01
GENERATE A BRIEF
Run the agent on demand
Runs for about a minute. Up to 6 model calls, every one shown as it happens. 2 runs per visitor per hour. You can stop it at any point.
Labs
Act
Autonomous agent
The Monday Brief
An agent that investigates at-risk accounts every week on its own schedule, and writes up what it found.
Early every Monday it checks account health, recent feedback, usage and open flags, decides what matters, and writes a risk brief a CCO can act on. Start a run yourself and watch each step as it happens. Past briefs are kept below.
02
RUN STATUS
Run #364
Live trace
I'll start by pulling all at-risk accounts and flagged accounts simultaneously, then dig deeper into the findings.
→ 10 items
→ 0 items
No flagged accounts. Now I'll pull health details, usage trends, and recent feedback for the top 5 highest-risk accounts in parallel.
→ keys: name, health, usage_trend_pct
→ keys: name, health, usage_trend_pct
→ keys: name, health, usage_trend_pct
→ keys: name, usage_trend_pct
→ keys: name, usage_trend_pct
→ keys: name, usage_trend_pct
→ 3 items
→ 3 items
→ 3 items
Excellent data. Tinker Supply Lab's feedback is actually positive despite a low health score and steep usage decline — that's an interesting nuance worth flagging. Let me now confirm I have enough to …
03
MONDAY BRIEF
The brief
Monday Risk Brief — All Segments
Week of 17 August 2026 | All Segments | Prepared for Chief Customer Officer
Situation Overview
Ten accounts are below the health threshold. No accounts are currently flagged for executive review. Three require immediate attention this week.
Top 3 Highest-Risk Accounts
1. Halberd Financial Group — Health 33 | Usage -18%
Active churn signal: the account has issued a two-week ultimatum, citing broken custom metrics in Fetch Radar, recurring report export failures, and a three-week-old unresolved critical support ticket. They are in live conversations with competitor Perch AI and have explicitly named competitor Coriolis as a benchmark for support responsiveness. Leadership contact has been demanded.
2. Nook & Bolt Technologies — Health 47 | Usage -12%
Confidence is eroding across reliability and AI accuracy. Scheduled reports are running hours late with no notification, Fetch Radar forecasts missed a predictable seasonal pattern, and the ops lead is actively evaluating competitor Perch AI on a peer recommendation. A compliance audit is also exposing gaps in Fetch Flow's data export capabilities.
3. Tinker Supply Lab — Health 53 | Usage -22%
The steepest usage decline in the cohort (-22%), though recent feedback is entirely positive — the account reversed a near-decision to switch to competitor Perch AI after a UI overhaul and pipeline fix landed well. The drop in usage needs an explanation before assuming recovery; it may reflect a workflow change rather than disengagement, but that must be confirmed.
Talking Points for the CCO's Team
1. Two accounts have named the same competitor.
Perch AI is referenced by both Halberd Financial Group and Nook & Bolt Technologies as an active or likely alternative. This is a pattern, not a coincidence. The team should understand Perch AI's current positioning on forecasting accuracy and support SLAs before engaging either account.
2. Support responsiveness is the proximate trigger at Halberd Financial Group.
The underlying product issues (metric drift, export failures) are aggravated by the perception that support is unresponsive. Halberd Financial Group contrasted response times unfavourably with competitor Coriolis's sales process. Resolving the open ticket is not enough — the relationship needs a reset at a senior level this week.
3. Tinker Supply Lab's recovery is fragile.
Recent product improvements stopped a churn decision, and the account's sentiment is positive. But a -22% usage decline and a health score of 53 do not yet reflect that recovery in the data. Without proactive outreach to validate adoption, this account could slide back silently.
Recommended Actions
| Priority | Account | Action | Owner | Deadline |
|---|---|---|---|---|
| Immediate | Halberd Financial Group | CCO or VP-level call this week; bring the product team lead for Fetch Radar; come with a dated fix plan for metric drift and export failures | CCO + Product | By Wednesday |
| Immediate | Halberd Financial Group | Escalate the three-week-old critical support ticket to a named senior engineer; close the loop before the leadership call | Support Lead | Today |
| This week | Nook & Bolt Technologies | CSM to run a working session on Fetch Radar forecast accuracy using the account's own seasonal data; address export format gaps with a near-term roadmap commitment or workaround | CSM + Product | By Friday |
| This week | Tinker Supply Lab | CSM check-in call to confirm whether the -22% usage drop is a workflow change or a disengagement signal; document outcome | CSM | By Thursday |
No accounts are currently flagged for executive review. Given the Halberd Financial Group situation, consider opening a flag now to ensure visibility if the two-week ultimatum is not resolved.
04
05
CALL THE BRIEF
Get up to speed via voice mode
The same brief, read aloud. Front Desk fetches this week's run and summarises it in a sentence or two, then takes follow-up questions on it, so you can ask what is driving a number instead of scrolling back for it.
Your browser will ask for microphone access.
Calls are capped at 5 minutes.
Other labs
The same customer feedback runs through each of these.
See
The Observatory
Every comment, mapped by theme. Explore the map AskThe Listening Post
Questions answered in customers' own words. Ask a question ConnectThe Signal Line
Your feedback, inside any AI assistant. See the tools ProveThe Proving Ground
Agents tested on hard conversations. See the verdicts TalkThe Front Desk
A voice agent you can call from the page. Talk to itThey're the working parts of Customer intelligence. To see what AI assistants say about your own business, get the free audit.