How should I deploy AI automation solutions for clients—specifically, who pays for APIs/platforms, whose email should be used, and who runs the automation? Please provide a step‑by‑step guide.
The client should cover the cost of any APIs or platforms used in the automation. You set up the automation using the client’s own email address. In most cases, you manage the automation on their behalf—that’s where your value lies. For a detailed walk‑through, check out my video titled 'Nick Sarif, how to manage an automation project'; it shows my step‑by‑step automation agency delivery process, which hasn’t changed in the last few months, so you can follow it directly.
He says clients should pay for APIs and platforms to remove liability and enable affiliate partnerships, which can add 3‑5% to gross margin (e.g., Zapmail 10%, Instantly 40%). This also minimizes handoffs because the client owns their own environment, making transitions easy and avoiding dark‑pattern lock‑ins. For email integration, use a dedicated client‑owned Google Workspace or Outlook account; only handle credentials during a kickoff call when 2FA is required. Finally, recommend containerizing services as a black box: the client pays, you do all the work, they only see input and output (e.g., lead‑gen service where they pay and receive booked meetings). This delivers clear ROI, reduces client stress, and simplifies delivery.
I don’t really do any deliveries. I build everything inside the client’s account. When it’s time to deliver, I just send them an email saying I’m done, plus a video recording of some documentation (maybe a Google doc) if we discussed it, and walk them through how to use the system. That’s delivery in a nutshell. Most people think they need to handle API keys, authorization, etc., but you don’t have to deal with most of that if you build inside the client’s environment. My preferred approach is to run through a kickoff call SOP: collect all needed credentials, access, emails, passwords, then build the automation directly in their Make.com (or similar) account. That’s it.
I don’t just sell workflows; that would make me a commodity provider. When I sell AI agents, I package them as a SaaS product, handling delivery and pricing accordingly. I’ve recorded several videos on the AI automation agency delivery process—search for ‘Nick Sarif manage an AI project’ to find step‑by‑step guides. I’ve also created a dedicated resource for this, which I point people to (you’ll have to Google my name to find it).
On the client's. I'd recommend looking into Claude Routines — a workflow offering Anthropic launched a couple months ago; it's essentially a skill that runs in the cloud on a loop (you set the loop interval, or trigger it via API). It costs more in tokens since you're running AI inference, but it's straightforward. Delivery works the same way as delivering an n8n or Make.com automation: build it in their workspace, then record a walkthrough video showing them the whole system end-to-end so they can run it independently with or without you — it's their IP, and making things stable long-term for the client is part of the job as a service provider. After the delivery video (also a great relationship-building moment), always try to upsell: point out something you noticed while building that could make them more money with a few tweaks, and offer a call to talk through it.
claude codeclaude routinesclient deliveryautomation ownership