How Sofía Martínez, a solo HR consultant in Buenos Aires, rebuilt her client acquisition from the ground up using the AI Sales Starter Kit — with zero ad spend, zero cold calling, and a system that now runs in 45 minutes a week.
Sofía Martínez had spent five years building a consulting practice that was widely respected by everyone who had worked with her — and completely invisible to everyone who hadn't.
Sofía specialized in HR strategy: helping growing companies build the people infrastructure needed to scale without imploding. Talent acquisition frameworks. Onboarding architectures. Manager capability programs. Performance review systems that employees didn't dread. She was genuinely excellent at it — her client retention rate was 94% over five years, and every single engagement had produced measurable outcomes she could speak to in detail.
The problem was simple and devastating: she had no system for finding the companies that needed her before they were referred to her by someone else. Of her twelve lifetime clients, ten had arrived through the same two LinkedIn connections — former colleagues from her time inside a multinational HR department who had moved into VP of People roles at startups. The other two were cold inquiries that had found her website by accident.
In early February, Sofía sat down with her financials and felt a specific, slow-building dread. She had three active clients. One was a six-month engagement entering its final stretch. Another was a project wrapping in four weeks. The third was a referral from eight months ago that had quietly become her anchor — a $3,200/month retainer that, if it ended, would take her income below what she needed to cover her studio apartment and her team of one (a part-time operations assistant she'd been promising to make full-time for two years).
"I had proven over five years that I could do the work. What I had never proven — because I had never actually tried — was that I could find my own clients."
Sofía Martínez — February, before implementationShe had made two attempts to change this. In September, she invested $1,800 in LinkedIn Ads over six weeks. She got eleven leads — three were genuinely interested, two had calls that went nowhere, and one turned into a $2,400 one-time project. Negative ROI. In November, she committed to posting on LinkedIn every day for a month. Engagement was strong. Comments from peers and admiration from other consultants. Zero client inquiries. Zero revenue from the effort.
The failure pattern was clear in retrospect: both attempts were strategies without systems. Isolated efforts with no follow-up infrastructure, no ICP to guide targeting, no structured conversion process, and no way to maintain relationships with the people who weren't ready to buy yet but might be in three months.
Two of three clients ending within 5 weeks. Revenue would drop from $7,400 to below $3,200 with no replacement in sight — and no system to generate one.
Ten of twelve lifetime clients from two LinkedIn contacts. One relationship pivot away from a business emergency, and no mechanism to change it.
Ads spent. Content created. Zero clients generated. Not because the channel was wrong — because there was no system behind the effort to convert attention into conversation.
Using the AI Sales Diagnosis from the Kit, Sofía identified six specific, interconnected problems. Each one had a cause — and a fix. Together, they explained precisely why five years of excellent work hadn't translated into a self-sustaining business.
Her LinkedIn headline read: "HR Strategy Consultant | Talent & Organizational Design." Her website: "Expert HR consulting for organizations that care about their people." Both descriptions are true — and both are invisible to a founder who doesn't think "I need HR strategy." They think: "We're scaling and half our new hires quit in month three." Sofía was answering a question no one was asking, instead of naming the problem everyone was feeling.
When asked to describe her ideal client, Sofía said: "Any company going through growth that needs HR support." This describes tens of thousands of businesses in Latin America alone. Without a specific ICP — industry, size, stage, decision-maker role, specific trigger event — no outreach message could feel personal, no content could resonate specifically, and no lead generation effort could be targeted with precision.
In her previous attempts, Sofía had sent a handful of cold LinkedIn messages — all variations of "I'm Sofía, I do HR consulting, I'd love to connect." No research. No specific observation. No hook that demonstrated she understood the recipient's actual situation. And critically: one message, no follow-up. No sequence. No mechanism for the reality that most interested prospects don't respond to the first message because they're busy, not disinterested.
When Sofía did get on calls, she was genuinely impressive. Prospects left feeling they'd spoken to a deep expert. The problem: she had no structured framework to move from problem diagnosis to clear next step. Calls ended with "let me think about it" because she hadn't created the conditions for a decision. Her estimated close rate: approximately 7%. She had never tracked it. She found this number both shocking and instantly recognizable.
