Skip to content
Academy
Marketing Academy · Field Work●Marketing Tools
All projects·From: Slack and Teams as a Marketing Operations Hub
MiniAudit· 25 minutes

The Firehose Diagnosis: Auditing TBO Tek's Alert-Routing Chaos

TBO Tek

Objective: Given a synthetic 15-row channel activity log, apply the lesson's three notification guardrails to decide which alerts should move channels, get batched into a digest, or lose their @mention entirely.

You're a marketing ops analyst at TBO Tek, the Gurugram-founded, Nasdaq-listed B2B travel distribution platform (BSE/NSE-listed, ~$1.79B IPO market cap) serving a 147,000+ travel-agent network. Five tools, HubSpot, a social listening tool, the ad platform, the CMS, and a partner-alert bot, all post into #general, and the team wants to know what to fix before people start muting the channel.

Score each row of the activity log against the three guardrails from the lesson (separate by urgency not tool, batch what can wait, only @mention what's actionable), then produce a triage list of moves, batches, and demotions.

Before you start

What you'll need

Free path (everything below is enough to finish)

Google Sheets
FreeImport, tag, and triage the channel activity log

Free, no account friction, filtering and column tagging is all this audit needs

The process

1 step

Step 01 of 01

Separating channels by urgency, not by tool

The lesson's three guardrails: separate channels by urgency (not by tool), batch what can wait, and reserve @mentions for alerts that need a decision within the hour.

Of 15 rows posting into #general, a CRM lead-alert bot pings @channel for every new signup, a weekly analytics digest also posts live, and a sentiment-spike alert has no @mention at all. Which of these is actually backwards?

Google Sheets— Import the synthetic activity log (columns: tool, alert_type, current_channel, frequency, has_mention) and freeze the header row.

Procedure

  1. Import the 15-row log and freeze row 1
  2. Add an `urgency` column: crisis (needs action within the hour), operational (needs eyes today), informational (can wait for a digest)
  3. Flag every row where `has_mention` = yes but urgency = informational as a demotion candidate
  4. Flag every row where frequency = live but urgency = informational as a batching candidate
  5. Flag the one row where urgency = crisis but has_mention = no as a promotion candidate
Sample output
TRIAGE RESULT (15 rows)

PROMOTE TO @MENTION (1)
  Sentiment-spike alert (Brandwatch) -> #brand-crisis, add @on-call-mention, currently silent

DEMOTE FROM @CHANNEL (6)
  New-signup alert (HubSpot) -> #general @channel every signup -> move to #pipeline-log, no mention
  ...5 more rows

BATCH INTO DIGEST (4)
  Weekly analytics summary -> currently live-posts -> switch to Friday 9am digest
  ...3 more rows

LEAVE AS-IS (4)
  Deal-stage change (HubSpot) -> #deals-active, correctly scoped, no change

Healthy

Only the genuinely time-sensitive row (sentiment spike) carries an @mention; routine signups and weekly summaries are demoted or batched.

Unhealthy

A live signup counter pings the whole channel while an actual reputation crisis posts silently, which is what the un-triaged log shows.

What this means

Urgency and mention-worthiness are not the same axis as 'how often does this tool fire' — a tool that fires constantly (signups) is rarely the one that needs a ping, and a tool that fires rarely (crisis alerts) usually is.

So what do I do about it?

SymptomActionEffort
6 of 15 alert types over-mention relative to their real urgencyStrip @channel/@mention from routine operational posts, reserve it for the crisis-tier row only30 min
A live sentiment-crisis feed has zero mention despite needing action within the hourAdd a targeted @on-call mention to the one row that actually needs it5 min
YouYou can do this yourself, no engineering access required.

Final deliverable

A triage table sorting all 15 alert types into promote/demote/batch/leave-as-is, with the target channel and mention setting for each.

See a reference example
Sample output
Freshworks marketing ops, channel triage (excerpt)

PROMOTE (1): Sentiment-spike alert -> #brand-crisis, add @on-call
DEMOTE (5): New-trial-signup pings -> #pipeline-log, drop @channel
BATCH (3): Daily ad-spend recap -> fold into Friday digest
LEAVE (6): Deal-stage change in #deals-active -> already correctly scoped

Success criteria

You're done when you can:

  • Every row is classified by urgency, not by which tool sent it
  • The one under-mentioned crisis-tier alert is correctly flagged for promotion
  • At least 5 over-mentioned operational alerts are correctly flagged for demotion