Method · Measurement

How to Measure If Your Content Is Working: Two questions, not twenty metrics.

Menno Kater Menno Kater · July 27, 2026 · 7 min read
TL;DR

Almost everything you're told to track predicts nothing. There are two questions worth asking: did people stay (retention and watch time), and did anyone move toward you (a DM, a reply, a warm sales call). Views and likes are weather. Signals are people. Kill the dashboard and run one monthly review built on a single question: what produced a signal?

Views are a scoreboard, not a signal

Open your analytics and you'll see a wall of numbers: views, impressions, reach, likes, followers, engagement rate. Every tool wants you to watch them, because a moving number feels like progress. It rarely is. A post can do 40,000 views and produce nothing. A post can do 900 views and put a client in your inbox. If the number that went up doesn't tell you which of those just happened, it isn't measuring what you actually care about.

Views are weather. They tell you it was sunny on Tuesday. They don't tell you whether anyone who saw you decided you were worth working with. For a founder building a personal brand, that decision is the entire point, and it lives almost nowhere on the standard dashboard. So the first move isn't to track more carefully. It's to stop treating attention as the outcome. Attention is the raw material. The outcome is a person moving toward you.

Views are weather. Signals are people. Only one of them pays you.

The two questions that actually matter

You can replace the whole buffet with two questions. First: did they stay? Of the people who started watching or reading, did a meaningful share get to the end. That's retention, and it tells you whether the thing was worth someone's time. Second: did anyone move toward you? A reply, a DM, a save, a follow that turns into a message, a sales call that opens with your name. That's the signal that predicts business.

Everything else is downstream of these two, or noise. Reach is only useful if the people you reached stayed. Followers only matter if some of them eventually move. Judge every piece of content against these two questions and most of your anxiety about "is this working" dissolves, because you finally know what you're looking at. If you're also wondering how big your audience needs to get before any of this pays off, we wrote about that separately in how many followers you actually need to get clients →.

Retention and watch time: the only 'reach' number worth reading

If you're going to read one platform metric, read retention. On YouTube it's average percentage viewed and average view duration. On short-form it's the completion rate and the drop-off curve. This is the leading indicator, because it answers the honest question underneath all the others: was this actually good enough to finish?

A high view count with a cliff in the first ten seconds means the algorithm pushed you and people bailed. That's not a win, it's a warning. A modest view count where people watch to the end means you made something worth someone's time, and the reach usually follows. Retention leads; views lag. When your views feel stuck, retention is almost always where the problem is hiding, which is why we treat it as the first thing to look at in why your views aren't growing →.

You don't need a chart for every video. You need to notice the pattern: which topics hold people, which formats lose them, where the curve drops. Make more of what holds. That single habit does more than any dashboard.

The signals that predict clients

Retention tells you the content was worth watching. Signals tell you it was worth acting on. These are the human responses that reliably precede money, ranked roughly by how much they mean:

  • A direct message. Someone took the effort to open a private conversation with you. This is the highest-value signal there is, even when it's just a question.
  • A reply that shows they understood. Not "great post" but a comment that references your actual point, or disagrees with it. That's someone thinking, not scrolling.
  • A save or a share. They filed you away or handed you to someone else. Quiet, but it means you were useful enough to keep.
  • A sales call that opens warm. The person on the call already trusts you before you've said a word. This is the one you're really building toward.

Notice what's not on this list: likes. A like costs nothing and predicts nothing. Track the four above and you're tracking the things that actually turn into revenue.

Why your best client will never show up in your analytics

Here's the part nobody tells founders. Your best client usually arrives with a sentence that no analytics tool can attribute: "I've been following you for a while." They watched twenty videos, never liked one, never commented, never clicked a link you can trace. Then one day they booked a call already sold. There is no row in your dashboard for that person. The single most valuable outcome your content produces is, by its nature, invisible to the software.

