case · 2026 role: product + interface duration: MVP status: working console

Source Hub CRM
1215 sources → 716 usable

A working console that turns a catalog of Telegram sources into a queue of signals for websites, bots, design, and automation work.

[ visual · source catalog → signal queue ]
01 / context

Why it
was needed

The problem is not that requests do not exist. They are spread across hundreds of Telegram sources and disappear in noise.

The project has a catalog of 1215 public sources, with 716 already selected as usable. But a list of channels does not sell; you need a screen with real signals.

Source Hub separates three layers: where to search, by which rules, and which messages already look like buying intent.

02 / solution

How the signal console
works

The system keeps source catalog, offer rules, and CRM queue separate so a person quickly sees who to write, why, and with what context.

  • I
    Source catalog
    Public Telegram spaces are collected and ranked in advance.
  • II
    Offer rules
    Separate profiles for websites, flower shops, bots, design, and automation.
  • III
    Request queue
    Hot, warm, and noise become a working list, not overloaded analytics.
  • IV
    Reason shown
    Every signal explains why it appeared.
  • V
    Manual control
    No auto-spam: the system helps decide, it does not send questionable outreach.
[ visual · source catalog ]
[ visual · hot / warm / noise ]
[ visual · reason shown ]
[ visual · manual decision ]
03 / result

What we could
measure

1215 sources → 716 usable. Here are verifiable outcomes and visual materials from the source projects.

1215
sources found
716
usable sources
3
signal groups
0
auto-spam
Source Hub exists not to read more Telegram, but to read less and see only messages with deal potential.
- project breakdown
04 / how it is built

Stack,
and why

01
Static frontend
The first console stays light while the backend is not needed.
02
Source catalog
Separates sources from found requests.
03
Rules config
Offers can change without rewriting search logic.
04
CRM queue
Requests become a work list instead of a text dump.
05
Manual decision
Keeps first-touch quality high.
05 / next work
Telegram Leads
Channel parser · scoring
06 / contact

Similar task? 30 minutes, no brief.

Describe what you need. I will say honestly whether I take it, how much it costs, and whether AI can speed it up.

book 30 min message on telegram