How do you measure search result sentiment over time?
Search result sentiment is measured over time by classifying every ranking URL on priority queries as positive, neutral, or negative, then aggregating into a rank-weighted score that accounts for position (higher slots count more). Capturing this data at a consistent cadence, weekly or monthly, produces trend lines that reveal narrative drift before it becomes obvious and validate whether specific interventions have moved the needle. The same approach runs inside AIQ™ for AI engine responses, producing per-engine sentiment trends alongside the SERP view.
Tracking sentiment over time converts a snapshot observation into an operational signal. The trend, not any single measurement, is what reveals whether the brand’s narrative is improving, stable, or deteriorating across both search results and AI engine responses.
Step 1: Capture every ranking URL at consistent intervals
For each priority query, IMPACT™ captures every URL appearing on page one and page two at consistent intervals. Because search result positions shift continuously, a single snapshot is not sufficient, the cadence (daily data with weekly or monthly reporting cuts) is what makes trend comparisons valid over time.
Step 2: Score each URL on its actual content
Each captured URL is sentiment-scored based on its actual content, headline tone, body framing, and overall impression, rather than on surface signals like domain name alone. The classification axis is positive, neutral, or negative. Consistent methodology across every measurement period is what keeps trend lines comparable.
Step 3: Aggregate into a rank-weighted SERP sentiment score
The per-URL scores are combined into an aggregate sentiment figure for the full SERP, weighted by ranking position: URLs in higher slots contribute more to the aggregate than lower-ranked results, reflecting their greater share of first-impression exposure. This weighted average is the number that gets trended over weeks and months.
Step 4: Apply the same logic to AI engine responses
Inside AIQ™, every response from the eight AI engines AIQ currently monitors is sentiment-scored per engine using the same positive / neutral / negative axis. The per-engine trend lines reveal which models are improving or worsening in their narrative treatment of the brand, and whether intervention work (new content, source-layer remediation, entity updates) is registering differently across engines.
What the trend lines reveal
- Narrative drift: A gradual shift in aggregate sentiment, even when no single new hostile article is obvious, indicates that the mix of ranking content is changing in a way that will eventually affect stakeholder perception. Catching drift early allows intervention before the picture becomes entrenched.
- Intervention impact: When sentiment moves in response to specific work, a new article ranking, a source-level remediation, an AI entity update, the trend line provides evidence that the intervention landed. Without a trend baseline, impact attribution is largely anecdotal.
- Per-engine divergence: AI engines can score the same brand’s narrative differently, since even minor variations in prompt phrasing can shift classification results. Tracking per-engine trends reveals which models require different source-layer or content approaches.
The discipline: consistency over precision
Sentiment classification is imperfect on nuance, context, and tone, no methodology eliminates all noise. The value of sentiment tracking in reputation work comes from applying the same classification consistently over time, so that movement in the trend lines reflects actual narrative change rather than measurement variation. The trend is the signal; no single snapshot should be treated as a definitive assessment.
Last reviewed: 19/05/2026
