What we see in AI prospecting
Interest in artificial intelligence has never been higher, and that cuts both ways for anyone selling AI to businesses. Opening a conversation is easier than in other sectors, but most decision-makers still don't know which concrete problem they want to solve or how much budget to assign to it. Prospecting that works doesn't sell “AI”: it sells a use case with a measurable outcome.
At Siete we work with AI, analytics and data consultancies in Mexico and Latin America. Because we have fewer engagements in this vertical than in software or agencies, this page combines case data with team estimates, labeled as such.
Reference benchmarks
- Qualified meetings per month: 12+ (case data), sustained across two business lines in parallel.
- Meeting quality rated by the client: 86% (case data).
- Time to get the sales engine running: under 4 weeks (case data).
- Positive reply rate: 3–5% of contacts reached (estimate). It is higher than in other technology sectors thanks to market interest, as long as the message is a concrete use case.
- Meetings that turn into opportunities: 15–25% (estimate). Lower than in software, because many meetings are exploratory.
What worked
- Use cases by function, not AI in general. Messages like “cut your month-end close time” or “predict customer churn” started conversations; “AI transformation” didn't.
- Reaching whoever owns the problem, not just IT. Operations, finance, sales and data leaders replied more than technology teams, who tend to receive more pitches.
- Separating training and services. If the company sells data training as well as services or staffing, each line needs its own ICP and campaign.
- Qualifying data maturity before the meeting. Asking what data the company has and who manages it avoided meetings with prospects who couldn't move forward yet.
- Proposing a small first step, such as an assessment or a scoped pilot, instead of a full project.
A typical situation
- A boutique AI and analytics consultancy in Mexico, with a validated product and happy clients but growth that depended only on referrals, got its sales engine running in under four weeks and sustained 12+ qualified outbound meetings a month across its two business lines.
Common mistakes
- Selling the technology instead of the outcome.
- Prospecting only CTOs and CIOs.
- Bringing a full project proposal to the first meeting.
- Counting every exploratory meeting as an opportunity.
How to apply this to your company
If you sell AI, data or analytics solutions to businesses, see how we approach lead generation for artificial intelligence companies or book a call with the Siete team to define the use cases and decision-makers with the most potential.
