Iron Goo
Iron Goo robot examines a glowing path from a smartphone to a warmly lit service business

What Is UX? A Small Business Journey Dissected

Atamyrat Hangeldiyev
Atamyrat Hangeldiyev
Systems Architect
March 4, 2026
On this page
UX

What is UX? User experience is the “user's perceptions and responses that result from the use and/or anticipated use of a system, product or service,” according to ISO 9241-11:2018. Those responses can occur before, during and after use. The consequence is that a search result sets part of the experience before someone reaches your site, and the confirmation and follow-up continue it after they submit a form.

Usability has a similarly precise scope. You can assess whether a particular person can achieve a particular goal in a particular context. Calling a website “usable” without naming any of those conditions leaves out the evidence needed to make the claim.

The definitions. ISO 9241-11:2018, clause 3.1.1, defines usability as the “extent to which a system, product or service can be used by specified users to achieve specified goals with effectiveness, efficiency and satisfaction in a specified context of use.” The standard's clause 3.2.3 gives the UX definition above, carried from ISO 9241-210.

Effectiveness means the “accuracy and completeness with which users achieve specified goals.” Efficiency is the “resources used in relation to the results achieved,” including resources such as time and effort. Satisfaction is the “extent to which the user's physical, cognitive and emotional responses that result from the use of a system, product or service meet the user's needs and expectations.” The context of use combines users, goals and tasks, resources, and environment.

Don Norman and Jakob Nielsen give a company-wide definition of user experience:

“User experience” encompasses all aspects of the end-user's interaction with the company, its services, and its products.

Don Norman and Jakob Nielsen, The Definition of User Experience

Norman and Nielsen distinguish UX from the user interface, or UI. The UI is the interface a person interacts with, such as the navigation, buttons and form fields; usability is a quality attribute of that interface. UX also includes the person's expectations and responses around the whole interaction with the company.

For the anatomy that follows, assume a hypothetical local plumbing business offers first-visit appointments. A customer uses a phone to book a visit for a leaking faucet. The task is done when the customer receives a confirmation stating the appointment details and what happens next; the business also sends a reminder and, where needed, follows up with a person. The anatomy and its one-hour checks are Iron Goo's method, built from standard observations, and no standard prescribes them.

A phone booking journey from search to follow-up, with context, speed, trust and accessibility along the path

Follow the path from the search result to the appointment follow-up. Context, speed, trust, accessibility and microcopy affect more than one stop along the way.

A customer journey is the sequence of interactions a person has while pursuing a goal. For this booking, it begins with an expectation from search, continues through the website and form, and reaches the confirmation and follow-up. Sources were read on 29 October 2026.

Part 1, the user and goal: name the booking the person needs

What it is. The user is the person attempting the task, and the goal is the result that person wants. In the hypothetical journey, the customer wants an appointment for a leaking faucet; the business wants a completed booking. Those goals overlap, but the customer also needs to know whether the plumber handles the problem and when a visit is available.

ISO's usability definition asks for specified users and goals for a reason. A form can work well for someone booking a routine visit and fail someone seeking an urgent repair, even when both use the same page. A task should be specific enough that you can tell whether the person completed it accurately.

The one-hour check. Write a sentence in this form: “A customer with a leaking faucet wants to book a first visit on a phone.” Replace the example with a real service and goal at your company. In Search Console, inspect the top 20 search queries, or read the first lines of the last 20 enquiry emails, and record the words people use to describe that need.

Compare those words with the service and booking labels on your site. You are looking for observable matches and gaps: does the site say “faucet repair” where customers ask for that service, and does it identify a way to book? Keep private enquiry details out of any shared audit notes.

The failure that costs customers. If the hypothetical plumber's site groups the relevant service under a vague internal category, the customer may never recognize that the business can help. A completed form cannot rescue a journey that the customer does not think applies to them.

Part 2, context: test the phone and conditions the customer has

What it is. Context includes the user's goals and tasks, available resources and environment. A phone, a mobile connection, limited time and unfamiliarity with plumbing terms all change what the same booking task demands. The customer may have one hand available and may not know whether the problem qualifies as an emergency.

A desktop screen in a quiet office is one context. It provides useful evidence about that context, but it does not establish that the phone journey works. Prior knowledge matters too: the person running the business already knows what its service categories mean.

The one-hour check. Open the GA4 Tech details report and inspect device category to see the mobile share for your site. Then use a phone on mobile data, hold it in one hand and start at the same kind of search result a customer might use. Give yourself the stated task without opening an internal price list or asking a colleague what a label means.

