I've been building AI customer service systems, and most clients came from Upwork and referrals — I've got a case study for a big property management company where a system led to big savings, and I built others for medical practices too. My issue is generating demand myself outside of referrals: is this sellable via cold outreach, and can it be turned into a front-end offer? I've sent thousands of cold emails with only one booked meeting, plus LinkedIn DMs and content, and I don't know if that's the right play or if I should pivot.
Right now you're focused on the savings, and I think that's the main thing you've understood automations to provide, but savings is a small fraction of the total value of an automation. Say you build a system that saves 10 salespeople 1 hour a week each — that's 40 hours a month, and if they're paid $50/hour that looks like a $2,000/month automation. That looks sizable, but it's only a small part of the value you're capable of providing. The big thing isn't the savings, it's the revenue: if you free up an hour of a salesperson's time and they can use it to close deals worth, say, $500 in that hour (amortized across their whole workload), you're not just saving $2,000, you're making them an extra $20,000. That's the difference between a savings mindset and a revenue mindset — it's opportunity cost. When I sell a cold email system, I don't say 'this saves you $50/month on cheaper leads,' I say 'this frees up 10 hours a month of your salespeople's time, worth $500/hour, meaning it makes you $5,000.' Reorient your pitch around revenue, not savings — savings only matters if you're working with a multi-billion dollar enterprise. On your second issue: 3,000 emails to one booked meeting is a 0.03% booked-meeting rate. Two meetings would be 0.06%, three would be 0.1% — that's totally fine, not the best ever (ideally 0.5-1%, a meeting every ~200 emails), but it proves it's possible. You've run the equivalent of a 10-minute mile; now make your copy better and run a 9-minute mile.
My view on AI consulting is that there’s no real difference—it’s just consulting repackaged. Whenever you see terms like AI automation, AI consulting, AI growth systems, or AI operators, they’re all the same thing; people are just trying to sell a business model and differentiate themselves in a crowded space, which in content is called packaging. At its core, this is foundational marketing: nothing has really changed since the SMMA craze; we’re still selling marketing systems, just rebranded as AI automation. Most AI automation sellers are actually selling marketing systems—content repurposing, social media automation—so it’s just repackaging. Regarding cost per acquisition for an AI agency client, it depends entirely on your lead‑generation mechanism. If you run an inbound funnel with massive distribution, your CPA is essentially just your time on new inquiries—near zero. If you’re doing Upwork applications at $2 each and need about 33 applications to land a client, your CPA is $66. If you run a cold‑email campaign spending $450 per month and sending roughly 3,000 emails to land three clients, your CPA is about $150. However, CPA alone means nothing unless you also know the order value or lifetime value of the clients. For example, if Pete’s CPA is $64 and he sells a product for $1,000 on average, his CPA is 6.4% of revenue. If Samantha’s CPA is $150 and she sells for $10,000 on average, her CPA is only 1.5% of revenue. You also need to factor in referrals and other variables to judge which business is truly more successful. After that, I mention I’m running out of time because I have a podcast with Jack Roberts. We’ve become fast friends after initially being hesitant to collaborate due to a scarcity mindset. We now chat regularly, run a podcast together, and plan to play video games like Halo, Fortnite, or League of Legends while streaming, which I think will be fun and more engaging. I’m even considering buying a gaming computer just to stream with Jack weekly. Ultimately, I’m moving into Twitch streaming to play video games all day.
ai consultingcost per acquisitionltvlead generation
First, this isn’t a full business model—it’s a go‑to‑market offer. Your offer is essentially: “I will guarantee you three to five qualified appointments per month using a predictable outbound machine that automatically researches companies and generates hyper‑personalized icebreakers for cold email.” The real business model sits behind that offer: you need to understand your customer acquisition cost versus lifetime value, how you staff and fulfill the service, and the economics after you have paying customers. The concept works; you can apply it to almost any niche if you guarantee a specific number of appointments. The math is simple: if you close, say, one in five meetings and you deliver five meetings a month, that’s one new client per month. At $10,000 per client and a $2,000 monthly cost, that’s a 5× ROI. To improve the offer, be very specific about the guarantee. Saying “3‑5 appointments” is vague; you’re really promising at least three. I’d suggest a clearer guarantee—e.g., 20 qualified appointments per month—with an expected 70‑80 % show‑up rate, and make it clear that the guarantee only covers booked meetings, not attendance. Regarding your concern about splitting attention across multiple businesses, I allocate roughly 60 % of my time to my media brand, 20 % to my agency, 15 % to my SaaS, and 5 % to side projects. Each business experiences diminishing returns after a certain number of hours (about four hours for media, two for agency, one for SaaS). My optimal strategy is to work up to that cap on one business, then rotate to the next, repeating the cycle daily. This “portfolio” approach to time mirrors investment diversification—spreading effort avoids the plateau effect and maximizes overall results.