"I don't like to chase people." This is how Sofía described her follow-up philosophy. The effect: every prospect who expressed interest but wasn't immediately ready to move forward was quietly abandoned. No sequence. No nurture. No re-engagement mechanism. In B2B services, the majority of revenue comes from follow-ups. She had eliminated that possibility entirely with a belief she'd mistaken for professionalism.
Sofía's pipeline management consisted of her inbox, her memory, and a sticky note on her monitor with three names on it. She could not tell you how many discovery calls she'd had in the past year, what percentage became clients, or what had happened to any lead after initial contact. Without data, she couldn't identify patterns, improve conversion rates, or understand where leads were actually dying.
Six separate problems, one single cause: Sofía had been operating as if client acquisition were something that happened to her practice rather than something her practice actively did. The AI Sales Starter Kit gave her the architecture to change that — systematically, layer by layer, in seven days.
Sofía followed the 7-Day Implementation Plan from the Kit exactly. Then she ran a strict weekly operating ritual for the remaining three weeks. Here is the full account — every significant decision, every number, and every moment that changed something.
Sofía started with Prompt #01 and spent two hours being pushed to answer questions she had spent five years avoiding. The AI forced specificity at every step: What exact job title signs the contract? What's the company's revenue range? What's the specific trigger event that makes them realize they have a problem right now?
The output that emerged was surgical: CEOs and Heads of People at B2B tech and professional services companies with 30–120 employees in Argentina, Colombia, and Mexico, raising Series A or experiencing 40%+ headcount growth, with new hire turnover in the first 90 days above 20% — a signal they had outgrown their informal onboarding and retention infrastructure.
This precision changed everything downstream. She now knew exactly who to search for, exactly what pain to name, and exactly why a message from her would be relevant today and not just theoretically interesting.
With the ICP defined, Sofía ran Apollo.io to extract 380 LinkedIn contacts matching her criteria — filtering by job title (CEO, Head of People, VP Talent), company size (30–120 employees), industry (SaaS, consulting, professional services), and geography. She supplemented with 195 leads from Google Maps targeting "tech companies" and "consulting firms" in Buenos Aires, Bogotá, and Mexico City using Outscraper. A final 105 came from Crunchbase — startups that had raised in the past 12 months, a high-signal ICP indicator she had never thought to use before.
By end of Day 3: 680 leads in her Google Sheet, all ten columns populated. Company name, contact, title, email, LinkedIn URL, website, source, and three AI-generated columns yet to come.
Sofía ran the AI Research Batch Prompt in Claude, processing 50 companies at a time. For each: an ICP fit score (1–5), the most likely decision-maker title, and one specific outreach angle — a particular detail about their recent growth, hiring announcements, or visible operational challenge that could be referenced in the opening line of a message. The output was 680 rows of scored, personalized prospect intelligence. It took approximately 3 hours — including coffee. Manually, this would have taken months.
Top 140 leads scored 4–5 became Priority Tier 1. The rest were divided into mid-tier sequences (Tier 2) and long-range nurture (Tier 3). She sorted the sheet by score and never looked at the bottom half before starting outreach.
Using Prompt #02, she wrote three versions of her cold outreach — problem-led, curiosity-led, and result-led — for both email and LinkedIn. Using Prompt #12, she built a 5-touch follow-up sequence for each channel. She configured HubSpot CRM (2.5 hours, Day 5), connected Calendly with her booking link in every CTA, and activated Instantly.ai with domain warm-up on two sending domains.
Day 7: Make.com automations built and tested — new reply → CRM status update + instant booking confirmation. First 80 emails to Priority Tier 1 sent. LinkedIn connection requests with personalized notes — 28 sent in one hour, each with a unique AI-generated first line referencing the specific prospect's company situation.
By Day 9, the first replies began arriving — and they were substantively different from anything Sofía had received from her previous outreach attempts. Not polite deflections. Actual engagement. One CEO replied: "How did you know we just went through a round of bad hires? This is exactly the conversation I've been trying to avoid having with my board."
By end of Week 2: 19 replies from 108 total outreach messages sent (email + LinkedIn combined) — a 17.6% reply rate. Eight were positively interested. Seven were "not right now but keep in touch" — all entered an automated 90-day nurture sequence in Instantly.ai. Four were genuine rejections. Six discovery calls were booked directly through Calendly. Zero scheduling friction.
