Skip to content
MakerGain
← Back to cases

He reported a 15% close rate, but customers left after three months. He rebuilt his AI sales agent to screen out buyers who are a poor fit

Edited by MakerGainPublished: Article updated:

Sohaib Ahmad’s DM Champ is an AI sales agent working across WhatsApp, Instagram, and other channels. It helps agencies and businesses handle incoming leads: restart conversations with old lists, assess potential, arrange the next step, and hand back to a person when stuck. He reported a 15% close rate, yet customers left after about three months. He attributed this to AI persuading people who did not truly need the product, then shifted optimization from conversion to customer fit and rebuilt the system to proactively screen out unsuitable buyers. Retention data after the rebuild have not been published.

According to Sohaib Ahmad’s account in a Starter Story interview, he used Meta ads to sell his own AI sales agent and reached a 15% close rate. Then he found that customers left after about three months—“every single time” (Starter Story interview[1]). His explanation was that AI had persuaded some people who did not really need the product.

That is worth a closer look coming from someone selling sales software. DM Champ is an AI sales agent operating across WhatsApp, Instagram, Messenger, SMS, web chat, and other channels. Its website positions it as a white-label platform for agencies: the Business tier for internal use starts at $97/month on monthly billing, while Agency, with white labeling and customer subaccounts, starts at $297 (dmchamp.com[2]). It sits at the point where leads are already coming in and support cannot keep up with replies.

Image: From the official product demonstration[3], obtained on 2026-11-18; official presentation retained as shown, for product illustration only. Dashboard amounts are part of official promotional imagery, not evidence of DM Champ revenue or actual customer payments.

The first sale came from an old list nobody had followed up

Ahmad found that position through his own work. He previously ran a marketing agency, working for FueGenix hair-transplant clinic. With clients such as clinics, leads arriving through WhatsApp, Instagram, and web chat went cold because teams could not keep up (Starter Story[1]).

The first sale was simple. He cold-emailed local clinics; three expressed interest, and he followed up with only one. That clinic had fewer than 200 old leads in a spreadsheet. After upload, AI contacted them one by one on WhatsApp. The interview says those conversations brought the clinic about $2,197.80 (€2K) in monthly recurring revenue. That is the client’s money, not DM Champ revenue. The interview gives only that figure, without breaking out how many appointments resulted.

The workflow matters. Those old leads had been dormant: names sat in a spreadsheet with nobody replying. AI took over reopening contact, assessing potential through conversation, and arranging the next step. People did not disappear. When conversations stalled or customers raised complex questions, the system still needed to hand back to a person; the pricing page lists “human takeover alerts” as a separate feature (pricing page[4]). In Ahmad’s account, then, this is about restarting conversations with an old list and identifying people who can be booked, rather than merely adding another talking window. As for how many customers it actually recovered, the public material offers only his account of client-side revenue, without an appointment count.

The $40,000 story, and what the customer actually paid

Ahmad recounted a win in the interview: for a service quoted at $40,000, AI rejected three requests for a discount, and the customer later returned and paid (Starter Story[1]). The anonymized WhatsApp record published on the website shows only one refusal and is labeled FueGenix Hair Clinic (dmchamp.com[2]). In it, AI quotes a starting price of $40,000 and asks whether that fits the budget. The reply is “I can go up to $10,000 max.” AI says pricing is fixed because of demand and quality, and invites the person to reconnect if their budget changes. Four days later, the person returns with another question. On day 14, they write, “I completed payment.” Beneath the screenshot, the website adds: Lead paid · $1,000 consultation.

This is marketing material published by the company, not independent documentation. The page establishes only that it displays a conversation refusing a discount, with a $1,000 consultation fee noted below. It neither independently proves payment nor rules out a later, separate transaction. What can be said is that the displayed $1,000 sale is not the $40,000 service quote. The customer returned after four days to keep asking questions, rather than AI returning to them. Nor is there evidence that AI prolonged the negotiation.

Sales went smoothly—then he realized they were the wrong people

The team designed conversations to be stateful: when to advance, when to step back, and when to say no. Ahmad says it can close sales. A 15% close rate looks attractive in a category selling through direct messages, but it remains founder-reported, without payment or CRM records to corroborate it.

The problem appeared three months later. Customers left, “every single time.” Ahmad’s judgment was that AI had sold to people who did not truly need the product—“high close rate, wrong customers.”

The AppSumo figures offer another example. A self-service listing brought about $21,978.00 (€20K) in revenue and 300 customers, with a 40% refund rate. He attributed that to pricing: the cheapest tier attracted tire-kickers who wanted to try it and refund. The interview does not specify whether about $21,978.00 (€20K) was gross before refunds or a net figure. What remained after 40% refunds cannot be calculated.

Both point to the same problem: closing quickly is not the same as finding the right people. His response was to change the optimization target from conversion to customer fit, rebuilding the system to proactively exclude unsuitable buyers. The public material does not explain the specific changes.

A signature demo, and an optional switch

Ahmad said in the interview that DM Champ uses its own AI to sell DM Champ: prospects complete a conversation with AI without realizing it is not human, then learn the truth. This is used as a product demonstration (Starter Story[1]).

It is a striking demo, but in his account it shows only that a conversation can briefly leave someone unaware that the other party is AI. That is not the same as conversion or retention. One condition must be noted: the pricing page lists “Optional AI disclosure notice” as an optional feature included in every plan, without specifying whether it is on or off by default (pricing page[4]). The seller uses it for a surprise moment; buyers face the same optional switch with an undisclosed default. That moment of “not realizing” remains Ahmad’s description, not an independent verification.

