Keywords don’t help organize information; topics do. Keywords can reveal different terms used to search for the same result, but they don’t capture the semantic relationship in the way that topics can.
If you’re like most content marketers, you’ve learned to view Search through the eyes of keyword research. That’s understandable given the history of Google.
In the past, the search engine focused on matching keywords on a page to a search query. However, Google continues to evolve, and what was valid five or ten years ago is no longer true. Take the case of keywords, a topic held in high regard by SEOs.
They are no longer the driving force behind Search. As Google’s John Mueller explained, “search engines will get better over time to understand more than just the words on a page.”
So, it’s time we have a serious conversation about the difference between keywords and topics. As you’ll soon discover, there’s a substantial variation in how they are used for content marketing.
What Is a Topic?
According to Oxford, a topic is “a matter dealt with in a text, discourse, or conversation; a subject.”
Let’s talk about my favorite subject, “apple pie.” There’s a lot to discuss, such as:
- Nutritional facts
- How to bake an apple pie
- Pie dough
- Baking a pie crust
- Parchment paper
- Pie plate
- All-purpose flour
- Ice cream
There’s so much more we could include in a discussion. When we put the subject into MarketMuse, this topic tool found 50 highly-related topics associated with apple pies.
Semantic relevance is key to understanding the narrative that this group of words tells. It’s also what sets apart topics from keywords, which we’ll get to in a moment.
What Is a Keyword?
Keyword is a term we often substitute for search terms. Oxford states that a keyword is “a word or concept of great significance.”
Once more, let’s take my favorite subject and plug that into a popular keyword tool to see what we get. This software offers four choices for the types of keywords returned:
- Broad match (variations of the original phrase in any order)
- Phrase match (the exact phrase in various orders)
- Exact match (use the keyword string in the exact order)
- Related (similar to the see keyword)
The first three choices are quite restrictive in that they require our initial phrase (apple pie) to be in the result. There’s no notion of semantic relevance there! The last choice offers a bit more flexibility. In theory, related keyword results don’t need to contain the initial phrase.
Let’s see how this looks.
The screenshot above shows the top ten results ordered by relevance. There are over 1,500 keywords that are used as search queries, meaning they all have a minimum monthly search volume. But let’s focus on those that are the most closely related.
You’ll notice that nine out of the top 10 contain the seed term “apple pie.” For those who are curious, I looked at the other 1,500 keywords in that list. What I found was that the majority contain either the term “apple” or “pie.”
What Is the Difference Between a Topic and a Keyword?
Let’s revisit the formal definitions of topic and keyword to see if there’s a discernible difference. Here are the two definitions side-by-side. What do you think?
Call it a matter dealt with in a discourse or a concept of great significance. To many, there may be no difference; it’s a matter of tomato vs. tom-ah-to.
Fair enough. But look at the difference in output between the two approaches.
Imagine writing an article about apple pie. Most likely, you’d offer a recipe including instructions on how to make the dough, how to slice the apples along with instructions on using an egg wash to make the crust shiny.
And let’s not forget the cinnamon or that it’s best served with a scoop of ice cream on the side!
With the keyword approach, you’ve got a list of phrases like:
- Apple pie pie recipe (no this is not a grammatical error)
- Apple pie com
- Apple pie recipe
- Apple pay recipe
How are you going to create content from that? (That’s a rhetorical question.)
You’re not going to optimize your post to rank for a bunch of keywords by adding more of them. You’re going to rank for a specific topic by covering them in the normal course of conversation.
Can you rank for multiple topics?
Absolutely. Although it depends on how close the topics are and whether you’ve covered any additional topics related to the subject.
Keyword Research vs Topic Research
Using a keyword tool for the topic “guitar types” will return a list of related keywords that are simply different ways of expressing the same search term. Here’s an example from a popular keyword research tool.
Here you’ll discover that most people use the search term “guitar types,” quite a number use “type of guitars,” while few use the term “different types of guitars” and even less so for “different kinds of guitars.” While all these may seem to be linguistically similar, you would need to verify by comparing the SERP results.
All this isn’t very helpful if you’re looking to create content around the topic of “guitar types.” And if you’re not careful, this can get you into a lot of trouble. In the past, marketers would create separate pages targeting each of these phrases with no regard for whether they served the same intent. This is where the whole keyword cannibalization problem started, with different pages competing against each other.
Occasionally, this type of research will surface something usable, such as “types of guitars for beginners.” But that’s more the exception.
Type “guitar types” into MarketMuse or another topic modeler and you’ll get related topics like “acoustic guitar,” “electric guitar,” “bass guitar,” “classical guitar,” “steel guitar,” “Les Paul,” “flamenco guitar,” etc. Now you have a map of topics around which to create content. Some of these topics will require multiple pages, in which case you can apply this process recursively. In the case of MarketMuse, it presents a list of related topics sorted by semantic relevance, which can help prioritize your content creation.
Why You Need to Start Thinking About Topics Instead of Keywords?
Keyword research doesn’t help create better content, and a bigger list won’t solve the problem. Fifteen hundred keywords (most of which contain either “apple” or “pie”) don’t help. Instead, they hinder the creation of expert-level content.
Topics help you focus on the critical matters that are dealt with in a conversation on the subject. Keywords? Not so.
That’s why writers struggle to fit keywords into their narrative. The goal isn’t to optimize keyword usage, it’s to ensure thorough coverage of a topic.
If that’s not enough to convince you, then take a moment to think about Google’s mission statement, which is to “organize the world’s information and make it universally accessible and useful.”
Last year Google announced a new “topic layer in the knowledge graph.” Notice that they don’t refer to it as a “keyword layer.” Is anyone willing to venture a guess why?
Here’s what a topic layer looks like. Available on mobile searches since last year’s announcement, it’s starting to roll out to desktop queries.
Google has organized the SERPs into several topics, all related to the queen of soul herself, Aretha Franklin. If you’re writing content on this subject or any subject for that matter, you will do well to take a topical approach
Theoretically, keywords and topics should be the same. But in practice, they are like night and day. On-page search engine optimization focuses heavily on keyword phrases. Unfortunately, keyword research tools don’t help you deal with those essential matters in your content. For that, you need a platform built from the ground-up with a topical focus.
Written by Stephen Jeske