Every Monday I used to spend about 90 minutes checking analytics. YouTube Studio, Instagram Insights, LinkedIn Analytics, Ghost's built-in dashboard, and a separate Google Analytics property for my personal site. Each platform has its own metrics definitions, its own date range defaults, and its own idea of what "impressions" means. None of them agree.
The frustrating part wasn't the time. It was that after 90 minutes I couldn't tell you which piece of content had performed best that week in any meaningful cross-platform sense. I had five separate numbers for five separate platforms with no common denominator. The review session produced data but not insight.
Why platform analytics are designed to be platform-specific
YouTube, Instagram, LinkedIn, and TikTok all have strong reasons to keep you inside their own analytics environments. The more time you spend in their dashboards, the more you understand their platform-specific signals — and the more you optimise for their platform specifically. This is good for the platform. It is not especially good for you as a cross-platform creator.
Platform-native analytics also define metrics in ways that serve their advertising products. YouTube's "impression click-through rate" measures how often people click your video after seeing the thumbnail in YouTube's recommendation feed. This is valuable for understanding your thumbnail performance but tells you almost nothing about the total reach of your content. Instagram's "reach" includes accounts that scrolled past your post for 0.1 seconds. LinkedIn's "impressions" count once per person per post, but their algorithm can show the same post to the same person multiple times on different days — a nuance buried in the footnotes of their measurement methodology page.
These aren't failures of measurement. They're accurate measures of platform-specific signals. The problem is that none of them translate across platforms, so comparing performance is almost impossible without a normalisation layer.
The two metrics that actually matter across platforms
After two years of cross-platform analytics reviews, I've settled on two metrics that hold meaning regardless of platform: engagement rate and content-to-conversion rate.
Engagement rate — total meaningful interactions (comments, shares, saves, link clicks) divided by total reach — is comparable across platforms when you're consistent about what you count as "meaningful." I exclude likes and reactions; they're low-intent signals that don't predict anything useful. Comments and shares are the engagements that indicate actual connection with the content.
Content-to-conversion rate is how often a piece of content drives someone to take a downstream action — subscribing to your newsletter, signing up for your product waitlist, clicking through to a sale. This requires UTM tracking on every link and a destination you control, but it's the metric that actually connects content output to business outcomes. Everything else is vanity metrics with varying degrees of usefulness.
How Deaku's unified analytics view works
Deaku connects to your channel APIs and normalises the underlying engagement data into a single time-series view. The platform doesn't attempt to create a single "score" — that would require choosing weightings that suit some creators and not others. Instead, it shows a per-piece breakdown: for each content item, across each platform where it was published, you see reach, engagement rate, and link clicks in a single row.
The comparison view becomes useful when you've published the same topic across platforms. When a piece about financial independence performs 4x better on YouTube than LinkedIn, that's actionable signal — it tells you something about where that topic resonates. When an opinion-format post outperforms a listicle format consistently across all platforms, that's an even stronger signal about your audience's preferences.
Deaku's analytics surface these patterns automatically after about eight weeks of connected data. The pattern cards appear in the insights panel: "Your tutorial-format videos outperform your opinion-format videos by 2.3x on YouTube, while the reverse is true on LinkedIn." That cross-platform pattern recognition is what takes 90 minutes manually and five seconds with unified data.
Revenue per post: the metric most creators skip
If you have any revenue connected to your content — sponsorships, affiliate commissions, product sales, newsletter subscriptions — revenue per post is the most important metric you're probably not tracking.
Revenue per post requires connecting your monetisation data to your content data, which most platforms make intentionally difficult. YouTube's revenue data lives in YouTube Studio. Affiliate commissions live in your affiliate network's dashboard. Sponsorship revenue lives in your inbox. Pulling these together manually is a project; most creators skip it entirely and optimise for views instead.
The problem with optimising for views is that views and revenue often diverge. A 200K-view video monetised through mid-roll ads might generate less revenue than a 40K-view tutorial where 12% of viewers click an affiliate link with a 30% conversion rate. The view count tells you one thing. The revenue per post tells you whether that content is actually building your business.
Setting up a useful weekly review: a practical framework
The goal of a weekly analytics review is to answer two questions: what worked, and why. The "what worked" part is quantitative — you're looking at engagement rate and, if possible, revenue per post. The "why" part requires your own judgment — platform algorithms, publication timing, the topic's relevance in the current news cycle, and the quality of execution all contribute.
Keep the review to 20 minutes. If it takes longer, you're looking at too many metrics. Start with a top-three list: which three pieces got the best engagement rate this week? For each one, write one sentence about why you think it performed. That hypothesis, built up over weeks, becomes your personal publishing theory — a model of what works for your specific audience. No analytics tool can build that model for you; they can only give you the data to test it against.
The patience problem with content analytics
YouTube videos peak in performance weeks or months after publication. LinkedIn posts peak within 24–48 hours. Newsletter open rates peak within the first hour. These are completely different time horizons, which means a seven-day analytics review genuinely misses most of YouTube's signal.
Deaku handles this with a 30/60/90-day rolling view alongside the weekly snapshot. A piece that looks average in the first week might show strong long-tail performance at 30 days — particularly tutorial content and evergreen opinion pieces that get discovered through search. Tracking both timeframes simultaneously is what separates platform-native analytics from a genuine content intelligence tool. The weekly data answers "what should I make this week?" The 90-day data answers "what format should I focus on this quarter?"