Record the device, connection and starting point alongside each observation. Check whether you can read the service description, choose an appointment and enter the required details without switching to a desktop. If GA4 has little or no useful traffic data, record that limitation and still perform the phone check.

The failure that costs customers. A booking control that is easy to operate with a mouse may require repeated taps on a phone. If the hypothetical customer is trying to arrange a repair between other obligations, each unclear choice adds effort to a task they expected to be quick.

What it is. Information scent is a person's imperfect estimate of the value a source will provide, based on its representation, such as a link label, summary or image. The concept comes from Pirolli and Card's information foraging theory. Raluca Budiu's explanation of information scent recommends labels that describe the destination succinctly and accurately. The customer uses those clues before deciding whether a click is worth the effort.

For the hypothetical plumbing visit, “Book a plumbing appointment” gives a stronger clue than “Learn more.” A search result also makes a promise: if it suggests faucet repair, the landing page should make that service identifiable without requiring the visitor to decode company terminology.

The one-hour check. Start at the search result, then inspect the landing page and home page on a phone. Ask a colleague unfamiliar with the site to look for five seconds and say where they would click to book the stated service. Record the label they choose and whether it actually leads toward booking.

Read the navigation and prominent links against the phrases collected from enquiries. List every “Learn more” or “More” link on the route and identify whether its surrounding text makes the destination clear. Open each candidate link and check that the destination delivers what its label suggests.

The failure that costs customers. The customer sees several identical “Learn more” links and cannot tell which one leads to a repair appointment. They may choose the wrong service, return to the search results or leave without discovering that the booking form exists.

Part 4, task flow: count every screen, tap and field

What it is. The task flow is the sequence of actions between identifying the service and completing the booking. Count the screens the customer sees, the taps needed to proceed and the fields they must complete. A step earns its place when it obtains information or makes a decision needed for the appointment.

Baymard reports that the average US ecommerce checkout displays 23.48 form elements by default, including 14.88 form fields. That is a reference point for a full ecommerce checkout, not a target for a service booking form. A plumbing appointment has its own information requirements.

The one-hour check. Starting at the landing page, write down each screen and count the taps required to reach the confirmation. Count each field separately, including optional fields, and mark where information is requested more than once. For every required field, write down the decision the business uses it to make.

Check whether the customer can tell which steps remain and whether a previous answer survives when they go back. If the sequence needs work, designing the conversion path provides a way to redesign the steps around the task. Keep your original count so you can compare the revised route with the one you observed.

The failure that costs customers. The hypothetical customer selects faucet repair, then has to describe the same issue in several required fields before seeing an appointment time. The repeated work can make the booking feel uncertain and more demanding than the customer expected.

Part 5, content and microcopy: answer the question beside the field

What it is. Content explains the service and its conditions. Microcopy is the short text that helps someone act, including button labels, field labels and instructions. On a booking route, these words should tell the customer what to provide and what the business will do with it.

A label such as “Phone number” identifies a field; a nearby explanation can say whether the number is needed to confirm the visit. A button labeled “Request an appointment” also sets a different expectation from “Confirm appointment.” Use the wording that matches what actually happens.

The one-hour check. Read every heading, field label, instruction and button on the phone journey aloud. Mark terms that require knowledge of the business, then check whether the page explains available appointment times, any known callout fee or price range, and the next step before the customer submits. If a price cannot be stated, say what information is needed before the business can provide one.

Ask a colleague to point to the text that answers two questions: “What am I submitting?” and “What happens after I submit it?” Note any answer that exists only in a staff member's head. The check takes only the time needed to walk the route and read what is on the screen.

The failure that costs customers. The hypothetical customer reaches a button labeled “Continue” without knowing whether it will display available times or send a request. If the page also says nothing about a possible visit fee, the customer has to commit personal details while an obvious question remains unanswered.

Part 6, feedback and errors: show what happened and how to fix it

What it is. Feedback tells the customer what the system is doing and what happened after an action. Jakob Nielsen's usability heuristics call for keeping users informed, preventing errors, and presenting errors in plain language that identifies the problem and suggests a fix. All three apply to the appointment form.

Nielsen's response-time limits provide useful distinctions: about 0.1 second feels instantaneous, one second keeps the flow of thought uninterrupted, and ten seconds is approximately the limit for keeping attention on a dialogue. For a longer wait, a person needs feedback about when the operation will finish. A silent button offers no such information.

