Skip to content
ROI Synth

Information architecture for SEO

Information architecture for SEO decides which pages a site should have, how they are grouped and how they connect. A good structure gives every important topic one clear page, places it under a sensible parent, and keeps it within a few clicks of the homepage.

By ROI SynthUpdated 11 min read

The problem this solves

Architecture problems are usually invisible in single-page audits. They appear when you step back: three pages competing for the same topic, important services buried five levels deep, or a blog that has become the de facto home of the company's expertise while the service pages say little.

Without a deliberate structure, content grows by accretion. New pages are added wherever there is space in the menu, overlapping pages split relevance, and search engines receive mixed signals about which page represents a topic. Visitors face the same confusion.

What the work includes

  • Inventory of existing pages mapped to topics and search intents
  • Identification of overlapping, missing and orphaned topics
  • Proposed hierarchy with parent and child relationships
  • Navigation and hub page recommendations
  • URL and breadcrumb conventions aligned to the hierarchy

How the work is done

  1. Map demand

    Group the searches and questions your audience has into topics and intents.

  2. Map supply

    Assign every existing page to one of those topics and flag gaps and overlaps.

  3. Design the hierarchy

    Decide the parent topics, the pages beneath each, and which existing pages merge, move or are created.

  4. Plan the transition

    Sequence changes, including redirects for merged or moved pages, so visibility is preserved.

When the site has grown without a plan

Creative workspace featuring a laptop, stylus pen, drawing tablet, and office supplies for remote work

Most websites start with a sensible structure and lose it gradually. A new service gets a page wherever there was room in the menu. A campaign landing page becomes permanent. Blog posts accumulate under categories that made sense three years ago. Each decision was reasonable at the time; together they produce a site where nobody can say which page is meant to answer which question.

Information architecture work steps back and decides the structure deliberately. On a coded website it has a particular payoff: the structure lives in routes, folders and a content model, so once it is right, the framework enforces it for every new page.

  • Two or three pages cover the same service in slightly different words, and none of them ranks well.
  • Important pages sit four or five clicks deep or are reachable only from the footer.
  • URL paths reflect old campaigns, an old CMS or an agency's folder habits rather than your services.
  • The navigation lists everything, so it highlights nothing.
  • You are planning new services, markets or a redesign and want the structure settled first.

What a good structure does

A good information architecture is not a diagram for its own sake. It is a set of decisions that make the site easier to use, easier to maintain and easier for search engines to understand. Four principles guide those decisions.

  • One clear home per topic

    Every important topic has exactly one page whose job is to cover it. Other pages link to that page instead of repeating it.

  • Parents that summarise

    Hub pages introduce their children, explain how they differ and link to each, so both visitors and crawlers understand the group.

  • Shallow paths to what matters

    The pages that bring business are within a few clicks of the homepage, through navigation a person would actually use.

  • URLs that describe the structure

    Readable, stable paths that reflect the hierarchy, so a URL tells you where you are and rarely needs to change.

From search demand to a page map

Hand demonstrating business planning using colorful sticky notes, perfect for creative and planning concepts

We start from evidence rather than from the current menu. Keyword and intent research shows the distinct needs your customers search for; your sales knowledge shows which of those needs matter commercially. We group needs that a single page can satisfy, separate needs that require their own page, and then arrange the resulting pages into a hierarchy.

The current site is then mapped onto the new structure. Each existing URL gets a fate: keep as is, keep and improve, merge into another page, or retire with a redirect. Nothing valuable is thrown away, and every change of address is recorded so links and search history carry across.

A hypothetical page map extract for a coded accountancy site
New pageCoversExisting URLs mapped to it
/services/tax-returnsPersonal and sole-trader tax returns/tax, /self-assessment-help, /blog/tax-return-service
/services/bookkeepingMonthly bookkeeping for small companies/bookkeeping, /services/bookkeeping-packages
/industries/contractorsAccounting for independent contractorsNew page (no existing equivalent)
/guides/vat-registrationWhen and how to register for VAT/blog/vat-guide-2021, /blog/do-i-need-vat (merged)

Building the structure into routes and content

On frameworks like Next.js the page map becomes the route tree, and the content model defines what each page type must contain. That gives the architecture teeth. A service page template can require a parent hub, a summary and a set of related services; breadcrumbs and internal links can be generated from the hierarchy rather than hand-written; and the sitemap can be produced from the same data, so it never lists a page the structure does not contain.

We implement the structure in your repository: route folders, typed content definitions, navigation data, breadcrumb generation and redirects from every retired URL. Your team then adds pages by filling in content, and the structure stays intact without anyone having to remember the rules.

Navigation that reflects priorities

Navigation is the most visible expression of information architecture and the most fought over. Every team wants its page in the main menu. A useful rule is that the primary navigation shows the routes most visitors need and that the business most wants to be found for; everything else is reachable through hubs and contextual links one level down.

