How do I find clients in the really early stages of your AI automation agency? I tried Discord and Reddit at no success for three weeks only 50 bucks per automation. 50 bucks per automation. Any advice? And um, how do I even ship them if I got one?
Thanks for the question. Shipping an automation isn't really needed—you don't need to ship it at all because the shipping happens during the building process. You start by talking with somebody who likes what you have to offer, get them on a call, discuss the problems and solutions, do quick math to see how much the problem is costing them, and pitch your solution at a fraction of that cost—usually around 30%. Once they agree, you send a proposal, they sign it and pay you a little money. I like doing a 50/50 upfront and then upon delivery structure for a small intro offer, then transitioning to a retainer fully paid up front at the beginning of each month. After they sign, you need to get them on a kickoff call to walk through expectations, scope, timelines, and confirm the scope again before starting work—this is your point of highest leverage. If there are project problems, deal with them now while you have some of their money and haven't done work yet, rather than at the end when they're holding money out on you. On the kickoff call, you also have them sign up to the necessary platforms (like Make.com or N8N) so they have everything they need upfront and you never have to bug them again for access; the only messages you'll get are positive happy updates. As a Make.com partner and NN verified creator, I get preferential treatment like a thousand free operations via affiliate promo codes, which adds 3‑5% to your margin because they use your affiliate links and trust you as an authority. You have no liability for recurring payments—they take that on. Before the end of the call, you have all the credentials so you never have to bug them again. That's how you actually go about doing the shipping. To find clients in the really early stages of your automation agency, join Maker School.
I think we should start by clarifying terminology: I wouldn’t consider someone a client until they’ve paid money—or at least committed to paying for something, like a software subscription. Until money changes hands or an agreement is signed, it’s not a real client. Many people think a deal is closed when the prospect says yes, but it’s not truly closed until they sign or, better yet, pay the invoice. Saying yes is different from actually paying you. You said you started automation in November, learned it in two months, and since January you’ve only got two clients in the process of closing. Over roughly 60–70 days, if you’d done a daily outreach minimum—say 10 personalized Loom videos a day via Upwork, DMs, Instagram, X, etc.—you’d have contacted about 700 people. Yet only two showed interest, which is about 0.3% (2 out of 700), or roughly 1 in 350. That’s extremely low; to get one ‘yes’ you’d need to hear roughly 349 ‘nos’. If you’d done consistent, reliable outreach using the mechanisms I show in Maker School and on my channel (daily Loom videos, etc.), even if you were terrible at it, it would be statistically unlikely to get so few results. Typical response rates are around 2%, which would be about eight times better than what you’re seeing. So you likely reached out to far fewer people—maybe around 80 total, or about one per day. To improve, first be very clear about what ‘closed’ means—wait for payment or a signed agreement. Second, set a clear daily outreach minimum, like sending 10 customized Loom videos a day on platforms such as Upwork, Instagram, X, or via DM. Third, standardize your sales process: use a repeatable roadmap (like my ‘Roadmap to $25k a Month with Automation’) that funnels prospects into a meeting, then a sales call, then a detailed proposal or scope of work, and follow the same steps each time. Doing this builds infrastructure you can iterate on—run it for 10 clients, tweak, run another 10, and you’ll see conversion rates improve. That’s what I’d do if I were in your shoes.
Start with a kickoff call after the client pays, where you help them set up any needed platforms and possibly share your affiliate or partner codes for discounts or revenue. Then state that you’ll build everything in their environment so they retain full control if they stop working with you. Build using their tools (they give you credentials), and deliver the project with a video SOP that walks them through the system and serves as documentation for future maintenance. Use that video as a retention mechanism, then upsell related projects based on the knowledge gained from the current job. Mention that you can save them $500–$10,000 per month with a simple system, and invite them to a quick call if interested. The loop gets you in and prints money; you can pitch a monthly upsell or another one‑time project.
Sure. When a new client signs a contract and the invoice is paid, the first step is a kickoff call. In that call you thank them, clarify the problem you’ll solve, reiterate the value you’re delivering, and cover timelines, communication channels, account sign‑ups, and any two‑factor‑authentication requirements. You walk them through signing up for every platform you’ll use, making sure all 2FA is set up, and you aim to have every account, email, and password ready by the end of the call—ideally stored in a password manager like 1Password, LastPass or Dashlane. Then you invite the client to your communication hub (Slack, email, WhatsApp, etc.) and establish a cadence for updates, such as a brief status every couple of days. From there you proceed with the automation build. The client pays for everything: the automation platform (Make.com), any niche tools you use (Naden, PandaDo, Instantly, Mailfinder, ClickUp, Slack, Apollo/Amplify, etc.). Because you have affiliate partnerships with most of these services, you include your referral links, which typically give you a 3‑5 % margin, providing a small side‑income that continues even after the project is handed off.
I start with a kickoff call where I deliver quick wins to minimize buyer's remorse, and I always take payment upfront before any work begins - never considering a client 'onboard' until the money is in my account. After receiving payment, I create a detailed proposal that serves as both a sales document and a roadmap. Roughly 60% of the proposal outlines the client's problems, demonstrating my understanding from any prior small project; about 20% covers the proposed solution; the remaining 20% discusses logistics, principally compensation. I typically pitch a large monthly fee - recently moving from $6,900 to around $12,000-$12,300 - and often add a revenue-share component to align incentives, inspired by the idea that sharing upside makes both parties more comfortable and can lead to more money. I discuss performance-based versus flat pricing, noting that beginners should lean performance-based, intermediates may go flat, and experts often return to performance-based because they have confidence in delivering results. Once the client agrees and pays, I invite them to a kickoff call to set communication expectations: I'm highly reachable via Slack (within 15 minutes) and hold daily office hours from 12 p.m. to 2 p.m. PT, plus a weekly strategy session. I ask clients to commit to the same recurring time each week for at least 50 out of 52 weeks, emphasizing consistency so they take the calls seriously and see ongoing value. During each meeting I review wins from the past week, outline what I'll do next, and address any questions or fires, ensuring I always show up with something tangible to justify the retainer. I then follow the roadmap, checking off the 45-plus tasks in order, and optionally sync with the client's project-management system (though I've moved away from using my own ClickUp board as the single source of truth). I also note that other agencies like DesignJoy use shared boards, but I prefer bringing a quick note of wins and next steps to each meeting.