The one-hour check. Submit the form empty and record which fields are identified. Enter a deliberately wrong phone format, submit again and check whether the message identifies the phone field and gives an accepted format. Then switch the network off before submitting, restore it afterward and check whether the form explains the failure and preserves what you entered.

Try a valid submission as well. Watch for a visible response after the tap, and check whether a repeated tap creates another request. Use a stopwatch for any noticeable wait; record what the screen says during it and whether the eventual result is clear.

The failure that costs customers. The hypothetical customer enters the appointment details, mistypes a phone number and receives “Invalid input” at the top of the form. If the form also clears their other answers, they must work out the problem and start again. Some will leave before the business ever receives the request.

Part 7, speed and reliability: measure the phone experience

What it is. Speed affects whether the customer can see and operate the booking route promptly; reliability affects whether it keeps working through submission. Core Web Vitals define three measures of loading, responsiveness and visual stability. Google's Web Vitals guidance sets good thresholds at the 75th percentile of page loads, segmented by mobile and desktop.

2.5 seconds
Largest Contentful Paint
200 milliseconds
Interaction to Next Paint
0.1
Cumulative Layout Shift
24 by 24 CSS pixels
Minimum pointer target

LCP describes when the largest content element renders. INP reflects interaction responsiveness, and CLS measures unexpected layout movement. A moving appointment button can cause a wrong tap even when the page eventually loads.

The one-hour check. Run the booking landing page and the form URL through PageSpeed Insights. Record whether each report has field data, then record mobile LCP, INP and CLS alongside the thresholds. Repeat the task on a real phone over mobile data and note slow loads, delayed controls, shifting elements and failed submissions.

Chrome's UX Report includes public pages and origins that are sufficiently popular; Google does not disclose the eligibility threshold, and a page cannot be submitted for inclusion. Many small business pages may therefore have no field data. In that case, the lab results and phone test are useful observations, but neither supplies a 75th-percentile measurement of your customers' experience.

A green lab score is not field data

A lab result describes a test under its test conditions. Core Web Vitals thresholds apply to field measurements at the 75th percentile, so record whether those measurements actually exist before calling the customer experience good.

If a slow template and the absence of a performance budget make the site itself the bottleneck, Foundation is a site rebuild with a 90 plus mobile Lighthouse score and performance budgets.

The failure that costs customers. The hypothetical customer taps an appointment time just as the page shifts and another control moves under their finger. If the form then appears unresponsive, they cannot tell whether to wait or tap again. Speed and stability have become part of whether the booking can be completed.

Part 8, accessibility: operate the same journey without a mouse

What it is. Accessibility asks whether people with different abilities and ways of interacting can use the journey. WCAG 2.2 became a W3C Recommendation on 5 October 2023. It added nine success criteria and removed the obsolete 4.1.1 Parsing criterion, according to the W3C's WCAG 2.2 overview.

For the booking route, 2.5.8 Target Size (Minimum), level AA, requires pointer targets of at least 24 by 24 CSS pixels, with exceptions including spacing. Criterion 3.3.7 Redundant Entry addresses asking for the same information again in a session. Criterion 3.2.6 Consistent Help requires help to appear in a consistent place when it appears on multiple pages, and 2.4.11 Focus Not Obscured (Minimum) requires a focused item to remain at least partially visible.

Text contrast under 1.4.3 Contrast (Minimum) is at least 4.5:1 for ordinary text and 3:1 for large-scale text. These are observable requirements, although checking them alone does not establish that the whole journey conforms to WCAG.

The one-hour check. On a computer, use the keyboard Tab key to move through the booking route and form without a mouse. Check that you can see the focused control, that a sticky bar does not hide it, and that you can reach and operate the submit button. On the phone, inspect small appointment and form controls; use browser developer tools to check target dimensions where needed, including whether an exception applies.

Run an automated accessibility checker and record its findings, then manually inspect labels, focus and contrast on the route. The WebAIM Million's February 2026 examination of one million home pages detected WCAG 2 failures on 95.9% of them, with an average of 56.1 detected errors per page. Automated detection has limits: an absence of detected errors does not establish accessibility or conformance.

The failure that costs customers. The hypothetical customer using the keyboard reaches a time selector, but a sticky booking bar covers the focused option. They cannot see which appointment they are selecting. The form may be technically present while the customer cannot complete it with confidence.

Part 9, trust signals: explain why the business needs the details

What it is. Trust signals help a customer decide whether to proceed with an unfamiliar business. For the hypothetical booking, a stranger may need a real business phone number and address, a clear explanation of the service, and a statement of how their number will be used. Those are practical checks, not a guarantee that every visitor will trust the business.

