The Medlar blog
Notes on AI that does the work.
Practical writing on AI operations coworkers, agency automation, lead research, and building AI you can actually trust.
Choosing your first AI automation: a 15-minute audit
You don't need a strategy. You need one good task. A quick way to find the highest-leverage thing to hand over first.
Read →Cold outreach that isn't spam: personalization at scale
Generic blasts get ignored. How an AI coworker researches each prospect so every message is specific — and you approve each send.
Read →Turning meeting notes into action with AI
The insights from a meeting are worthless if nobody acts on them. How to go from recording to owned tasks automatically.
Read →CRM hygiene on autopilot
A messy CRM costs you deals. How an AI coworker keeps contacts, companies, and deals current without the manual upkeep.
Read →How to measure the ROI of an AI coworker
Hours reclaimed is the obvious metric. Here's a simple way to know whether an AI coworker is actually paying for itself.
Read →Data privacy and AI coworkers: what to know
Connecting an AI agent to your business data is a trust decision. The questions to ask — and the answers that should reassure you.
Read →From solo operator to a small team: scaling ops with AI
The work that used to justify your first ops hire can now run on an AI coworker — so you hire for growth, not for grunt work.
Read →Connecting your stack: Gmail, Slack, Sheets, and beyond
One agent that reaches across all your apps beats a separate bot for each. How the connected-app model works — and why it scales.
Read →Competitor benchmarking with AI
Find out where a site really stands by comparing it to living, successful peers — not a list of dead domains.
Read →Credit-based pricing, explained
Why an AI coworker is priced in credits, what a credit buys, and how top-ups keep heavy weeks flexible.
Read →AI transparency: why every step should be logged
A black box that gives you an answer isn't good enough for real work. Here's why the log is the product.
Read →Recurring reports that write themselves
The highest-leverage automation is the one you set once. How scheduled AI runs quietly do the weekly work.
Read →How to brief an AI agent: the one-sentence skill
The better your brief, the better the result. A short guide to writing sentences an agent can actually run.
Read →AI coworker vs Zapier: when to use which
Rule-based automation and AI agents solve different problems. Here's the honest dividing line.
Read →SEO audits in minutes: what to check and fix
The on-page and performance signals that actually move rankings — and how an AI coworker surfaces them with ready-to-paste fixes.
Read →Inbox triage with AI: reclaim your morning
The inbox is the tax you pay before real work starts. An AI coworker can sort, route, and draft — you just approve.
Read →The case for human-in-the-loop AI
Autonomy is exciting until an agent sends the wrong email. Approval gates are what make AI safe to actually use at work.
Read →Lead research on autopilot: finding and enriching prospects
Prospecting is search plus enrichment plus judgment. An AI coworker can do the first two at scale and leave the judgment to you.
Read →How agencies can automate client reporting with AI
Weekly client reports eat hours of skilled time. Here's how to hand the whole loop to an AI coworker — with your approval on what goes out.
Read →What is an AI operations coworker?
Not a chatbot, not a workflow builder — an agent you brief in a sentence that does real back-office work through your apps.
Read →