Field notes from building theagency47, frameworks we use in client engagements, and arguments we want on the public record. New post every 2–4 weeks. Follow via RSS.
theagency47's blog is the long-form publication of founder Christos Papadimitriou, covering AI agent design frameworks, engagement methodology, agency-operations lessons, and arguments about the AI services category in 2026. Posts are published every 2 to 4 weeks. Topics span how-to (specs, evals, monitoring), strategy (pricing, positioning, vertical packs), and field notes from real client engagements where the relevant party has consented to share details. All posts are also distributed via the RSS feed.
theagency47 · Updated May 2026For 20 years outbound targeted by industry, size, and location. In 2026 the winning teams target by observable behaviour. Here are the five public signal types that predict a B2B decision better than firmographics.
The 2026 AI agent payback numbers cluster around 4–9 months, but the spread within a function is wider than the spread between functions. Here is the honest per-function breakdown.
Multi-agent orchestration is the loudest agent architecture in the 2026 vendor conversation. It is also the wrong first move for almost every SMB. Here is the failure pattern and the alternative that works.
The average knowledge worker with a good AI agent deployment saves 6.4 hours a week. Senior practitioners save 10-12. Almost none of that shows up on a P&L unless someone measures it deliberately.
The AI agent hype started in engineering. The adoption data says something different: 9 out of 10 real deployments are running finance, marketing, ops, and support workflows. Here's what that means for how SMBs should be buying.
The old assumption was: rank on Google, get cited by AI. The 2026 data destroys it. 28.3% of ChatGPT's most-cited pages have zero Google traffic — and the SMBs winning AI citations are not the ones winning Google.
Two years ago, 'high stakes AI' meant a boardroom fight. In 2026, 62% of SMB leaders are 'very confident' delegating high-stakes work to agents. The trust curve did not happen by accident.
The AI agent market is finally boring. 80% of deployments work. The interesting question is not why the 80% succeed — it is why the 20% fail. Four recurring patterns account for almost all of it.
Knowledge workers lose ~1.8 hours a day hunting for information. The 'second brain' idea promises a fix. Here's what the research says about scaling it from one person to a whole company, and where Obsidian fits.
An eval suite is what separates a demo from a deployment. Twenty test cases, four buckets, a clear rubric, automated re-runs on every change. Here's how we structure it.
Five agents, three firms, real production. The pack works, but not in the order you would expect, and three of the agents needed a redesign by month two.
An AI consultant gives you a strategy deck. AI agents give you working software. Both can be the right buy, but only if you know which problem you actually have.
Four frameworks solve four different problems. The right question is not which is 'best', it is which one fits the agent you actually need, with the lock-in profile you can live with.
Returns cost ecommerce brands $10–$20 in labor per ticket and 4–8 minutes of a support rep's day. An AI returns agent handles 75–85% of them end-to-end. Here's the actual workflow.
ChatGPT Team and Claude Projects sit on a salary. A custom AI agent replaces a workflow. The cost curves diverge after month four, and the ROI math is not what most buyers think it is.
Most AI agent failures come from putting the wrong agent on the wrong work. The 3-tier model (Executive, Operational, Task) splits agents by what they're actually good at.