to fit best with the business. Organizing UX people or teams to cover a business area in the firm enables them to develop expertise (relationships, processes, tools, and terminology) and carry it from project to project.
Managing multiple products avoids the tedium of working on solely one product, but enables the team to build a reputation and good working relationships, leading to greater influence. The longer-term engagement enables them to focus deeper on workflow and have a strategic point of view. It opens the potential to suggest that UX could deliver more value by doing more work on project X funded by project Y. Ideally, allocate “10% time,” where team members can work on fixing problems or developing new ideas.
Inevitably, cost cutting concerns raise the question of off-shoring UX. Why pay North American wages when there are people willing to do the job for a third the cost? Many large outsourcing firms have a design or user experience offering, why not use them when when the developers may already be off-shored?
I’ve been unable to hire at the same skill level with off shoring companies, but the real challenge is simple project management. Design resources are active during the formative phase of a project, when clear communication is most needed and requirements are in flux. Waiting 24-48 hours to learn if the request was understood is an order of magnitude slower (and thereby more expensive) than a head shake during a meeting.
When Diana Vreeland said “Pink is the navy blue of India,” she wasn’t thinking web design, but a user experience is often defined by shared cultural norms. Good design takes into account the intangible essentials. The best designers are plugged into the cultural currents and apply them to the job at hand. Amazon lets people tweet their product purchases. Is is appropriate for pharmacy orders? How much visual priority should news be given on a page? Many failed projects could have been fixed by asking basic questions such as “Do people really want that?”
Many firms understand that Agile development is difficult with a team in multiple physical locations, not to mention time zone and language or cultural challenges. Outsourcing works best in a waterfall process with tightly documented deliverables and less dependency on communication. Unfortunately, design operates in an agile mode at all times. In a knowledge worker field like design, it is not enough to have one senior “thinker” and 10 “doers.” The thinking IS the doing.
The best way to integrate off-shore talent is to supplement a team, with a local lead who can break design problems up across a team and coordinate efforts. A good example would be to extend a design idea across a defined workflow, or develop a set of icons. Most outsourcing firms recommend this structure for developers. There are many projects that are simply extensions of previous work. If you have tight standards and quality control, this model can work well.
Ultimately, though, the more important UX is to a new project, the less successful outsourcing is likely to be. One-third as expensive costs more if it takes three times as long.
If I had a dollar for every time I’ve been asked, “We’d love to have UX, but how do you pay for it?” Funding an internal practice is inevitably the hardest problem, but it is how a UX manager earns their stripes.
There are two main ways to fund a design group–centrally, by some overarching part of the firm, or by the various projects that the team will support. Central funding has its benefits–you don’t have to worry about justifying the cost of design on a project or the headcount with many stakeholders. It is often easier to start a group with the support of senior executives who may be concerned about the customer experience across products.
I’ve come to prefer a hybrid model in order to build UX deeper into the company. Central funding creates a competition for the “free” resources and creates a perverse value on the service–that is, $0.00. This is important, because people value what they pay. Everyone knows development and QA are significant costs. If UX does not cost, there is no need to plan for it during yearly budgeting, which means no money for the team.
Additionally, it can be hard to justify hiring a person in the central group even if another group is willing to fund the person. Bureaucratic delays can make UX integration across the firm much harder. “Free” but unavailable is also hard to take seriously. Central funding is definitely needed for centralized tasks like creating style guides and exploring new design ideas that would not be supported by any one project.
The ideal is to have an understanding with finance that the group will be housed in one location but have the actual funding for the people distributed across projects around the firm. You are looking for something like insurance, not actual dollars, from the central funder. In time, they will see how in-demand UX people are. The 70:30 split we discussed earlier helps here as well.
A good team who knows its company still faces the significant challenges. Team member burnout is a real problem. Working on the same problem area for years causes fatigue and sloppiness; one solution is to plan to rotate team members from area to area. Often, this needs to happen before there is a glaring problem, like a project delayed or someone quits.
It is human nature to put off a team change when there is work to be done. Unfortunately, there is always work to be done. My experience has been that the key is that no client likes to replace a known resource with an unknown, even if they are stronger or more well suited. A solution is to plan ahead and let them get to know the replacement well in advance. Ideally, bring on the replacement to assist for a few months. My motto with clients is “No surprises.”
UX projects can have the reputation of being expensive, due to additional team members, and the additional thought put into them. This can be a poor fit when the need is defining the basic problem and sketching a solution. Offering an “innovation” UX , in contrast to regular projects can be a powerful tool to get UX in at the conception of an idea.
