Tineco put customer language and paid performance on one evidence table.
Customer conversations and media performance stopped living in separate narratives, giving the overseas marketing team one place to investigate a shift and shape the next brief.
2 hours
Time to investigate a performance shift end to end
2 weeks
Weeks between a theme first appearing in reviews and reaching a brief
9 in 10
Claimed theme-to-performance links that survive review
“First system that puts VOC and paid data on one evidence table.”
Company
TinecoFloor care · consumer electronics
Region
Overseas markets, US and Europe
Team
Overseas marketing
Use case
Voice of customer · Paid performance
10use cases running on PanoverseClient statement
01USE CASE
Investigate a performance shift in hours, not days
The problem
Customer language lived in conversations and media performance lived in dashboards. Each was useful, and neither alone explained why a message had started to move the market.
The result
The team investigates from one shared fact base and writes the next brief without translating between a VOC report and a media report.
Time to investigate a performance shift end to end
3 daysBefore Panoverse
2 hoursToday
Goal
Make a customer theme and a performance movement checkable against each other, instead of reconciling two reports that were never built to meet.
Deployed
Customer conversations, reviews and support language on one side; creator and paid delivery on the other, joined on the same timeline at the account's own daily granularity.
How it works
The join is on time and product line, not on sentiment. A sentiment score compresses the thing worth reading — the actual words — into a number that cannot be argued with, and an argument is exactly what a brief needs.
Both source trails are retained behind every recommendation: the conversations that made the theme, and the delivery rows that made the movement. A reader who disagrees with the link can open both sides and say where it breaks, which is the only form of evidence that survives a disagreement.
Result: held until the client approves it for publication.
02USE CASE
See a customer theme five weeks before it hits the numbers
The problem
By the time a theme is large enough to move a performance number, the window in which a brief could have been written around it has mostly closed.
The result
Themes are graded on rate of change against the account's own preceding weeks, so a theme is raised while it is still small enough to be early.
Weeks between a theme first appearing in reviews and reaching a brief
9 weeksBefore Panoverse
2 weeksToday
03USE CASE
Stop calling two things that rose together a cause
The problem
Two things that rose in the same month are not related by that fact, and a deck that places them side by side invites everyone to assume they are.
The result
The alignment is stated as dates and retained rows, with an explicit note of whether the account was active in that window.
Claimed theme-to-performance links that survive review
3 in 10Before Panoverse
9 in 10Today
04USE CASE
Write the brief in the words customers actually used
The problem
A brief written in internal category language competes against copy written in the language customers actually use, and loses to it on the same page.
The result
The brief carries the verbatim phrasings the theme was built from, so the writer starts from what was said rather than from a summary of it.
Briefs written from verbatim customer language
14%Before Panoverse
86%Today
05USE CASE
Send each complaint to the team that can actually fix it
The problem
Both arrive in the same channel and read the same way, and the team that can fix one has no authority over the other.
The result
Complaints are separated by whether the language describes the object or the claim, and each set is routed to the team that can act on it.
Complaints routed to the wrong team
38%Before Panoverse
9%Today
06USE CASE
Find out whether the market repeats your claim back
The problem
A claim can run for a full campaign without ever entering the vocabulary customers use, and nothing in the delivery data reports that.
The result
The claim is tracked as a phrase in customer language over time, so adoption and non-adoption are both visible as a series.
Campaign claims that showed up in customer language within 8 weeks
1 of 6Before Panoverse
4 of 6Today
07USE CASE
Stop averaging markets that are arguing about different things
The problem
An account-level summary averages markets that are arguing about different things, and the average describes none of them.
The result
Language is held at product line and region, so a theme that is local stays local instead of being read as a global signal.
Regional themes mistaken for global ones, per quarter
5Before Panoverse
0Today
08USE CASE
Open both sides of any recommendation you disagree with
The problem
A recommendation that cites a theme and a number, with neither openable, has to be taken on trust — and taken on trust is how a wrong one survives.
The result
Every recommendation links to the conversations and the delivery rows it was built from, on both sides of the join.
Recommendations challenged and overturned in review
40%Before Panoverse
12%Today
09USE CASE
Retire a message when the market stops repeating it
The problem
A message is usually retired when the team tires of it, which is not the same date as when the market did.
The result
Retirement is proposed against the phrase's own frequency curve in customer language, with the turn shown rather than asserted.
Messages retired on evidence rather than on fatigue
22%Before Panoverse
81%Today
10USE CASE
Run the whole investigation in one table
The problem
Every switch between a VOC tool and a media tool costs a reconciliation, and the reconciliation is where most of the investigating time went.
The result
The investigation runs on one table, so the time goes into the question rather than into making two systems agree.
Tools opened per investigation
4Before Panoverse
1Today
02WHAT RUNS TODAY
The system that is deployed
Panoverse aligned an emerging customer theme with the matching performance window and retained both source trails behind the recommendation.
03STARTING POINT
The state before any of it
Customer language lived in conversations while media performance lived in dashboards. Each source was useful, but neither alone explained why a message was beginning to move the market.
04HOW IT WAS MEASURED
What the number rests on
Evidence kept
Customer conversations and paid-media performance
What fired
Customer theme aligned with a performance movement
What changed
Investigate and brief from one evidence table
What was measured
Both source trails retained in the recommendation
Caliber
Client statement
The result shown is drawn from a client statement or from a backtest against the client's own history. Results vary with data, execution and market conditions.
05HOW THE TEAM WORKS NOW
The operating model
The team could investigate from a shared fact base and write the next brief without translating between separate VOC and media reports.
