BuildsClose

Ticmint Signals

An always-on listener that finds people publicly frustrated with ticketing platforms and hands the team a ranked daily lead feed.

BetaInternal to TicmintBuilt Sept to Nov 2026

What it is

Signals listens for people complaining about Eventbrite, Ticketmaster, BookMyShow, Ticket Tailor, FIXR, vivenu, Townscript and others. It qualifies and scores each post, works out how the person can be reached, and gives the growth team a daily ranked lead feed with reply drafts a human approves.

The workflow is listen, qualify, enrich, recommend an action, then engage. Nothing is ever posted automatically.

The problem

Buying intent for a ticketing platform shows up in public as complaints, questions and tender notices, scattered across Reddit, review sites, social posts and RFP portals. Nobody had time to read all of it, and anonymous posts gave no way to reach the person. Signals reads it daily and says what to do about each one.

How it works

  1. ListenA collector runs daily at 2:30 AM UTC across Reddit, web search, review sites, RFPs and job posts, plus LinkedIn, X and Trustpilot.
  2. CleanJunk and review-shaped posts are filtered, duplicates removed and every permalink checked.
  3. Qualify and scoreAI classifies sentiment (frustrated, curious or actively shopping), intent, pain point and segment, then gives a 1 to 5 lead score.
  4. Work out the routeFrom profile data and context it extracts name, role, company and public contact details. Each lead gets a route, a confidence level and a recommended next action. A route is never just yes or no.
  5. Enrich on demandFor identified companies, Apollo looks up the decision-maker when a teammate asks.
  6. Draft, then a human decidesReply Copilot writes three drafts per lead: helpful with no Ticmint mention, conversational and direct. A person edits, approves and posts.

Architecture

Raw posts become scored signals, then leads with a route. Humans sit at the end, not the middle.

Impact

78daily collector runs
34actionable leads, about 30%
81signals that are market intelligence, no route yet
9source types
From raw post to a hot lead

3 September to 4 November 2026.

Raw posts721
Qualified signals115
Hot leads88
Where the signals come from

Qualified signals by source.

Reddit
49
RFPs and tenders
27
Web
24
TrustRadius
5
Trustpilot
3
X
3
LinkedIn
2
G2
1
Job posts
1
Which competitors they mention

Signals by competitor, as logged. Another 9 Eventbrite signals are shared with a second competitor.

Eventbrite
39
FIXR
10
vivenu
10
Townscript
6
Ticket Tailor
3
Unclear
23
How the 34 actionable leads can be reached

The other 81 are market intelligence with no route yet.

Company identified
21
Public reply possible
12
Direct contact
1

The engine reliably produces a steady pipeline of qualified, competitor-specific pain signals. The bottleneck now is team follow-up, with 114 of 115 still marked new, and identity coverage on anonymous sources.

Built with

TanStack StartReactTailwind and shadcnLovable Cloud: Postgres, auth, scheduled jobsLovable AIFirecrawlApifyApolloSlack

Decisions that mattered

  • Human in the loop for all outreach. There is no automated commenting anywhere.
  • Contactability is never binary. Five routes, each with a confidence level.
  • Only @ticmint.com Google accounts can sign in, and paid lookups are gated on the server.
  • Three reply tones per lead, so the helpful one never has to mention Ticmint.

What is next

  • Tune or replace the LinkedIn collector.
  • Raise identity coverage, since 70% of signals have no route yet.
  • Score on pain, intent, recency, reachability and fit together.
  • Get the team to work the backlog and record outcomes, so meetings and wins can be measured.

Sources: Ticmint Signals project overview, Nov 2026

Say hi

Starting something, or want to talk growth or building? Write to me.