Lead search, qualification and outreach in one workspace, with a person approving every send.
Prospecting usually spills across a map, a spreadsheet, a CRM and a pile of browser tabs. Leadesse keeps the search, the reason a company fits and the next message in one product.
It is a Kavaliero product. I built the platform end to end: sourcing, scoring, sequences, the review queue and billing.
- 11interface languages
- 3ways to find an email, tried in order
- 0 to 100fit score, with a written reason
- 40API route handlers
Finding
Search by place
A query combines keywords, location and radius. Results come from Google Places, are enriched in batches and deduplicated against leads the team already holds.
Three ways to an email
The website is scraped first. If that fails, name and domain patterns are generated and checked by DNS and SMTP. A lookup service is the last resort. The lead records which one worked.
A score with a reason
Each lead gets a fit score from 0 to 100 and a sentence explaining it, so the list can be judged before anyone spends time on outreach.
Sending
Sequences in the reader's language
Multi-step emails are drafted per lead in 11 languages, with tone set per campaign.
The draft waits for a person
Every draft lands in a review queue with the lead, subject and score in view. Approve, edit or reject without leaving the screen.
Sent in business hours
The processor works out the recipient's local time from their location and holds anything outside nine to six.
Under the surface
Untrusted by default
Scraped pages are treated as untrusted input in every prompt, and user-supplied URLs pass an SSRF guard before the scraper touches them.
A browser pool that sleeps
Headless Chrome runs at most five scrapes at a time, queues the rest and shuts down after five idle minutes.
Teams and billing
A user is provisioned from the auth token into a personal team inside one transaction. Plans and credits are defined once and enforced through Stripe.
