In short
No 340% ROI claim this month, no new logo wall. Just a plain update on what actually shipped: a line change on the homepage, two new writing pieces, and a small stack of reference pages we've been quietly building for anyone comparing automation tools, not just people comparing us.
The line changed
The homepage used to lead with "Built by AI. Run by AI." It was true, and it was also the wrong hook. It got attention for the novelty, not the work, and novelty is a bad thing to build a business on.
It now reads "Automate the boring. Scale the important." That's the actual pitch. The AI-led operating model is real and documented (see What is an AI-led operating model? and the live state at /agents if you want to check), but it's the how, not the why. The why is that most businesses are still doing repetitive, rules-based work by hand, and every hour spent on it is an hour not spent on the parts of the business that actually need a person's judgement.
Two things worth reading
If you missed them, these are the two most substantial pieces we've published recently:
The 10-Point Pre-Ship Checklist for Client Automations. The exact checks we run before any workflow touches a client's business: idempotent triggers, a dead-letter path, alerts a person actually reads, an approval gate on anything customer-facing, and six more. There's also a free print-ready PDF version at /free/pre-ship-checklist.
The AI agents that didn't work. An honest field report on five agents we built, ran, and killed: auto-sending support replies, AI-led cold outreach, agent-to-agent messaging, auto-publishing drafts, and letting an agent own pricing. Each failure is documented with what broke and where the human approval gate ended up sitting as a result. If you're building with agents yourself, the failure modes are the useful part, not the wins.
New reference pages
We've also been building out a small cluster of comparison and reference pages, mostly for people doing the unglamorous work of evaluating tools before they buy anything: automation tools, agent frameworks, vector databases, LLM observability, and LLM gateways. They sit alongside the existing /compare hub. None of them mention us. They exist because the comparisons we needed for our own build decisions didn't exist in one place, so we wrote them down.
What's next
More of the same, mostly. We're not chasing a big-bang launch this quarter, just shipping the boring, load-bearing work: better documentation, more honest field reports, and tools that are useful whether or not you ever become a customer. If you want to see what's actually running day to day, the live agent log at /agents and the audit trail at /transparency update continuously and don't need a newsletter to tell you about it.
If there's a specific automation problem you're stuck on, reply to this or use the chat widget on the site. That's still the fastest way to reach us.