An “innovation” project starts off with a three to five week boot camp to develop the product vision from an elevator pitch to testable prototype or a presentation to request funding. These projects tend to be a lot of fun as well!
Keeping fresh and staying connected with UX, design, and technical developments outside the company can be a challenge. Many companies block access to social media tools and design websites. These rules have the (unpopular, but real) benefit of keeping people focused (it can be amusing to read mid-day tweets by consultants who are working “full time” on your project), but also blocks out many design-focused sites.
A team that shares links has a healthy culture that spreads good ideas and innovative design. Collecting these in an UX newsletter email makes it easier to share with design-interested colleagues, and keeps design in the conversation. Talk to the corporate security to get the top design sites unblocked. It can be surprisingly easy–often they are simply caught in a blanket block licensed from the firewall vendor. Little details like this can make a big impact to employee morale.
“Going native” is what happens when UX’ers understand and accept unchallenged why certain business rules are required and why new approaches are impossible. As representatives of the user, the team must refresh themselves. Good ways to do this are to listen in on customer service calls, visit company stores, and observe real customers. Bringing clients along can be a great team-building outing; many head office execs rarely get enough time with customers to talk about their products.
Development is an ongoing problem. UX’ers who aren’t growing feel like they are stagnating. The single best way is to support their development in ways that help others and build an ecosystem. Encourage white paper writing and presenting at conferences. Learning by teaching is a tried and true method. UX groups have the benefit of an audience with similar interests.
An integrated internal UX team is critical to organizational success, and the stakes are higher in larger enterprises. An internal practice that builds lasting relationships, provides thought leadership, and acts as trusted advisors provides long-lasting value to the firm. As the digital space becomes increasingly human-centric, and organizations evolve offerings around consumer need, the internal user experience agency plays a significant part in delivering both short term wins as well as long term success.
When UX’ers talk, they tend to talk about process, but the ability to deliver an innovative user experience starts before kickoff and lasts after the launch. Repeatable success in UX depends on the right culture. This is particularly important in enterprise scale organizations, with long-lasting relationships.
Having worked as a consultant, at an agency and in-house, I’ve observed that the organizational location and economics of the user experience team can make or break them. When should you bring in an outside team, and when should you hire an individual employee? When might you want to grow an in-house agency?
As firms digitize their business, user experience has gone from marketing to a core business function. Financial service companies have come to embrace this (see “Your interface is your company”). When your products are invisible and complex, web and mobile interfaces define your customers’ opinion. This increased value means increased responsibility: Designers need deep business knowledge, not just wireframing skills. Full “domain knowledge” starts with knowing the basic terminology to business rules, previous project successes and failures, and regulations.
Intimate organizational insight is critical to UX. In the course of a project, UX professionals dig up data and identify solutions to problems beyond the immediate project. A UX future vision accelerates short-term delivery, but also drives the product roadmap. By understanding how the business works, strategic UX’ers can connect them to fundamental insights on how to deepen customer relationships and win.
How to figure out the right team for a new project?
Here are three things to think about for a project’s user experience team strategy.
Domain complexity. The more difficult a project is to learn by a new person the less happy enterprises are with “turnover.” New people can mean more training, delayed projects, and missed deadlines.
Lifespan of project/portfolio. Generally, single marketing campaigns don’t need as much investment into future-proofing. On the other hand, software can live for years; it’s worth investing to make it scalable, consistent, and avoid design entropy.
Scale. An organization’s scale determines the amount of impact a single UX practitioner can have. On small projects, a single designer can do it all, but in a large organization, a few scattered people will have difficulty influencing business strategy or maintaining a consistent UX strategy.
Because of these structural differences, agencies and in-house groups have different strengths and are suited to solve different problems.
An individual consultant can be an effective solution for an experienced client, but too often a consultant’s input does not get the traction it deserves. The temporary nature of the engagement makes it difficult to know the business in depth or to earn relationships that can influence the project. There is a high risk of being relegated to “surface” design. This is the most difficult position to extend into wide ranging influence, as a consultant often lacks the standing to create standards, nor the scale to work on many projects.
One area an individual consultant can have influence is speaking. There are many talented UX people who are happy to adapt their UX conference speeches to a business audience.
There are many related subjects, from mobile trends to analytics to SEO that overlap with user experience. Showing the connection to the latest hot topic to the fundamental user centered design process helps show how UXD helps accomplish a goal that executives have.
With the right introduction, these external experts can demonstrate best practices that can gain a competitive advantage.