Baymard's checkout research reports that 19% of US online shoppers who abandoned during checkout did not trust the site with credit card information, while 40% cited extra costs that were too high. Those figures exclude the “just browsing or not ready to buy” segment; the page does not state the sample size for those reasons. They describe ecommerce checkout, so they cannot tell you what percentage of plumbing customers will abandon a booking form.

Baymard also reports a 70.22% average documented cart abandonment rate across 50 studies, with its page last updated on 22 September 2025. Use metrics that matter for a small business to decide which numbers to track for your own journey. A booking request and a shopping cart are different tasks.

The one-hour check. Before entering a phone number, inspect what the form and nearby page say about contact. Check for the business's actual contact details, a truthful price range or explanation of how pricing is supplied, a source for any reviews shown, and a privacy line explaining the proposed use of the number. Follow the contact and review links to confirm that they lead where their labels say.

Ask a colleague who has not used the site to identify who will receive the request and what happens next. Write down any answer they cannot find on the route. If the form asks for information the business does not need to arrange the visit, identify why it is there.

The failure that costs customers. The hypothetical customer is willing to describe the leaking faucet but sees no explanation for a required phone number and no clear business contact details. They may leave at that field because they cannot judge what contact to expect.

Part 10, after the task: confirm the appointment and follow through

What it is. The confirmation screen, confirmation message, reminder and human follow-up belong to the experience because ISO includes responses after use. A customer who submits a request still needs to know whether it arrived and whether the appointment is actually booked. A successful form submission and a confirmed appointment can be separate events.

The confirmation should reflect the real process. If the business must review a request before assigning a time, say so and state when the customer should expect a response. If a time is confirmed immediately, show the details needed to recognize and attend the appointment.

The one-hour check. Book yourself through the real phone journey, using a test appointment that your team can identify and remove. Time the interval from submission to the confirmation screen and from submission to the confirmation message. Check the service, date, time, contact details and next action against what you entered.

Record when a reminder and any human follow-up arrive, and compare them with the timing promised to the customer. The initial submission check can fit inside an hour; a reminder scheduled for a later day requires a later observation. Use measuring UX and fixing what fails to track the journey over time.

A business owner tests a booking on a phone while a colleague times each step

The owner follows the booking route on a phone while a colleague records the time and observes each step. Include the confirmation in the timed journey.

The failure that costs customers. The hypothetical customer submits the form and sees only a blank page. An email arrives later saying the business will call, but it gives no expected time. The customer cannot tell whether to keep waiting or contact another plumber.

Watch five comparable people attempt the journey

Your own check identifies visible defects, but you already know the business and its terminology. Recruit people comparable to the intended customer, give each the same goal without telling them where to tap, and watch what they do. Record the starting point, where each person hesitates, the errors they encounter and whether they reach the stated completion condition.

In Jakob Nielsen's five-user finding, the Nielsen and Landauer model is N(1 - (1 - L)^n), where N is the total number of usability problems, L is the proportion of usability problems discovered while testing a single user, and n is the number of users. Nielsen gives 31% as a typical value of L, averaged across many studied projects. With five users, the arithmetic produces about 84% to 85%:

Estimated proportion of problems found with five comparable users

1 - (1 - 0.31)^5 = 1 - 0.69^5 = 0.8436

That estimate depends on comparable users using the site in fairly similar ways. It is not a promise that five people will uncover 85% of the problems on your particular booking form. The finding also says at least 15 users are needed to discover all usability problems, and recommends spending a 15-person budget on three iterative studies of five: test, fix and test again. The five-user rule does not apply to quantitative studies of usability metrics; Nielsen recommends testing 20 users for those studies.

Distinct user groups need their own coverage. Nielsen suggests three to four participants per group when there are two groups, and three per group when there are three or more. For practical session setup and note-taking, use UX research without a research team. After the first sessions, fix the observed obstacles and give the revised booking task to the next participants without coaching them through it.

Free ebook

Put your next steps in order

Build on what you have just read with a free ebook on semantic SEO and AI visibility. Get worked examples, decision rules, checklists and a do-this-week action in every chapter, plus a six-week plan for your site.

By signing up and confirming your address, you consent to receive the free ebook plus one email every two weeks with 3 or 4 new Iron Goo videos, guides and articles, and nothing else; unsubscribe any time with one click.

Related in UX

Ready to move?

Tell us where your business is today. A written assessment lands in your inbox within two business days.

Talk to us