Design-Centering Styles

Overview

Whose interests should shape a design when the customer, the company, the technology, and the designer all pull in different directions?

Whose interests should shape a design when the customer, the company, the technology, and the designer all pull in different directions?

Four competing design styles, each named for what actually drives its decisions: user-centered, company-centered, technology-centered, and designer-centered. Customer-Centered Design is best understood by contrast with the latter three, which it argues against directly, and by its debt to the first.

User-centered design

The closest relative: an approach to engineering useful, usable computer systems pioneered in the 1980s. Customer-Centered Design builds on it but goes further, fusing usability concerns with marketing and business concerns — balancing business goals, marketing goals, usability goals, and customer-experience goals together. On the Web, the harder problem isn’t attracting an audience but converting visitors into customers and bringing them back — a concern that lies outside what user-centered design, focused on task usability, was designed to address.

Company-centered design

The needs and interests of the company — its internal structure, its product taxonomy, its vocabulary — dominate the site instead of customer needs. Two characteristic failure modes:

Brochureware: sites organized around internal corporate structure with sparse real information about products or services, ignoring what the Web as a medium can actually do. Tolerable only as a short-lived first step, never as an end state.
Insider jargon: assuming customers share the company’s internal vocabulary. Worked example: a buyer shopping for a digital camera who just wanted to know “which camera is best for an amateur,” confronted instead with unexplained terms like CCDs, FireWire, and uncompressed TIFF mode — confusing and alienating rather than informative.

Technology-centered design

Sites built around whatever technology is novel or available, with little up-front research into business or customer needs — animation, audio, video, and streaming for their own sake. Treats technology as the end rather than a means to an end. Was common in the Web’s early days and has become less common as the medium matured, but the failure mode (a flashy, slow, hard-to-use site) recurs whenever a new technology trend appears.

Designer-centered design

Also called ego-centered design: the creative and expressive needs of the design team take priority over the needs of customers — captured by the (real, quoted) attitude “the less they talk to us, the better it is. We know what’s best.” Can produce sites that are technically and visually impressive, sometimes deliberately so as a brand statement, but the slow downloads, poor cross-browser support, and learning curve that often accompany that focus are a poor fit for any e-commerce or informational site whose business depends on repeat visitors. More tolerable for art-oriented sites where the visual statement is the point.

Why the contrast matters

Each of the three non-customer-centered styles inverts the same relationship: company-centered design gives no thought to why a customer is visiting or what they want to do; technology-centered design treats technology as an end instead of a tool; designer-centered design subordinates customer needs to the design team’s expressive needs. Customer-centered design flips this — customers and their tasks come first, and technology is one of several tools used to serve them.

Customer-centered design flips this — customers and their tasks come first, and technology is one of several tools used to serve them.

Answer common objections

Nine objections organizations raise against investing in customer-centered design, each rebutted:

“Good design is just common sense.” If that were true, bad sites wouldn’t be so common. Design teams aren’t their customers — they know too much about how a site works internally to see it the way a newcomer does — and intuition alone reliably fails to predict real customer behavior.
“Only experts create good designs.” Experts apply the techniques faster and more rigorously, but the techniques themselves — observing, talking to, and getting feedback from customers — are learnable by anyone willing to do the work.
“Interfaces can be redesigned right before launch.” This assumes the site already has the right features, built correctly — two risky assumptions that get far more expensive to fix the closer a project is to completion (see the cost argument in Iterative Design). Constant feedback throughout the project, not a pre-launch scramble, is what de-risks this.
“Good design takes too long and costs too much.” It adds real up-front cost, but it reduces back-end costs (help-desk load, returns, ongoing maintenance from customer dissatisfaction) and can lower total project cost by catching problems while they’re still cheap to fix.
“Good design is just cool graphics.” Aesthetics matter — good visuals communicate how to use an interface and set an impression — but are only one part of what a design must address. Customer goals, understanding, tasks, and context all sit outside what visuals alone can fix.
“Web interface guidelines will guide you to good design.” Guidelines are a useful checklist against minor implementation problems, but they don’t address what features a site needs, how it should be organized, or how pages should flow into each other. They don’t help with trade-offs the way customer-centered principles, processes, and patterns do.
“Customers can always rely on documentation and help.” Customers have little patience to read documentation just to use a Web site; help pages are a last resort for an already-frustrated customer, not a substitute for a usable design. Trust and Credibility makes the same point from the other direction: clear, honest content earns more trust than a fallback explanation customers have to go dig for.
“Market research reveals all customer needs.” Market research captures attitudes and stated intentions well but doesn’t reliably predict actual behavior, and turning research straight into a feature checklist risks cluttering the design with requests that satisfy individual askers while hurting the experience for everyone else. What customers say needs to be balanced against direct observation of what they do.
“Quality assurance groups ensure Web sites work well.” QA is technology-driven — it verifies the build matches the spec — not customer-driven, and it typically happens only after the site is built, when it’s too late to fix design mistakes. It catches coding errors, not design errors.

Design teams aren’t their customers — they know too much about how a site works internally to see it the way a newcomer does — and intuition alone reliably fails to predict real customer behavior.

A companion list in the source, “Top Ten Signs That Things Are Going Badly,” collects verbal tells of an organization falling into these myths in practice — e.g. “We don’t need user testing, I’m a user and I find it easy to use,” “We’ll just put an ‘Under Construction’ sign there,” and “We need a splash screen.”

Principles

Processes

Sources

The Design of Sites: Ch. 1 — Customer-Centered Web Design: More Than a Good Idea (Ch. 1) is this page’s source — the four contrasting design styles (user-, company-, technology-, and designer-centered) and their characteristic failure modes, and the nine rebutted objections to investing in customer-centered design.