For larger sites we often recommend a hub-and-spoke pattern: a small number of hubs in the main menu, each introducing its group of pages with a short summary of each. That keeps menus short on phones, gives search engines clear topical groups, and gives visitors a place to compare options before choosing.

How the work runs

Information architecture is delivered within the SEO offer and usually precedes major content work, because writing pages before the structure is settled leads to rewriting them.

Typical sequence
WhenWhat happensWhat you receive
Weeks 1–2Crawl of current site; search demand grouped by intent; workshop on commercial prioritiesCurrent structure diagram and demand groups
Weeks 3–4Proposed page map and URL scheme; mapping of every existing URLPage map and URL mapping for review
Month 2Routes, content types, navigation, breadcrumbs and redirects implementedChanges in your repository; redirect map tested
Month 3 onwardsNew and merged pages written in order of priorityPages published and listed in the progress sheet

How we judge whether the structure works

Structural change shows up in measurable ways. We track click depth of important pages before and after, the number of pages competing for the same queries, indexed pages per section, and impressions and clicks for the hubs and their children. Redirects are monitored for errors and chains after launch.

Because restructuring changes many URLs at once, there is often a short period of fluctuation while search engines process the redirects. We set expectations about that in advance and report it plainly, so a temporary dip is not mistaken for failure — or a real problem mistaken for a temporary dip.

When your own pages compete with each other

One of the clearest signs of weak architecture is cannibalisation: several of your pages appearing, alternately, for the same searches, none of them consistently. Search Console shows it when you filter a query and see impressions split across two or three URLs, with the ranking page changing from week to week.

It usually has an innocent history. A service page was written; later a blog post covered the same service in more depth; later still a landing page was built for a campaign. Each is decent on its own. Together they divide the signals — links, engagement, relevance — that a single strong page would collect.

The architecture answer is to choose the page that should own the topic, merge the best material from the others into it, and redirect or re-focus the rest. A blog post that duplicated a service page might become a narrower guide on one aspect of the service, linking back to the main page as the place to buy. The goal is not fewer pages for its own sake; it is one clear answer per question.

A hypothetical restructure

A hypothetical engineering consultancy has a Next.js site built in stages over five years. It offers structural surveys, planning drawings and building regulations support, in three regions. The current site has a single services page listing everything, a separate page for each region that repeats the service list, a blog with forty articles, and a projects section organised by year.

The new structure gives each service its own page with the questions buyers ask about it, under a services hub that explains how they differ. Regions become a section describing coverage, travel and local experience, each linking to the services offered there rather than repeating them. Projects are reorganised by service, so a reader looking at structural surveys sees relevant examples. The forty articles are mapped: twelve support specific services and are linked from them, twenty are merged into five comprehensive guides, and eight are retired with redirects.

The route tree in the codebase mirrors the new map: one dynamic route for services, one for regions, one for projects nested under their service, and one for guides. Breadcrumbs and hub listings are generated from that tree. Every one of the old URLs has a redirect target recorded in a single file, reviewed before launch and tested after it.

What we need from you

Architecture decisions depend on knowledge that is not in any crawl: which services you want to grow, which are being retired, how you describe your offer to clients, and which audiences you serve differently. A one-hour workshop early in the work usually settles the big questions; the rest is handled in comments on the page map.

We also need access to analytics and Search Console, so decisions about merging or retiring pages take their current value into account, and repository access to implement the routes, content types and redirects. If your team has naming conventions for URLs or components, we follow them.

Rules for URLs that last

URLs are part of the architecture that visitors, search engines and other sites all depend on. Changing them is possible with redirects, but every change has a cost, so it pays to design them once with a few durable rules. On a coded site these rules are enforced by the route structure and by a slug function in the content model.

  • Lowercase words separated by hyphens, describing the page in plain language.
  • One level of folder per level of hierarchy that matters to the visitor, and no deeper.
  • No dates, IDs or campaign names in paths for pages meant to last.
  • One trailing-slash convention, enforced with redirects for the other form.
  • Stable slugs: a page keeps its URL when its title is edited, unless there is a reason to change both.

What a good hub page contains

Hubs are where architecture becomes visible to readers. A weak hub is a grid of logos or a list of links with no explanation. A strong hub helps someone choose: it says what the group covers, explains how the options differ, and gives each child page a sentence that tells the reader whether it is the one they need.

For search engines, a strong hub concentrates relevance for the broader topic and passes it to the more specific pages beneath. For your team, it is a natural place to answer comparison questions — which service is right for whom — that otherwise end up scattered across sales emails.

  • A clear introduction

    Two or three sentences on what the group covers and who it is for.

  • A short entry per child

    A heading and a sentence or two that would help someone choose, linking to the child page.

  • Comparison where it helps

    A small table when the differences between options are what readers need most.

  • A next step

    A way to ask for help choosing, for readers who still are not sure.

The deliverables, listed

Architecture work produces a structure you can see in the code, not just a diagram.

