The Graph That Should Be Front-Page News

Every so often the Earth produces a signal that is impossible to ignore. This graph is one of them. It shows sea-surface temperatures in the Niño 3.4 region of the equatorial Pacific, one of the most important parts of the Earth's climate system. Each blue line represents a different year since 1982. The red line is this year. It doesn't just set a new record. It has departed entirely from the range of previous observations.If this graph represented stock market prices, a new Olympic record or a

From Feeds to Trails

To design the future of social media, rethink the interface before the algorithm.

How the Substack feed is learning to understand your reading journey

Modeling sequences of user behavior makes discovery feel alive

Permissioned Data Diary 4: The Big Picture - Daniel's Leaflets

A special edition of the data diary that sketches out the rough shape of where we're heading.

Bluesky feed engagement — image-count analysis

The left chart tiers authors by actual follower count (at analysis time). The right chart tiers authors by median likes per post — a proxy for effective reach. Interesting divergence: under the follower definition, medium-size accounts show no 4-img penalty; under median-engagement, the pattern is different.

Here’s a metric that I bet you aren’t using… I want you to check it out, and start using it. It could be a real wake up call about how inactive your users are. I’m talking about the XD28 metric. A… | Moe Ali | 17 comments

Here’s a metric that I bet you aren’t using… I want you to check it out, and start using it. It could be a real wake up call about how inactive your users are. I’m talking about the XD28 metric. A lot of teams have never bothered to collect this data - but you can get started after 28 days. Here’s how the metric works👇 This metric looks at a user and determines how many days they used your product/feature in a 28 day cycle. If your user used your product just once in the last 28 days, they’re identified as 1D28. If they use it 5 days out of 28, they’re categorized as 5D28 and so on… Then you create a bucket with each number of days and how many users used the product for each number of days. For example, 17 users belong in the 5D28 bucket and 500 users belong in the 24D28 bucket. You can use this to create a distribution chart. You should be using these figures to calculate the average AND median number of days that users are using the product in a given 28 day cycle. Ideally you want people to be moving from a lower number of days to a higher number. E.g. If your average number of XD28 is 5 days, you want to aim for 15 days in a year’s time. This is an interesting metric because often you will find users at one extreme of this distribution curve. And it can be a real wake up call for Product Managers. | 17 comments on LinkedIn

it's my following and i want it my way! - matt's not blog

a late response to the social graph talk from last week

bluesky visualizers

See the connections between the last 150 people you followed (if they followed each other as part of their last 150 follows)

bsky charts (@bskycharts.edavis.dev)

posting unofficial charts of bluesky network activity 🤖📈 more at https://bskycharts.edavis.dev/bluesky-day.html chart guide: https://edavis.dev/notes/bskycharts-guide/