Case 02 · GMV Monitor / Advertising Analytics

GMV means gross merchandise value: the total value of transactions.

Thousands of creative rows.
A clearer next decision.

Problem
Across TikTok Shop campaigns, thousands of creative records made review priorities hard to identify.
Decision
Designed a path from campaign totals to creatives, videos and daily evidence.
Verified outcome
Verified queries across 109 campaigns using live account data.
My role
Requirements · Product & UX direction · AI-assisted development lead
Production stack
Google Apps Script · JavaScript · Official TikTok Ads MCP
Advertising Analytics Web App · GMV Monitor · Actual app screen · Business values, accounts and dates redacted
Actual app screen · Business values, accounts and dates redacted
Campaigns queried
109Verified with live account data
Creative records handled
~34K rowsIncludes video–product combinations
Workspaces
4Overview · Diagnosis · Growth · Reports
FROM DATA TO DECISIONS

From displaying more data
to directing attention.

An aggregate ROI cannot tell a team which creative to inspect. The product moves from an overview to a focused shortlist, then to the content behind the numbers.

  1. 01

    Overview

    Review campaign performance and changes

  2. 02

    Shortlist

    Narrow creatives with search and spend filters

  3. 03

    Inspect

    Compare video context and daily trends

  4. 04

    Share

    Export consistent summaries and CSV reports

PRODUCT IN USE

An analysis experience
beyond the headline number.

Settings, results and content stay connected, with an information hierarchy that works across desktop and mobile.

Performance diagnosis · Actual app screen · Business values, accounts and dates redacted
Performance diagnosis ↗Current settings remain distinct from historical results.
Daily creative trends · Actual app screen · Business values, accounts and dates redacted
Daily creative trends ↗Move from period totals to daily evidence for review.
Mobile hierarchy · Actual app screen · Business values, accounts and dates redacted
Mobile hierarchy ↗Key metrics and the next action remain easy to find.
PRODUCT JUDGMENT

AI can write the code.
Metrics still need judgment.

A useful dashboard also prevents misleading interpretations. Data definitions and decision boundaries are part of the product.

01

Keep totals separate

Search and top-100 filters narrow the creative list. They do not silently change the denominator for campaign-wide performance.

02

Preserve uncertainty

Unavailable data and errors remain unknown, not zero. Current targets are not retroactively applied to historical results without setting history.

03

Separate analysis from action

The app focuses on querying, analysis and reporting. A person reviews the evidence and makes advertising changes in TikTok.

CONSTRAINT → DECISION → VERIFICATION

An API constraint.
A decision we could verify.

AI assisted implementation and review. At runtime, official data is collected and calculated in code; advertising decisions remain with the operator.

01 / CONSTRAINT

The request was rejected

The official API rejected a daily-dimension and video-filter combination. The feature needed a different supported query path.

02 / DECISION

Keep the source meaning intact

Query supported period reports, match the same video and preserve daily results separately from the period report.

03 / VERIFICATION

Reconcile before presenting

In the recorded sample, daily spend and order totals matched the period report. A small revenue difference was retained rather than overwritten.

Based on the recorded integration check in the detailed case study. This sample does not establish agreement for every campaign or period.

PERFORMANCE, OBSERVED

Less waiting
on repeat review.

Show campaign performance first, then request selected creatives and daily results on demand. Reuse only validated, complete results to address correctness and repeat-query latency together.

1,973 campaign creative rows

First collection16.302s
Cached query1.558s

Daily results for one video

First collection19.117s
Cached query1.329s

Observed server execution times: first collection versus a cached repeat query for the same feature. These are sample observations, not end-to-end page times or multi-user averages.

MY ROLE × AI

Translate a business question
into a buildable system.

I led AI-assisted development by connecting requirements, priorities and acceptance criteria, then checking the output against actual screens and data.

01

Own the problem and criteria

Turn PM requirements into features and flows. Choose official data, a read-only scope and usability priorities.

02

Implement and cross-review with AI

Use Codex and review agents across collection, caching and UI. Improve the product through observed errors and feedback.

03

Make validation part of delivery

Use recorded results from 286 passing regression tests, live-data and CSV reconciliation, and browser checks to assess behavior and performance.

From collecting data
to supporting decisions.

A product that connects operational understanding, information design, official data integration, performance and data integrity.

Scope and evidence

Grounded in app captures, development records and verification results. The 286 tests refer to a recorded passing regression run. Deployment and live queries with the administrator account are documented. Colleague mobile access and multi-user concurrency remain to be validated. About 34K rows does not mean unique videos. No ad revenue uplift or team time-saving percentage is attributed to the app. Initial large collections still have latency.