Customer-Centered Design

Overview

Should design decisions follow the customer’s existing knowledge and expectations, or the design team’s internal conventions and guesswork?

A design approach that builds decisions around the customer’s existing knowledge, expectations, and goals rather than around internal conventions or guesswork. A site is customer-centered when it scores well on five factors: content, ease of use, performance, brand value/trustworthiness, and satisfaction. The term customer is used deliberately instead of user: it covers anyone who uses or depends on the site (administrators, partners, managers, producers — not just end-customers), and signals that the design has to account for more than usability alone (trustworthiness, brand value, how well the company’s other customer-facing channels work). This is a deliberate variant of Donald Norman‘s earlier user-centered design — the same underlying priority of people before internal convention or aesthetics, applied to a broader group and a broader set of concerns.

Two pillars, put into practice through processes and patterns

Customer-centered design rests on two pillars: Know Your Customers — understanding people, their tasks, their technology, and their social context before designing anything — and Iterative Design — keeping customers involved throughout the design and implementation process via repeated prototyping and evaluation, illustrated below. These two pillars are the book’s principles; processes (Web Site Development Process) describe how to put them into practice across a project, and patterns (the lettered pattern groups, A–M) are the reusable, tailorable solutions principles and processes are applied through. Principles arbitrate between process activities (e.g. whether to do one more paper-prototype iteration or move to a high-fidelity build) and between competing pattern-based solutions to the same problem.

Why it matters: the business case

Customer-centered design isn’t purely a quality nicety — it’s argued to directly affect revenue. One cited case: an unnamed IBM site redesign that simplified and made navigation more consistent (after IBM noticed customers leaned heavily on search and help because they couldn’t otherwise find things) led to a sharp drop in search/help reliance and a 400% jump in online sales within a week of launch. More generally, the source’s own research found that a 5% increase in customer satisfaction was associated with a 25%-or-greater increase in revenue — driven by easier-to-find products (more repeat visits) and fewer support contacts/returns (lower cost). Yahoo! is offered as a standing example: its popularity is attributed less to visual design than to a fast, text-heavy, easy-to-use interface shaped by ongoing field studies, interviews, and usability evaluations baked into its development process.

A 5% increase in customer satisfaction was associated with a 25%-or-greater increase in revenue — driven by easier-to-find products (more repeat visits) and fewer support contacts/returns (lower cost).

It’s also explicitly contrasted with three design styles it argues against — company-centered, technology-centered, and designer-centered design — and defended against nine common organizational objections to investing in it; both covered in Design-Centering Styles.

In practice: keeping customers involved

People arriving at a Web site bring prior experience with the Web and with physical-world objects (e.g., recognizing shaded buttons as clickable, see Action Buttons); a customer-centered design deliberately reuses those expectations instead of inventing unfamiliar interactions.

In practice, this means keeping customers in the loop throughout the design process rather than only at the end. A worked example in the source: a designer noticing checkout abandonment in server logs, consulting relevant design patterns (e.g., Process Funnel) to generate candidate fixes, sketching alternatives, then validating them by recruiting a small number of representative customers (five, in the example) to walk through both the existing and the redesigned flow. Observing real customers surfaced problems the design team had not anticipated — e.g., a key button positioned below the fold was invisible on smaller/older monitors, causing users to click the wrong link and lose entered form data when they fell out of the flow.

This also points to a methodological caveat: when the observer is present during testing (as in informal in-home sessions), participants may complete a task despite real usability problems, an effect attributed to testing bias — task completion alone shouldn’t be read as evidence that a flow is easy to use.

Task completion alone shouldn’t be read as evidence that a flow is easy to use.

Patterns

Processes

Sources

The Design of Sites: Ch. 1 — Customer-Centered Web Design: More Than a Good Idea (Ch. 1) is the source for this page’s five-factor definition of a customer-centered site, the deliberate “customer” vs. “user” framing, and the business case above — the IBM redesign’s 400% sales jump and the 5%-satisfaction/25%-revenue finding.

The Design of Sites: Ch. 2 — Making the Most of Web Design Patterns (Ch. 2) supplies the principles/processes/patterns three-part structure described above, and the worked example of a designer diagnosing checkout abandonment and validating a redesign by testing it with five recruited customers.

The Design of Sites: Ch. 3 — Knowing Your Customers: Principles and Techniques (Ch. 3) is the source for the Know Your Customers pillar as described here: understanding customers’ people, tasks, technology, and social context before any design work begins.

The Design of Sites: Ch. 4 — Involving Customers with Iterative Design (Ch. 4) is the source for the Iterative Design pillar and for the testing-bias caveat — that a customer completing a task while being observed in an informal session doesn’t by itself prove the flow is easy to use.

The Design of Everyday Things (Wikipedia) is the source for the Overview’s note on user-centered design — Donald Norman’s coinage of the term in The Design of Everyday Things.