The reply quality itself was evidence the ICP and personalization were working. Three of the eight positive replies mentioned that Sofía had named something specific — a recent job posting for five simultaneous roles, a LinkedIn announcement about Series A funding, a blog post about company culture during growth. The specificity of the opening line was doing the work she had previously hoped charisma and reputation would do alone.
Monday morning of Week 2: Sofía added 90 new leads to the database — her first weekly ritual execution. Scored them. Loaded the top tier into Instantly sequences. The database was growing on a controlled rhythm for the first time in five years of practice.
All six discovery calls happened across Days 15–20. Before each one, Sofía ran Prompt #17 (AI Pre-Call Briefing) — 90 seconds of input, 3 minutes of reading, and she had: a situation brief on the company, five tailored questions for this specific prospect, and written responses to their three most likely objections. She described the experience of these calls as "completely different" from anything before — not because the prospects were easier, but because she arrived as the most prepared person in the conversation.
Call 1 — SaaS CEO, Buenos Aires: The company had raised Series A six months prior and hired 22 people in four months. Turnover in month one was visible in their LinkedIn activity — several "We're hiring!" posts for the same roles. Sofía had noted this before the call. She named it in her opening question. The CEO said, "We've talked to three consultants. You're the only one who actually researched us." Closed at $5,200/month on the call. Contract signed same day.
Call 2 — Head of People, consulting firm, Colombia: Strong fit, clear pain. "I need to think about it." Sofía — using the objection framework from Prompt #16 for the first time — asked directly: "Of course. What specifically would you like to think through? I'd rather address it now." The real hesitation emerged: board approval needed. Sofía offered a one-month pilot structure at a reduced entry point. Reframed the ask. Signed three days later: $3,400/month, pilot converting to full engagement in Month 2.
Call 3 — CEO, professional services firm, Mexico City: Closed immediately. $4,800/month. He said he had been trying to solve the same onboarding problem for two years and had given up on finding external help that actually understood the operational context. "You're describing my company," he said, fifteen minutes in. "When can we start?"
Call 4: Closed. $3,400/month. A referral from Call 1's CEO — who had already mentioned the engagement to a peer. The outbound system had generated its first organic referral in Week 3.
Call 5: Wrong fit — company was pre-revenue, not Sofía's ICP. She ended the call gracefully in 14 minutes. The ICP scoring had flagged this one as Tier 3; it had been fast-tracked by a contact introduction. She tightened the scoring criteria.
Call 6: "Let me discuss with my co-founder." Entered a 6-touch automated follow-up sequence. Would close in Week 4.
By Week 4, the system was running without requiring Sofía's constant attention. Sequences were active across 1,240 leads. HubSpot was updating statuses automatically via Make.com. Buffer was posting her LinkedIn content twice a week from the batch session done in Week 1 — and those posts were generating inbound comments from people who matched her ICP.
Five more discovery calls came from new outreach — four converted. The co-founder from Call 6 returned after receiving a strategically timed follow-up email addressing exactly the concern he'd raised: a case study about a similar firm that had run a pilot before committing. He signed at $4,200/month. Two prospects from the Week 1 nurture sequences re-engaged on their own, requested calls, and both closed.
Her two original clients, both of which had been scheduled to end, were still active. One had extended after Sofía — for the first time — proactively proposed a continuation plan with a specific new objective framed in outcome language (Prompt #09: Upsell Designer). The other had referred two peers. Both were in active discovery.
By Day 30: 11 active paying clients. $31,200 in monthly recurring revenue. Three months earlier she had been deciding whether to let her part-time assistant go. She hired her full-time on Day 28.
Every tool Sofía used is in the Tier 1 stack from the Kit. Her first new client covered the full annual cost of the stack in its first payment. Here is precisely how each tool was deployed.
Instantly.ai ($37) + Claude Pro ($20) + Apollo basic ($47) = $104/month. HubSpot CRM, Calendly, Make.com, and Buffer were all free at Sofía's usage level. Her first new client — $5,200/month — covered the annual cost of the entire stack in its first payment. ROI on Year 1 tooling investment: approximately 600×.
The results below are not the product of luck, timing, or an unusually easy market. Each metric is the direct output of a specific system decision. That is what makes them repeatable.
