Blog or Knowledge Base: How to Organize Expert Content on a Website

Why do good articles sit on page seven after a year?

A business starts writing articles and sets up a blog for them. Each new article appears at the top; the older ones slide down. After two years there are forty articles. The most thorough text, which answers the most frequent customer question, sits on the fourth page of the overview, between a trade fair announcement and a New Year greeting. Search may still find it. But whoever looks for it on the website does not.

This is not a failure of writing but of organization. This analysis therefore answers one question: How should a small or mid-sized business organize its expert content on the website, as a blog by date, as a knowledge base by topic, or as a mix? To answer it, we first clarify the terms and a model from information architecture. Then we weigh up both forms, present a simple model for assigning content and describe how an existing blog is rebuilt.

A blog asks what is new. A knowledge base asks what holds. Most customers are looking for the second.

Terms: blog, knowledge base, topic page

A blog is a section of the website in which articles appear in chronological order, newest first. Each article has a date and often categories and tags. A knowledge base, sometimes called a guide section or expertise section, orders content by topic. The entry point is not a list of the latest articles but an overview of topics. A topic page is the overview page of a topic that introduces the question and links to all in-depth articles.

The article SEO for SMEs describes this order as the pillar-cluster principle: one overview page, the pillar, covers a main topic in overview. Several articles, the cluster, go deeper into sub-questions. All are linked to each other and to the overview page. A knowledge base is the collection of such topic areas.

The thinking model: exact and ambiguous organization schemes

Information architecture, the discipline of how content is organized and made findable, has distinguished two groups of organization schemes since the standard work by Rosenfeld and Morville. Exact schemes divide content by a clear-cut feature: alphabetically, by location or by time. Every item belongs in exactly one place, and whoever knows what they are looking for finds it quickly. Ambiguous schemes divide by meaning: by topic, by task, by audience or in hybrid forms. Here someone has to decide where an item belongs, and the searcher has to understand the division.

Krzesaj (2024), in a literature review with examples from websites in various sectors, examined how these schemes affect the quality of the information presented. According to it, exact schemes such as chronological order mainly improve features such as timeliness, unambiguity, speed and verifiability. Ambiguous schemes such as ordering by topic or task mainly affect relevance, comprehensibility and credibility. The author names the limits himself: a limited number of examples and no measurement with users. He also observes that company websites often structure content by their own organization, for example by department, rather than by visitors’ questions.

For expert articles, this means: the chronological order of a blog is unambiguous and easy to maintain, but it answers a question visitors rarely ask, namely what was written most recently. Ordering by topic requires decisions, but it matches the question visitors arrive with: what does this business know about my problem?

Two views: lively blog or organized library?

The first view speaks for the blog. A blog shows at a glance that the business is active and thinks regularly. It can be set up immediately with any common website system, and new articles need no decision about where they belong. Besides, attention to new things declines over time anyway. Wu and Huberman (2007) showed on a news platform with around one million users that interest in new items fades according to a regular pattern, as if there were a natural time span after which attention wanes. Whoever regularly brings something new renews this attention.

The second view speaks for the knowledge base. The finding by Wu and Huberman applies to news, whose value lies in its novelty. An article on what maintenance costs or how to recognize a leaking roof does not lose its value after a week. It is searched for over years, and its place on the website should reflect that. A date above such an article also raises a false question in the reader’s mind: is this still current?

I consider the second view the more accurate one for most businesses, without dismissing the first. The expert content of small businesses mostly answers questions that hold for a long time. It belongs in an order by topic. There is news too, but less than a blog suggests. It needs a small place of its own, not the main stage.

The stream-and-library model: which content goes where

For assigning content, I use a simple model, the stream-and-library model. It separates two kinds of content. In the stream flows content whose value lies in its timing: a trade fair appearance, a new regulation, a new employee, commentary on a current event. It may pass by. In the library stands content whose value lies in the answer: what does it cost, how does it work, which is better, how do I recognize a problem? It is ordered, linked and maintained.

The test question for every piece of content is: will this still hold in two years if it is checked once a year? If yes, it belongs in the library. If no, in the stream. An article on a new obligation, such as e-invoicing, can appear in both: as a short news item in the stream when the rule is adopted, and as a maintained answer in the library that describes what businesses have to do.

How the knowledge base and sales work together

A knowledge base organized by topic has a further advantage that goes beyond the website. Sales can use it. The article B2B content marketing describes the question shelf, which sorts articles into three compartments: before the conversation, between conversation and quote, before the decision. A knowledge base ordered by topic and, within topics, by these compartments becomes a reference work for sales. Whoever wants to send a fitting article after a first conversation finds it in a minute instead of leafing through a blog.

The connection to the service also becomes easier. Every topic page refers to the service the topic belongs to, and every article to its topic page. The article Internal linking describes these paths with the four-way rule: up to the overview, across to the neighboring page, down to the in-depth page and to the service. In a blog in which articles are connected only by date and tag, these paths rarely arise by themselves.

Short questions that do not need a page of their own belong neither in the blog nor in the knowledge base but on the FAQ page. How it is built and how it refers to longer answers is described in the article The FAQ page.

Rebuilding an existing blog without losing visibility

Many businesses already have a blog with dozens of articles. Rebuilding it into a knowledge base does not have to change everything at once. What matters is that existing addresses are kept or cleanly redirected, because visitors from search arrive via them today.

1. Sort the inventory

   List all articles with views from the last twelve months and assign each to the stream or the library.

2. Form topics

   Group the library articles into three to six topics that match the business’s services.

3. Write topic pages

   For each topic, an overview page that introduces the question and links to all articles with descriptive link texts.

4. Merge and delete

   Articles that answer the same question are merged into one. Outdated news items without views may go.

5. Secure addresses

   Every changed or deleted address gets a redirect to the closest page.

6. Assign maintenance

   Every topic page gets a responsible person and a yearly review date.

The redirects are the most delicate step. The article Planning a website relaunch describes the redirect list, which applies here too. Whoever merges articles usually gains more than they lose, because two half-answers no longer compete with each other.

The lesson beyond the case

In the end, the question of blog or knowledge base is a question of how a business understands its own knowledge. As a stream of articles that appear and pass, or as a stock that grows and is maintained. Whoever sees their knowledge as a stock writes differently: less, more thoroughly, with the expectation that an article will still be read in three years. How topics are selected for this is described in the article Editorial plan for articles; how an article lives on in several formats afterwards, in the article Repurposing content.

The limits are clear. The distinction between organization schemes is a thinking model, Krzesaj’s work a literature review without measurement with users, and the study by Wu and Huberman concerns news on one platform, not businesses’ expert articles. How many visitors a rebuild brings cannot be derived from them. What is robust is the conclusion from the model: content that holds for a long time deserves an order that keeps it findable for a long time.

Conclusion and what you can do this week

The guiding question was how a business should organize its expert content, as a blog by date, as a knowledge base by topic or as a mix. The answer: for most small and mid-sized businesses, a knowledge base by topic fits, supplemented by a small section for news. The stream-and-library model separates the two with one question: will the content still hold in two years? Information architecture explains why: an order by date is unambiguous but rarely answers the visitor’s question. An order by topic requires decisions and in return makes answers findable.

• List all existing articles and assign them to the stream or the library.
• Form three to six topics that match the services.
• Write a topic page with links to all articles for the most important topic.
• Link every library article to its topic page and the matching service.
• Merge duplicate articles and redirect old addresses.
• Show the date of the last review on library articles.

A blog shows that a business writes. A knowledge base shows what it knows.