  • A page map with every proposed page, its purpose and its parent.
  • A URL mapping for every existing page: keep, improve, merge or retire.
  • Routes, content types, navigation data and breadcrumbs implemented in your repository.
  • A tested redirect file for every URL that changes.

How architecture connects to other work

Architecture is shaped by keyword and search intent research, which shows the distinct needs the structure must serve, and by a content audit, which decides what existing pages are worth keeping. It is expressed through an internal linking strategy, which turns the hierarchy into paths readers and crawlers can follow.

When a restructure changes many URLs, the launch is handled with the same care as site migration SEO: a complete redirect map, tests before release and monitoring afterwards. And if a redesign is already planned, architecture work belongs at the start of website redesign SEO, before wireframes fix the navigation in place.

Which offer includes information architecture

Information architecture is part of the SEO offer, from $200 per month, and of All-in-One SEO + AEO + GEO, from $300 per month. On a small site the design and implementation fit within one or two months; larger restructures are phased, with the most valuable sections first.

If the restructure is significant enough that most templates would change anyway, the Next.js + TypeScript Website Build, from $1,000 one time, can deliver the new structure and the new site together, with the redirect map included.

Concerns about restructuring

What owners and marketing leads usually want to know before changing URLs.

  • "Will changing URLs hurt our rankings?"

    Done carefully, with one-to-one redirects and updated internal links, most visibility carries across. We change URLs only where the benefit is clear.

  • "Can we keep our current design?"

    Yes. Structure and design are separate decisions. Most restructures change navigation and URLs, not the look of the site.

  • "Who decides what goes in the menu?"

    You do, with evidence. We bring search demand and usage data; you bring commercial priorities.

  • "What about our blog?"

    Articles are mapped like any other page: kept, improved, merged into a guide, or retired with a redirect.

How progress is evaluated

  • Number of topics with competing pages before and after consolidation
  • Click depth of priority pages in crawls
  • Impressions and clicks per topic group rather than per page

Who this is for

  • Sites that have grown without a plan
  • Businesses adding new product lines or services
  • Teams preparing a redesign

Limits and honest caveats

Common questions

Is a flat site structure better for SEO?

Keeping important pages within a few clicks of the homepage helps, but a structure with no hierarchy makes topics harder to understand. The goal is short paths to important pages and clear grouping, not the fewest possible levels.

What is information architecture for SEO?

Information architecture for SEO decides which pages a site should have, how they are grouped and how they connect. A good structure gives every important topic one clear page, places it under a sensible parent, and keeps it within a few clicks of the homepage.

What does a information architecture for SEO engagement include?

The work includes inventory of existing pages mapped to topics and search intents, identification of overlapping, missing and orphaned topics, proposed hierarchy with parent and child relationships, navigation and hub page recommendations and URL and breadcrumb conventions aligned to the hierarchy.

What problems does information architecture for SEO solve?

Without a deliberate structure, content grows by accretion. New pages are added wherever there is space in the menu, overlapping pages split relevance, and search engines receive mixed signals about which page represents a topic. Visitors face the same confusion.

How is information architecture for SEO carried out?

It runs in 4 stages: map demand, map supply, design the hierarchy and plan the transition. Group the searches and questions your audience has into topics and intents.

Who is information architecture for SEO a good fit for?

It suits sites that have grown without a plan, businesses adding new product lines or services and teams preparing a redesign.

How do you measure the results of information architecture for SEO?

Progress is measured with number of topics with competing pages before and after consolidation, click depth of priority pages in crawls and impressions and clicks per topic group rather than per page.

What are the limits of information architecture for SEO?

Restructuring can cause temporary volatility while search engines reprocess the site. Navigation must serve users first; an SEO-only hierarchy rarely works.

Can ROI Synth carry out the information architecture for SEO work, or only recommend it?

Implementation is included. Within the agreed scope we fix templates, metadata, internal links and redirects, and build the landing pages ourselves, on your existing platform or on a Next.js and TypeScript build when a rebuild is part of the project, starting with steps such as “Inventory of existing pages mapped to topics and search intents”. Every completed item is recorded with evidence in a weekly progress sheet.

How long does information architecture for SEO take?

The 4 stages (map demand, map supply, design the hierarchy and plan the transition) fit the proposal's four weekly phases. When the agreed scope can be delivered in a month, the proposal commits to completing it in the first month, subject to timely access and approvals. Search engines then take weeks or months to re-evaluate the changes.

Sources and further reading

Contact us about Information architecture for SEO

Tell us about your website and goals. We will say plainly whether this is the right offer, what it would include, and what it can and cannot achieve.

Prefer the full contact page?

Fields marked with an asterisk are required.

Your website's situation, the services and markets you want to grow, and any deadlines. Up to 5,000 characters.

Your details are used only to respond to this general enquiry. See the privacy policy.

Discuss information architecture for SEO for your business

Tell us about your website and goals. We will confirm whether this work is the right priority, what it would include and how it would be delivered.