A thesis before a draft
Why every piece of writing the studio ships starts with one sentence — a thesis the work is willing to defend — and how we kill drafts that cannot produce one.
There is a particular kind of article you can spot from a single paragraph in. It does not have a thesis. It has a topic.
The difference looks small and it is enormous. A topic is what the article is about. A thesis is what the article is saying about that topic. "Pricing pages" is a topic. "Most SaaS pricing pages bury the wrong number" is a thesis. The article with a topic has nothing in particular to argue, so it gestures at the subject for a thousand words and ends with a soft platitude that nobody would disagree with. The article with a thesis has something specific to defend, so every paragraph is doing work: setting up the argument, supporting it, or naming the obvious counterargument and answering it.
We will not draft a long-form piece until we have a thesis sentence we are willing to write across the top of the document. Not as the headline — the headline is a different problem. As the working hypothesis. The thesis is one sentence and it has to satisfy three tests. First, it has to be specific enough that you can imagine someone disagreeing with it. "Good content drives growth" is not a thesis because no one disagrees. "Most B2B blogs would perform better as a single thoughtful page than as a stream of weekly posts" is a thesis because people will disagree. Second, it has to be true. We cannot defend it if we do not believe it. Third, the source — the expert we are interviewing, the data we are reading, the experience the team has — has to actually support it. We have killed thesis statements at this step because the supporting evidence we expected to find did not exist.
The ratio in our studio is roughly one thesis statement killed for every three accepted, in the brief stage. That sounds wasteful and it is the opposite. It is much cheaper to kill a thesis on day one than to write fifteen hundred words of an article that does not have one and discover the problem at the end.
If you take one thing from this note, take this. The next time you commission a piece of writing, ask the writer for one sentence — the thesis — before you read the first paragraph. If they cannot produce it, the piece is not ready to be written. If they can, you can pre-approve or kill the article in thirty seconds instead of thirty minutes of reading.