Microsoft's new Copilot looks just like Claude. That's not the part that matters.
Microsoft relaunched Copilot with Home, Code and Autopilot. It looks a lot like Claude, but the real story is distribution: Microsoft is already inside the business.
Remco Livain
I folded twenty thin FAQ posts into one hub page: query augmentation, entity chips and canonical tags instead of a link list
Thirty short FAQ posts sat in Search Console as crawled-not-indexed. Here is how I rebuilt the cluster around the questions people actually type, used Google's own image filter chips as the entity checklist, and consolidated with canonical tags instead of deleting anything.
Would you trust Meta's AI agent with your inbox? My take on the Zuckerberg interview
I watched Joanna Stern's Zuckerberg interview on camera. Why trust is the issue Meta can't duck, why an AI that corrects itself isn't a feature, and why Meta gets too little credit for WhatsApp's encryption.
Our logo was the old domain, set in type: what a rebrand actually costs in code, redirects and brand strings
The migration plan listed nineteen files carrying the old domain. There were thirty-one. Plus 1,176 redirects, 22 brand phrases across 15 files, and one keyword that had to survive the sweep untouched.
The Prime Day freeze: why waiting for the peak costs you more than you think
Prime Days are two days. The freeze before them and the slump after them last three weeks. A model of waiting vs. ramping up early, and why Amazon's 30-day reference price scares brands.
Meta's Muse Charm is an AI Tamagotchi. The real play is the conversation.
Meta put its Muse AI assistant on a keychain. Why gamifying the assistant is clever, why the daily conversation is the prize, and why distribution, not product, decides Meta's hardware.
Amazon blocked Meta's AI shopper. It's not about security.
Amazon cut Meta's Muse agent off its marketplace, citing security. The real threat is to the browsing, ads and impulse buys the marketplace runs on, and what every online shop should do about agentic commerce now.
I raised a client's Google Ads budget 59% while the account could see one lead in sixteen: how to scale on purpose with broken attribution
Impression share lost to budget was 8.2% on weekdays and 14.5% on Sunday — the caps were closing exactly where the demand was. Three conditions that make scaling with broken attribution a decision rather than a gamble.
I tested my new Claude skill against an agent that didn't have it — and the control group found the saving my skill missed
The skill was 35% faster and used 12% fewer tokens. The baseline agent, with no instructions at all, found €10.51 the skill walked straight past. Why the no-skill run is a second opinion, not just a scoreboard.
A price-list download is not a lead: how to measure a channel whose revenue arrives in two years
The first fully traceable enquiry-to-quotation chain I have for a natural-stone supplier: a form fill in August, a phone call four weeks later, and a road surface due to be laid in 2028. Why quotation volume is the leading indicator and revenue is not.