Ahmad also described a concentrated burst of sales from a warm channel. A team member knew a Dutch business coach, who ran a 60-minute webinar demonstrating DM Champ to 300 attendees. According to Ahmad, within two minutes of the offer going live, Stripe received $20K and 80 customers joined. That combines borrowed trust with a moment compressed into two minutes. It shows a complex product can be explained to a warm audience, not that cold traffic will convert at the same rate.

Agencies pay the platform fee; Ahmad says Meta template charges are DM Champ’s largest cost

The three self-service monthly tiers are Business at $97, Agency at $297, and Agency Unlimited at $497. Agency includes 10 customer subaccounts, with each additional one at $27/month; Agency Unlimited has no customer limit. Annual billing is half price, at $582, $1,782, and $2,982 per year, respectively, and carries a 90-day money-back guarantee (pricing page[4]).

That structure makes one thing clear: whoever pays the platform need not be the person served by AI. Agencies apply their own domain, logo, and Stripe and charge end customers what the website describes as a typical $200–$500/month, keeping the difference. The AI chats with end customers on WhatsApp, while the bill goes to the agency. Business explicitly permits only internal company use, not resale. The split between agencies and businesses using it internally is undisclosed; what is clear is that pricing places agencies in the payer’s role. For the same stretch of work—existing customer conversations needing follow-up—another official option is RuleInside. It focuses on e-commerce calls and chat, also covering WhatsApp, overlapping with DM Champ’s social-direct-message entry point but with a different emphasis. Managed plans start at about $436.26 (€397)/month, normally with a one-time about $1,318.68 (€1,200) setup fee per agent. Starting before November 30, 2026 waives implementation for one agent on Lite and Starter. A monthly minute pool is included, and the provider handles implementation, maintenance, and tuning. A about $108.79 (€99)/month self-service version lets users build their own agents with no setup fee (RuleInside pricing page[5]). Compared with DM Champ Agency, the distinction is more than the monthly fee: DM Champ gives agencies control over white labeling, customer subaccounts, and configuring their own Meta/Twilio or buying credits. RuleInside includes “build it and run it for you” in its managed monthly fee. Both are their respective websites’ accounts, not evidence that one performs better.

The cost structure runs against intuition. In an August 2026 AppSumo Q&A, Ahmad said AI was not the largest expense. Meta’s WhatsApp template fees ranked first by a substantial margin and were carrier costs that could be passed to customers; AI came second (AppSumo Q&A[6]). The pricing page aligns with that: for DM Champ-managed numbers, Meta fees are deducted from purchased credits by sending volume, with a small billing fee added to Meta’s rates. Bringing a Meta app or Twilio account connects payment directly to the carrier.

In the same Q&A, he said the company was cash-flow positive before AppSumo and treated AppSumo proceeds as one-time revenue, not operating funds. He also said he had released a proprietary model with benchmarks comparable to or better than some midrange frontier models, and that moving to its own infrastructure had substantially lowered operating costs. The latter has no independent benchmark or cost data; for now it can only be recorded as his claim. The Q&A provides no complete profit-and-loss statement, so “cash-flow positive” does not mean profit is known.

A figure anchored in March 2026

As of the March 20, 2026 interview, he reported DM Champ revenue of about $10K/month, serving 150+ businesses. Those are self-reported figures at the interview date, not reconciled against one-off campaigns, refunds, costs, or profit. The 150+ businesses are not broken down into lifetime deals, trials, and agency deployments, so $10K cannot be divided by them to produce average customer spending.

In an industry that treats close rate as the report card, Ahmad changed his optimization target from “more sales” to “fewer mismatches.” Existing sources provide no retention data showing whether customers stayed after the rebuild. He has changed the exam, but has not yet turned in its score.

Continue reading

AI commerce
A product research tool built in 48 hours and the equity deal that gave it distribution

Hasaam Bhatti had no computer science degree. None of the 10 to 12 products he had built reached production. While running Amazon brands, he spent 20 to 30 hours researching each product idea and copying data into spreadsheets. He pitched Legacy X on building Launch Fast in 48 hours. It brings market and product research onto the Amazon search page, where private-label sellers can use it before ordering inventory. At their joint launch, he traded equity for access to Legacy X's existing distribution, reaching thousands of its active sellers directly rather than building an audience from scratch.

AI commerce
The $500 arrived before the AI caller was written

Setter AI takes over follow-up after a form submission. The website says it sends a first message via WhatsApp, text, or web chat within ten seconds, continues following up, writes bookings to Calendly or Google Calendar, and passes results to the CRM. The founders originally wanted cold-call automation, but before coding launched a landing page using an indexed domain, Ahrefs keywords, and a fake ElevenLabs AI-caller demo, receiving $500 according to their interview account. The interview calls this a refundable-deposit framework but does not say whether the money was refunded or counted as revenue; the intent was to test businesses' willingness to pay.

AI commerce
Charging an average of $30 per post and requiring at least 100 creators per campaign: Stack Influence turned micro-influencers into a volume business

Stack Influence turns micro-influencer posting into a managed service: brands upload product information and posting guidelines, the platform matches creators according to campaign size, charges an average of $30 per completed post, and requires a minimum campaign size of 100 people. Gasner says the company first built internal tools for its own e-commerce business and then turned them into a platform; he acknowledges that male-oriented and very narrow categories are hard to run. TrustMRR's publicly disclosed Stripe-verified revenue scale is not the same as net revenue or profit.

AI supportAI searchAI commerce
When a Visitor Clicks Handoff to a Human, Chat Data Sends a Colleague an Email with a Login-Free Link

Chat Data trains company documents, webpages, and files into an AI customer-service agent that can be deployed to websites and messaging channels, and uses a login-free email link to connect visitors who ask for a human. It charges by message credit and supports custom models and MCP hosted by Composio; the TrustMRR page lists MRR of 17,645 US dollars and 304 active subscriptions, with a team of 2–5 people.