This is the attribution gap, and for a founder brand it's not a rounding error, it's the main event. The generic advice tells you to track conversion rate as if the buyer's journey were a tidy funnel with clickable steps. It isn't. Trust builds silently over months, then converts all at once. If you only trust what the tool can attribute, you'll conclude your content isn't working while it's quietly doing the one job that matters.

You close the gap with one question, asked on every sales call: "How did you first hear about me, and what made you reach out now?" The answers are your real analytics. When people keep naming a specific video, a specific format, or "your stuff kept showing up," you've found what's working, no dashboard required. This is exactly how our own case reads. Two founders grew a community from $12.5K to $80K per month over seven months, almost all of it organic, mostly through YouTube and Instagram, going from around 50 to roughly 320 members. That growth was signal-driven, not view-driven. You can read the full $12.5K to $80K per month story → if you want the mechanics.

The monthly review: one question, one action

Daily analytics-checking is a nervous habit, not a strategy. Content works on a slower clock than the dashboard refreshes. So check on the clock that matters: once a month, sit down for twenty minutes and answer one question. What produced a signal? Which posts got DMs, thoughtful replies, saves, or a call that opened warm. Line them up next to which posts held attention to the end.

Then take one action: make more of what produced signals, less of what produced only views. That's the entire method. Not a scorecard, not a KPI deck, just a repeated bet on the things that moved real people. Over six months this compounds into a content style that consistently produces buyers, because you've been steering by the one variable that predicts them.

Here's what to keep and what to drop:

Stop watchingStart watching
Total views and impressionsRetention / average view duration
Likes and follower countDMs, thoughtful replies, saves
Engagement rate"How did you hear about me?" on calls
Daily analytics checksOne monthly review, one question

The small dashboard that's actually enough

You don't need software for this. You need a note with two columns. One column: the pieces that held attention this month, pulled from retention. The other column: the pieces that produced a signal, pulled from your inbox and your call notes. Where a piece shows up in both columns, that's your template. Do more of that next month.

That's the honest dashboard. It fits on an index card, it can't be gamed by a viral fluke, and it points straight at revenue. Everything the big guides pile on top of it is there to make measurement feel rigorous, not to make it true. If you want the same discipline applied to what you actually publish rather than what you measure, that's the work we do with founders, laid out in see how we work with founders →.

And if you would rather have this measuring done for you: here is how a personal branding agency reports on what converts.

How do I know if my content is actually working?
Ask two questions instead of tracking twenty metrics. Did people stay, measured by retention and average watch time, and did anyone move toward you, measured by DMs, thoughtful replies, saves, or a sales call that opens warm. If both are happening, it's working, even when your view count looks unremarkable.
What content metrics actually matter for getting clients?
Retention is the leading indicator, because it tells you whether the content was worth someone's time. The lagging indicators that predict revenue are human signals: direct messages, replies that show real understanding, saves, and calls where the person already trusts you. Views and likes cost nothing to give and predict almost nothing.
Why can't I track where my clients come from?
Because the highest-value buyer usually arrives having watched you silently for months without ever liking, commenting, or clicking a trackable link. Trust builds invisibly and then converts all at once, so no analytics tool can attribute it. You close that gap by asking every new client how they first heard about you and what made them reach out now.
How often should I check my content analytics?
Once a month, not once a day. Daily checking is a nervous habit, and content works on a slower clock than the dashboard refreshes. Spend twenty minutes monthly answering one question, what produced a signal, then make more of that and less of what only produced views.
Are views a vanity metric?
On their own, yes. A high view count with a steep early drop-off means the algorithm pushed you and people left, which is a warning, not a win. Views only matter when paired with retention, and they only become valuable when some of those viewers turn into signals like a message or a warm call.
Stay in the loop

New cases and systems in your inbox, a few times a month. No funnel tricks, unsubscribe anytime.

Want a second pair of eyes on whether your content is actually producing business?

Tell us what you're posting and what's coming back, and we'll tell you honestly whether it's working, even if that's "not yet."

Let's talk