A strong employee can positively influence their area of business; however, they also often lack the influence to change business strategy. Ironically, if they are successful at promoting a user centered philosophy, they are unable to satisfy the resulting demand.
Many firms employ isolated UX professionals in various departments, which can make it difficult to define and enforce standards and best practices. Individual user experience practitioners can face limited career paths and pressure to compromise design principles.
Many firms have UX interest groups where people talk shop and share techniques. One approach is to convert this to an “action” group to influence overall strategy. Bring in external speakers, write an analysis of the firm’s user experience, create a user experience mailing group for interesting topics. Find like-minded executives who could champion user experience.
Agencies are an excellent option for a blue-sky rethinking of a product, for crossing lines of business, and providing a neutral third party. Not knowing the business requirements, laws, or what was tried before naturally encourages new thinking. Employees may be less optimistic, or perhaps too realistic, for radical change. Additionally, there is implicit perceived value in the neutrality of an outside opinion, especially from a brand name consultancy; internal stakeholders are more likely to accept mediation with an outsider. Additionally, agencies can deliver a large team quicker and easier than hiring consultants. Agencies can provide trend insights from scanning the field across clients to understand what is being emphasized.
However, the effectiveness of external agencies can be constrained by simple economics. An agency team costs significantly more per person. This limits the type of projects they can execute. Multi-year projects with multiple releases are often not cost effective. Small projects are similarly not possible due to the team approach of agencies: You can’t just hire one person from an agency. Sort of like Goldilocks, projects too long or too small are left without UX assistance.
The nature of working contract by contract requires agencies to focus on different aspects of a project, for example, making elaborate presentations to help the client feel they have gotten value for their money.
Fixed-price contracts force enterprises to work to a rigid schedule, which can be good and bad. Spending money on an external group can bring focus, but it is rare that all groups in a large organization operate on the same schedule. There are usually several large initiatives fighting for attention and core developer resources at any one time.
Doing a large overhaul ensures that some of the teams will not be ready to work during the time the agency is there. An agency is often long gone by the time a software application is launched, preventing usability testing for the next phase and eliminating the chance to fix the UX for challenges encountered during development.
Being external increases the difficulty of getting to know the client’s business. Information is shared less freely with outsiders and access to users is more tightly controlled. This can be as simple as getting a laptop on the local network or as difficult as being licensed or having security clearance. Each barrier reduces the efficiency of a temporary worker. They have to be twice as fast for each hour getting up to speed, travelling, and making presentations. Collaboration is sometimes hindered by an “us vs. them” attitude with agency people working in their office and the client in theirs.
Agencies may execute short-term projects effectively, but being temporarily engaged limits their effectiveness over the long term. Their recommendations may never be built if there is no one championing them in the company. The need for an impressive “big reveal” presentation at the end of the project can get in the way of a spirit of open, iterative design. They are well positioned to create a style guide but poorly positioned to see that it gets distributed, adopted, and maintained over time. An agency is great to make a slogan like “Quality is Job 1” but would struggle to make quality the top priority across a company.
The in-house agency merges aspects of the external agency (scale, coordination, career path, best practices, and standards) with aspects of working in-house (stability, domain knowledge, personal relationships with partners) that can cause change over the long term. Investing in a UX department demonstrates the firm’s commitment to its customers, but it is often simply a practical decision.
In this model, user experience people are located in one group but are assigned to projects when needed. This enables the team to provide the core UX service to projects across an enterprise, without the higher costs of an entire agency team or tie up headcount on a project that does not need to hire a full time UX professional.
Workflow projects with complex business rules are best done inside the company. The business knowledge discovered during analysis is precious and expensive. Stakeholders rarely agree to be interviewed again because a person leaves mid project. Getting a UX professional up to speed is slow and expensive. Hiring an agency is a sure-fire way to lose this information when a team member is moved to another project.
Worse yet, given the higher turnover rates in agencies, the knowledge invested can be permanently lost. From remembering why decisions were made six months earlier to knowing the rules of the business, this is life force of a project. Documentation can help, but mid-project most of the information is held in the heads of the team.
The centralized group provides services that no other model can. It can maintain design standards to give the customer a consistent experience and reduce duplicate work. This reduces costs and improves quality. Members share in-progress work to the group, so a client benefits from the experience of the whole group.
Each of these models has strengths and weaknesses. Assuming the same competency of the people, an in-house agency provides the best long term value to the enterprise from its ability to engage with complex problems and influence the organization widely. As firms recognize the competitive advantage of customer experience, the question becomes: how to make this vision a reality.
The next article will cover building and growing a UX practice that thrives in an enterprise and delivers business value.