Your navigation is probably describing the company, not helping the customer

A visitor lands on a small business website looking for a practical answer: fix a leaking roof, compare bookkeeping options, book a consultation, find the right product, or understand whether the business can handle a particular situation. The website often answers with a menu such as Services, Industries, About, Resources, and Contact.

That menu may be accurate. It may also make the visitor do unnecessary translation work. They must decide which internal service name matches the job they have in mind, then work out whether that service applies to their situation.

The better answer is not to remove services from the site. It is to make customer tasks the main route into the site, while preserving a clear services section for visitors who already know what they want.

This is the distinction that matters: organize the first decision around the customer’s need, then explain the service that meets it. A commercial electrician might lead with “Plan a new installation,” “Fix an urgent fault,” and “Maintain an existing system.” Each path can then introduce electrical design, installation, repairs, or maintenance as the relevant services.

That structure reflects how people actually approach a purchase. GOV.UK’s content guidance says people visit to complete a task, such as applying for a licence or finding out how to vote, and recommends basing user needs and content on actions. Its design principles likewise begin with identifying user needs rather than starting with the organization’s structure. (guidance.publishing.service.gov.uk)

For a small business, the practical lesson is simple: if a first-time visitor has to understand your vocabulary before they can recognize their own problem, the navigation is asking too much.

The research points to findability, not menu fashion

Task-led navigation works when the tasks are real and specific

“Customer tasks” can become just as vague as “Solutions” if the business invents them in a conference room. “Grow better,” “Transform your business,” and “Build confidence” are not useful navigation labels. They describe aspirations, not actions.

A useful task is concrete enough that a customer might say it aloud. “Get a quote for a kitchen remodel” is useful. “Improve your home” is not. “Prepare for an IRS notice” is useful. “Find peace of mind” is not. “Choose a payroll system for a five-person company” is useful. “Optimize operations” is not.

The label should also give the visitor a reliable sense of what lies behind it. Nielsen Norman Group describes this as information scent: the wording should help people predict whether a click will take them closer to what they need. Its guidance warns that abstract labels such as “Explore” and “Connect” provide little context, while specific task language can work well when it matches the user’s goal. (baymard.com)

This is where many task-based redesigns go wrong. The team replaces service names with polished marketing language and calls the result customer-centered. It is still company-centered, only less obvious.

Use the language customers already use in search queries, sales calls, support emails, and intake forms. A customer who searches for “emergency furnace repair” should not be routed through “Comfort Solutions.” A founder looking for help with a website that attracts the wrong leads should not have to infer that “Digital Growth Architecture” means positioning and conversion work.

If the task is too broad, make the destination page do the sorting. A page titled “Choose the right bookkeeping help” can distinguish monthly bookkeeping, catch-up work, payroll support, and financial reporting. The navigation does not need to expose every service in the first menu. It needs to help the visitor identify the right branch.

Task-led entry points beat service-led menus for discovery, but not for every visitor

Do not force every visitor through the same customer journey

The strongest argument against a purely task-based menu is that customers do not all arrive at the same stage of understanding.

Some visitors are exploring a problem. They know the symptom, not the solution. They need a task-led path that helps them decide what kind of help is appropriate. Others arrive from a referral, a proposal, a search for a named service, or a previous conversation. They are looking for confirmation and proof, not orientation.

A site that hides its services in the name of customer empathy creates friction for that second group. Someone who has been told to “ask Acme about payroll processing” should be able to find Payroll Processing without answering a diagnostic questionnaire first.

This is why a hybrid structure is usually the right choice for service businesses:

* Use task or problem language in prominent homepage pathways and relevant landing pages.
* Keep a plain-language Services or What We Do section in the primary navigation.
* Link each task page to the specific service, proof, process, and next step that support it.
* Let industries, customer types, or use cases provide alternate routes when those distinctions genuinely change the offer.
* Use search for visitors who arrive with a precise term, and make the results understandable.

Gartner’s B2B research reinforces the need for this balance. Buyers increasingly prefer to research without a sales representative, yet self-service buyers can be more likely to regret a purchase when the website does not provide enough information or support for evaluation. That suggests the job of navigation is not merely to point to a service page. It must help people understand fit, compare options, find evidence, and know what happens next. (gartner.com)

For a small business, this may mean a homepage with pathways such as “I need help now,” “I’m planning a project,” and “I want ongoing support,” alongside a direct “Services” link. That is more useful than making visitors choose between internal departments they have never heard of.

Build the structure from customer language before touching the menu

1. Collect the jobs customers actually describe

   Review search queries, contact forms, sales notes, support conversations, and proposal requests. Look for repeated verbs, symptoms, situations, and desired outcomes.

2. Separate situations from services

   A situation such as “I need to replace an old roof” is not the same thing as a service such as “roof installation.” Keep both available, but do not confuse them.

3. Draft plain-language entry points

   Write labels that describe a recognizable customer job. Remove claims, slogans, and internal terminology that do not help someone choose a path.

4. Map each task to a destination

   Every task-led page should answer whether you can help, what the relevant service includes, what it costs or depends on, what proof exists, and how to begin.

5. Keep direct service access

   Add a services hub or service links so referred visitors and experienced buyers do not have to navigate through exploratory content.

6. Test the structure without visual design

   Use tree testing to ask people where they would go to complete specific tasks. Nielsen Norman Group distinguishes this from card sorting: card sorting helps discover possible groupings, while tree testing evaluates whether people can find items in a proposed hierarchy. ([nngroup.com](https://www.nngroup.com/articles/tree-testing/?utm_source=openai))

7. Revise from behavior, not stakeholder preference

   If people repeatedly choose the wrong branch, change the label or grouping. Do not defend a term because it is accurate inside the business.

The menu is only the beginning of a task-based site

A task-led navigation system fails if the pages underneath it are still organized around the company. A visitor clicks “I need a repair” and reaches a generic service page with a paragraph about expertise, a list of capabilities, and a contact form. The route changed, but the customer still has to do the work.

The destination should continue the task. State who the page is for, what situation it covers, what the business will do, what information the customer needs to provide, and what a realistic next step looks like. If the visitor needs to compare options, show the comparison. If they need to know whether you serve their location, say so early. If price depends on an assessment, explain what determines it.

The same principle applies to ecommerce and larger catalogs. Baymard’s research found that users repeatedly abandoned sites when they could not find products, despite the sites being major commercial properties. Its category research also found that many sites place promotional content ahead of the subcategories users need to choose the right path. The implication for a small business is direct: navigation pages should help people choose, not make them walk through the company’s preferred presentation order. (baymard.com)

Practitioners report the same pattern in other settings. Scottish Government intranet teams shifted from organizing around functions to organizing around user needs, using card sorting and tree testing to validate labels and structure. The resulting intranet ranked among the strongest tested for task completion and task time. GOV.UK’s navigation research likewise used repeated rounds of task-based testing, changing the structure as users showed what made sense to them. (nngroup.com)

You do not need a government-sized research program. You need a handful of representative tasks, people who resemble your customers, and enough discipline to observe where the structure makes them hesitate. Test the current menu before redesigning it. Then test the proposed menu with the same tasks. The comparison will tell you more than a room full of opinions.