Sofía's discovery call close rate went from approximately 7% to 46% in 30 days — a 557% improvement. She didn't become a different person. She didn't take a sales course. She arrived to every call with a pre-built briefing, a structured question sequence, and a prepared response to every likely objection. Preparation is not a personality trait. It's a system input. The Kit makes it automatic.
These are not general marketing insights. They are the specific, hard-earned lessons from 30 days of building and running a real AI-powered sales system inside a solo consulting practice — applicable to almost any service business.
Sofía spent five years believing that a broad ICP protected her from missing opportunities. In reality, it made every outreach message sound like it was written for nobody in particular — because it was. The moment she narrowed her ICP to a specific company profile, growth stage, and trigger event, her reply rate went from 0% to 17.5% on the first batch. She reached fewer people. Every single one of them felt like she was speaking directly to their situation. Specificity does not shrink a market. It focuses a message until it becomes impossible to ignore.
Sofía's previous cold messages opened with her credentials and service description. Her new messages opened with a single, researched sentence: a reference to a recent hiring announcement, a specific content gap on their website, a detail that proved she had actually looked at their company before reaching out. Only the first 2–3 sentences were unique. The rest was templated. That single change drove the entire improvement in reply rate. The rule: earn the right to be read before asking to be listened to.
Three of Sofía's eleven clients in Month 1 came from leads that had initially said "not right now" and entered automated nurture sequences. They re-engaged on their own timeline — 9, 14, and 22 days after initial contact — after receiving follow-up messages that delivered relevant value without pressure. In her previous approach, she would have sent one message and quietly given up, calling it respect for the prospect's time. In reality, she was abandoning relationships that needed maintenance, not abandonment. The automated sequence kept the door open. They walked through it when they were ready.
Sofía had been having discovery calls for five years with a 7% close rate. In Month 1, using the AI Pre-Call Briefing prompt before every call, she closed 46% of them. No new sales techniques. No personality change. No charisma training. She simply arrived to each call having already thought through the prospect's specific situation, their likely concerns, and the five best questions to move the conversation forward. The outcome was that she sounded like the most prepared expert in the room — because she was. The system manufactured that preparation in 90 seconds. The performance came from what she did with it.
Her second new client in Week 3 began as a "let me think about it." Using the objection framework, Sofía stopped — for the first time in her career — and said directly: "Of course. What specifically would you like to think through? I'd rather address it now than have you stuck on something I can answer clearly." The actual concern surfaced immediately: she needed board sign-off before committing. Sofía restructured the engagement as a pilot that didn't require full approval. Signed three days later. Then expanded two months after that. The old Sofía would have said "Take your time!" and never heard from her again. The new question was not a sales technique. It was honesty — and the Kit made honesty the default response.
On Day 30, Sofía tracked every minute she spent on sales activity. The total: 47 minutes. During those 47 minutes: she reviewed metrics in HubSpot, added 90 new leads to the database, adjusted the subject line on one email sequence that had a lower-than-average open rate, and read and replied to four warm inbound responses. The system sent 312 emails, updated 44 CRM records, posted two LinkedIn pieces, scheduled three calls, and triggered one onboarding flow — without her involvement. She had spent five years doing all of that manually. The Kit didn't make her work less. It made her work land harder. That is what compounding looks like when it's finally working for a business instead of against it.
"I used to feel guilty about not doing enough sales. Now I feel clear about exactly what the system does and exactly what I need to do. That clarity is worth more than any single client it generated."
Sofía Martínez — Day 30 after implementationEvery prompt Sofía used. Every automation flow. Every module. Every checklist. All of it is inside the AI Sales Starter Kit — ready to implement in the same seven days, for any consulting or service business, regardless of size or stage.
Fictional case study · Results representative of real-world system implementations · AI Sales Starter Kit · 2025
Es un escenario representativo basado en la metodología Souing aplicada en empresas del sector logístico. Los nombres y datos específicos son ilustrativos para proteger la confidencialidad de los clientes.
Los resultados dependen de factores como el compromiso del equipo, la calidad de datos y el tamaño de la empresa. Los números mostrados representan rangos típicos observados, no garantías contractuales.
El proceso de implementación estándar es de 30 días. Los primeros resultados observables (leads activos, reuniones agendadas) suelen aparecer en la tercera y cuarta semana del proceso.