Reputation monitoring that catches every negative review

People choose an internet provider by reading map listings. A weekly agent walks the company's listings and its competitors', and delivers the whole week of fresh criticism, with links, in one message.

A regional internet and cable TV provider. Twenty-five years in the market, competing against a national operator in the same city.

Telecom network equipment — reputation monitoring case study
0

negative reviews missed

3

listings watched: the company's and two competitors'

Monday

the report arrives on its own

52 weeks

of rating history accumulated

Problem 1

A bad review works against you every day it sits there

A regional operator competes with a national one. Someone moving house, or unhappy with their current provider, opens a map app and compares listings. Keeping track meant manually walking your own listing and your competitors' across two map platforms — so fresh criticism was easy to miss, and the later the reply, the worse the company looked to everyone still choosing.

Solution 1

One message, every Monday morning

The agent walks the listings on both map platforms — the company's and its competitors' — and sends the manager a single Telegram message: rating movement for the week, and every fresh negative review with a direct link to it.

Collects reviews and ratings: own listing and competitors'
Counts the weekly delta — how many reviews arrived, where the rating moved
Pulls out everything rated three stars or below, with links
One Telegram message, Monday morning, no one has to remember to look
Keeps up to 52 weeks of history — the trend is visible, yours and theirs

An honest boundary

What this system deliberately does not do

There is no public API for posting replies to reviews on these platforms. So «AI answers all your reviews automatically» means, in practice, working around the platform's rules. The honest model is the one built here: the agent monitors and alerts instantly, a human replies quickly and in the company's own voice. The listing looks alive because a person is on it — the machine just makes sure nothing is missed.

Problem 2

A web enquiry goes cold faster than anyone gets to it

Someone requests a connection and keeps comparing providers. If the enquiry lands in an inbox with no context — which plan they were looking at, what they answered in the quiz — the manager burns the call re-asking. Meanwhile the competitor has already phoned back.

Solution 2

The enquiry reaches a manager in seconds

The moment the form is submitted it fans out: to the managers' Telegram group with the plan, the quiz answers and the ad source; to a control channel so the form is visibly alive; and into a spreadsheet and call tracking. There is nowhere left for it to get lost.

Before

Seeing a new review means walking listings on two platforms by hand

A negative review sits unanswered for weeks

Nobody knows what competitors' ratings are doing

A web enquiry waits until someone notices the email

After

The whole week of fresh criticism in one Monday message

The link is right there — it can be answered the same day

Their movement sits next to yours in the same report

It reaches a manager in seconds, with full context

What changed

0

negative reviews missed

Everything three stars or below reaches the report with a link

One look

instead of a manual round

Own dynamics and competitors' in a single message

52 weeks

of rating history

Reputation trend visible over time, yours and theirs

Seconds

to reach a manager

With the plan, the quiz answers and the ad source attached

Under the hood

What this is built from

  • Scheduled collectors for two map platforms
  • Weekly digest delivered to Telegram
  • Lead routing: Telegram, control channel, spreadsheet, call tracking
  • Python, systemd timers

Let’s talkabout your bottleneck

Hiring, a project, or a second opinion on an agent design — Telegram is the fastest way to reach me. I read everything myself.

mail [at] artemborisov.ru

Based in Russia (UTC+12) · open to relocation