Asset-based outreach requires more time and energy upfront; it yields higher conversion but costs more money and results in fewer leads per day due to bottlenecks. In contrast, a book-a-call CTA converts lower but costs less money because you don’t need to generate assets, and the main bottleneck is scraping Apollo, which costs about 75 cents per 1,000 leads. You can get an Appify subscription and pay per use of your $26 credit (or less with my Nick 30 code). I'm 66 years old, learning this every day, and I'm enthusiastic about it. Regarding guarantees, I offer a Vappy bot that handles calls and orders via a POS system; I guarantee it will save you X, Y, or Z in dollars or hours, or you don't pay. On Maker School: it's purpose-built for people with zero sales experience, those at the start line of their business lacking daily accountability, sales systems, or pitch experience. If you have zero sales experience, this is the system for you. The main expenses are cold email and the Upwork community (the third lead-gen approach I discuss, which is free but time-intensive). Upwork averages $2-$2.50 per application; I recommend sending 10 applications per day ($25/day) for 30 days. In practice most people send only three to five per day, which still yields ROI because you spend the money gradually and can recoup it when a client pays a few hundred dollars. Instantly costs between $40 and $100 depending on the plan. Lead costs are about 75 cents per 1,000 leads, and with my Apify coupon they're even less; you may have some upfront fees, budgeting a few hundred dollars if you're going to be consistent, otherwise it could be around $100. As for Russell's question about free value offers in cold emails: the best offer is to promise to make or save the prospect a specific amount of money, and if you don't deliver, they pay nothing; you'll keep working until you achieve at least the minimum result, then pitch them something else, and even if they don't fall in love with you, you won't charge them. This kind of offer might have a 30-50% refund rate, but it generates strong front-end interest, especially for beginners who gain conversations and skill rapidly. Finally, offering to implement a simple workflow for free can be effective as long as the workflow is personalized and directly tied to a measurable return on investment, balancing risk mitigation with the guarantee.
google consoleasset based outreachbook a callupwork
Well, you can sell literally whatever you want, but my goal is always to get as close as possible to revenue. What I mean is I can literally guarantee them revenue — I've done that before. I say, 'Hey, I will guarantee you $15,000 a month in recurring revenue in the next 60 days, or you don’t pay a cent.' That sounds pretty sexy. If a business makes $45,000 a month (about $500–600k a year) and someone says, 'Hey, I want to guarantee you another $180–200k a year equivalent, and if I don’t achieve it, I won’t charge you a cent,' wouldn’t you be at least somewhat interested? It almost sounds too good to be true, and that’s the closest revenue‑based front‑end offer you can give. Many people don’t want to offer revenue because they feel they lack the ability to deliver it, which is fine — I felt that way too. So they step back one level and guarantee leads instead. For example, a common front‑end offer is: 'I will book you 20 meetings in the next 60 days for immigration, visa, or documentation services, or you don’t pay.' From the client’s perspective there are three outcomes: zero meetings (no cost, no harm), 1–19 meetings (you get prospects and likely close a few), or over 20 meetings (you make money and pay a small fee). That creates an ‘all roads lead to yes’ scenario. Whatever your front‑end offer is — revenue, leads, or something related — people are far more likely to say yes than if you’re selling something complicated. As for the backend offer, Simon, it will probably be something just related to document processing. I don’t know the niches very well, but I did used to work with an immigration lawyer in Vancouver, so I know it depends on location and where you want to do business. The best way to figure it out is to ask a client who signs up for your front‑end offer: 'Can you give me an example of everything you do to fulfill a customer?' Then you can say, 'Hey, I could probably do that — let me try and create a simple clawed routine or something.' Those are very simple and easy to do.