Pick one buyer group you can reach
Start with a role, industry, and size band. “Finance leads at mid-sized distributors who reconcile invoice exceptions” is a workable list. “Any company that needs automation” is not. Find buyers through your own network, industry associations, relevant communities, and public company pages. Respect each channel's rules and people's contact preferences.
Track why each person is relevant. A list of 30 thoughtful prospects is more useful than hundreds of scraped contacts with no evidence they own the job.
Send a research-led message
Use a short note: “I am studying how [role] handles [specific workflow]. I saw that [relevant context]. Would you be open to showing me how the last case moved through your team? I am not asking you to buy anything on this call.” Personalize the context truthfully; do not pretend to have seen a private problem.
When you have a real pilot to offer, say so plainly. Avoid a message that disguises a sales pitch as research. One respectful follow-up is enough to learn whether the contact is interested.
Move from interview to paid pilot
After you have seen the workflow, offer a limited result: one team, one dataset, and a defined turnaround. Put the fee, support expectation, data handling, and end date in writing. If the buyer cannot pay yet, ask for a concrete alternative commitment such as a scheduled review with the budget owner.
A paid pilot is useful because it exposes the buyer's real priorities. But do not force a product too early: if interviews reveal a different bottleneck, rewrite the offer and retest.
Count customers only when they are customers
Keep separate counts for replies, interviews, pilots, and paying customers. A prospect who says “interesting” is not a customer. Ask a successful pilot buyer whether they want the next cycle and who else owns the same workflow. Referrals work best when the first result was genuinely useful.
If outreach receives no replies, reconsider the buyer or message before buying ads. If conversations happen but pilots do not, revisit the pain, switching cost, and price. Pair this plan with the idea scorecard to decide whether to keep investing.
