Best of the inbox: what the filter actually sees

Best of the inbox: what the filter actually sees
Modern filters are ML models reading engagement, not keyword lists from 2009. The best explainers on how that changes what you optimize.

🤖 Filtering
— A practitioner explainer on engagement-based filtering — opens, replies, deletes-without-reading all feed the model. Made the cut for shifting focus from words to behavior.
— Microsoft/Gmail bulk-sender requirement docs — the one-click-unsubscribe and complaint-rate rules now enforced, not suggested.
— A 'spam words are mostly a myth' teardown — why obsessing over 'free' wastes time vs fixing engagement.

📐 Signals
— A list-engagement-as-ranking-signal piece — the clearest case that your sender reputation is mostly your audience's behavior.

Saved you ~2 hours of editing out the word 'free.' Optimize the behavior, not the vocabulary.