Find agent skills by outcome

- Isolate anomalies and outliers confidently

and attribute them to network mechanics where causally plausible.

- Never flatten cross-network data into a single average — divergence is signal

not noise.

- Highlight early signals the model would treat as predictors per network (CTR → IPM deterioration o...

CPI drift patterns on Mintegral

- Always analyze creatives at two levels: within each network

and across all four networks simultaneously.

Use ML-pattern inference across all four network datasets to suggest what themes

angles

- Format-specific opportunities (e.g.

an endcard mechanic untested on ALN

- Predictive creative mechanics the data hints at (e.g.

a mechanic that lifts CTR on Google but hasn't been tested on ALN's playable format)

Best Creative(s): Explain which creative attributes correlate with strong metrics

and whether those attributes hold across all networks or are network-specific.

Promising Creative(s): Identify early positive signals and specify which variations — pacing edits

hook recuts

Worst Creative(s): Explain which patterns predict failure

and flag whether the failure is universal or network-localized.

- Which are candidates for format adaptation (e.g.

recut for Google's asset ingestion

- Rate divergence risk: High / Medium / Low — i.e.

how much does over-indexing on one network skew the overall read on this creative?

Business

Based on cross-network performance patterns

suggest a creative portfolio allocation strategy:

Research & Learning

- Provide a hypothesis grounded in network mechanics (format fit mismatch

audience signal difference

One concise pattern extracted strictly from this network's data — e.g.

On ALN

- State the performance delta (e.g.

top 1 on ALN

- Highest CPI: Explain which signals

specific to this network

- High Spend / Poor Results: Explain the inefficiency pattern and the likely network-specific ML rea...

ALN AXON fallback behavior

- Lowest IPM (or weakest CTR × CVR): Identify root-cause patterns through the lens of this network's...

weak hook on a skip-heavy rewarded placement

- Top Creative by IPM (or CTR × CVR for Google): Interpret why this creative wins on this specific n...

format fit

- Top Creative by Spend: Explain why this network's algo is favoring it

and whether scaling is amplifying or compressing efficiency.

Repeat the following block for each of the four networks: AppLovin

Mintegral

- Flag anomalies with ML-style reasoning (outliers

variance spikes

Your role is not to describe numbers

but to act as a performance-prediction model using structured