Category · 219 posts
"AI for business" usually means one of three things: vendors pitching $200K platforms, consultants selling 60-page strategy decks, or operators wondering whether any of it actually works. This cluster is the honest middle path.
Practical guides on AI implementation, automation, and consultant vetting — written for $1M-$50M operators who want to turn AI into infrastructure their team actually uses every day. No "AI transformation" buzzwords. Specific costs, real workflows, vetting questions that filter out the bad consultants, and the cases where AI isn't ready yet.
Quick Answer
Curated by Justin McKelvey · 219 posts in this cluster · Updated October 2026
The honest answer on AI property valuation in 2026 — how AVMs actually work, where they fall apart (condition, renovations, micro-location, thin markets), how AVMs differ from a CMA and an appraisal, what AI really does in mortgage underwriting, and the draft-and-approve valuation workflow for agents.
Not the coolest workflow — the highest hours per week × repetition × low judgment. A scoring framework you can run yourself in 30 minutes, worked examples of what wins (intake and follow-up) and what always loses (pricing, hiring), the first-automation candidate by business type, and what not to touch first.
AI for bookkeeping in 2026 — the honest answer on whether it can replace your bookkeeper, the four things it genuinely does well, the three ways it quietly goes wrong, what the draft-and-approve setup actually costs in money and weekly minutes, and what this means if you're the bookkeeper.
Executives don't need to learn AI. They need to make five decisions about it: what to fund, what to refuse, who owns it, what the board gets told, and what 'we use AI' means in the P&L. Everything else belongs to the people who run the workflows. A fractional CTO's version of the executive AI briefing, without the slide deck.
Leaders keep asking which AI tool to learn. Wrong question. Your job is deciding what the company uses AI for, what it never uses it for, how work gets re-scoped when a task drops from days to minutes, and how you judge output you can no longer produce yourself. The decision rules, the meeting habits, and what to stop measuring.
A manager's AI job is not prompting. It's deciding which of your team's tasks AI takes, setting the review standard for what comes back, and re-scoping people's work when a two-day task becomes a twenty-minute one. Here is the weekly practice, the tools that matter, and the mistakes I watch managers make.
As of September 2026: Opus 5.5 $4 in / $20 out per million tokens (launched September 22), Sonnet 5 $2 / $10, Haiku 4.5 $1 / $5, Fable 5.1 $10 / $50, cache hits at 10% of input on most models, Batch API half price. The rate table is the easy part. What three real workloads cost, the tokenizer change that quietly inflates bills, and the price cut Anthropic made permanent.
Yes — Qwen Chat is free with no paid consumer tier at all, and the open-weights models are free to run yourself. What ended in April 2026 was the free developer API, replaced by a one-time 70-million-token trial. What free covers, what the API costs, and the data question a US business should ask first.
A Claude Skill is a folder with a SKILL.md file: a description that tells Claude when to use it, and instructions it follows when it does. Here's how to write one, where it lives in Claude Code and Claude.ai, how Claude decides to load it, and the three skills every business owner should build first.
Microsoft already signed the BAA — it's in the Data Protection Addendum you accepted at purchase, and Microsoft 365 Copilot is named on the in-scope list. So the contract isn't your problem. Two other things are, and Microsoft prints one of them in a footnote.
No AI model is HIPAA compliant. A signed Business Associate Agreement is — and Anthropic won't sign one for Claude Pro, Max, or Team. As of October 2026 it covers the API and Enterprise plans, once an owner switches it on. Here's exactly which tiers of Claude and ChatGPT are covered, which ones aren't, and why the business plan you bought is probably the wrong one.
Anthropic doesn't publish a message count for any Claude plan — not Free, not Pro, not Max. What it publishes are multipliers and two meters. Here's how the 5-hour window and the weekly cap actually work, what each tier is reported to get, and what to do the third time you hit the wall.
Yes, genuinely — n8n's self-hosted Community Edition is free with unlimited executions, which almost nothing else in automation offers. The bill just moves somewhere else. Here's what free costs in practice, what n8n Cloud runs, and how to tell which one you actually want.
Yes. DeepSeek's chat app is free with no paid tier at all — no Plus, no Pro, nothing to upgrade to. That's rarer than it sounds, and it's why the interesting question isn't the price. Here's what free actually costs a business, and the one version of DeepSeek that's free AND private.
Most Claude Projects tutorials are feature tours. Here's the part that actually pays: a Project is a place to park the context you re-type into a chat box every single day. Set one up properly once and every future conversation starts already knowing your business.
AI Automation
4 posts
AI Design Tools
6 posts
AI Visibility
15 posts
AI for Founders
3 posts
Business Growth
4 posts
Founder Sales
2 posts
Fractional CTO
14 posts
Fractional Product
1 posts
Go-to-Market Strategy
3 posts
Lean Software
2 posts
Pricing Strategy
2 posts
Product Leadership
6 posts
Vibe Code Rescue
76 posts
Book a free 15-min strategy call. Specific recommendation, no pitch.
Book a Call