Skip to content

The Content Audit: How to Learn From What You Have Already Posted

Before you plan new content, spend an hour with what you have already published. It will tell you more than any strategy framework.

By Justin DeMarchiJanuary 21, 20265 min read

Most founders who have been posting on LinkedIn for six months or more have a useful data set they are not using.

They think about content planning as a forward-looking problem. What should I post next? What topics am I missing? What formats should I try?

Looking backward first answers those questions better than any strategy session.

What a Content Audit Actually Is

A content audit is a structured review of what you have published, with the goal of identifying patterns you cannot see post by post.

It is not about finding your best posts. It is about understanding why some things worked, why others did not, and whether your current content is actually serving your positioning or drifting away from it.

Done well, it takes about an hour and produces a clearer picture of your content strategy than most founders have from months of posting.

What to Look At

Start with the last three months of posts. Anything older than that is hard to compare because your positioning and audience may have shifted.

Pull each post into a simple document or spreadsheet with four columns: the topic or category, the format, the engagement level (high, medium, low, using your own judgment), and one sentence describing the main point.

Once you have that list, you can see the patterns.

Which topics are overrepresented? Most founders discover they have been writing about one of their three pillars consistently and essentially ignoring the other two. The imbalance is invisible in the day-to-day posting cadence but obvious in a list.

Which topics are underrepresented? There is almost always a category your ICP cares about that you have not touched. Finding that gap is one of the most actionable outputs of a content audit.

What format are you actually using? Most founders think they are using a variety of formats. The audit usually reveals they use one format for 70% of their posts. Sometimes that is fine. Sometimes it signals a comfort zone that is limiting reach.

The Performance Pattern

Look at the top five performing posts and ask what they have in common.

Not the topics. The structure. Did they all open with a scene? Did they all make a counterintuitive claim? Did they all include a specific number? Did they all have a clear, direct close?

Then look at the five lowest-performing posts and ask the same question.

The pattern between top and bottom performers is usually more structural than topical. The same subject matter can produce wildly different engagement depending on how it is framed. The audit surfaces the structural patterns that are producing your results. For a deeper framework on evaluating individual posts, see how to know if a LinkedIn post is actually good.

The Positioning Drift Check

Read your last thirty posts as if you were a potential client seeing them for the first time.

Ask one question: what would someone conclude about what this person does and who they serve?

If the answer is specific and accurate, your content is doing its job. If the answer is vague or if it points to a version of your positioning that is six months out of date, you have drift.

Positioning drift is extremely common. A founder refines their ICP, shifts their angle, or develops a new area of expertise. The content often lags behind by months because new thinking takes time to work its way through into what gets posted. This drift is one of the structural reasons B2B content fails.

The audit makes the drift visible so you can close the gap.

What to Do With the Output

An audit should produce three to five concrete decisions, not a full content strategy overhaul.

Pick one underrepresented topic to add to your rotation. Change one structural habit that your bottom-performing posts share. Identify the one format you have been avoiding and commit to trying it once in the next month.

Small adjustments compounded over time produce better results than a complete strategy reset every quarter. A regular audit is one of the maintenance practices that keeps a founder-led marketing system performing over the long term.


Frequently Asked Questions

How often should I do a content audit?

Every three to four months is a reasonable cadence. That gives you enough new content to generate meaningful patterns without letting the drift go unchecked for too long. Some founders do a lighter review monthly and a deeper audit quarterly.

What tools do I need for a content audit?

None that are specialized. Your LinkedIn activity feed shows your posts with engagement data. A simple spreadsheet or document to organize the information is enough. The value is in the pattern recognition, not in the tools you use to capture it.

What if I have not posted enough to do a meaningful audit?

Twenty to thirty posts is a reasonable minimum. Below that, the patterns are not statistically meaningful. If you are newer than that, focus on building the volume first. Run the audit once you have two to three months of consistent posting behind you.

Should I delete underperforming posts after the audit?

No. The audit is for learning, not for pruning. Old posts get almost no views and do not affect your current reputation. Deleting them removes the historical data you would need for future audits. Leave the archive intact.

Can a content audit help me identify topics I should stop writing about?

Yes. If a topic is consistently low-performing and not strategically necessary to your positioning, the audit gives you permission to deprioritize it. The insight comes from seeing the pattern across multiple posts rather than judging any single one. One underperforming post on a topic proves nothing. A consistent pattern across eight posts on that topic is a real signal.

Justin DeMarchi
Written by

Justin DeMarchi

Senior B2B operator and founder of DUO. Eight-plus years running marketing and content systems for brands in tech, SaaS, and AI.

More in