Conversion Performance Analysis

CRO Analysis examines conversion funnel behaviour to identify why visitors complete or abandon each step—delivering specific, actionable recommendations to improve conversion rates.

Quick start

1.     Open the Conversions Dashboard and select a conversion path

2.     Set the date range (30 days minimum recommended)

3.     Click AI Analysis

4.     Review step-by-step results as they appear (typically 10–30 seconds per step)

5.     Add priority recommendations to Kanban for implementation

Understanding the analysis

CRO Analysis runs in two phases. First, it examines each step in the conversion path individually. Then it analyses the complete journey to identify cross-step patterns.

Step analysis

Each step receives focused analysis using three data sources:

Cohort comparison — Compares behaviour between sessions that convert and sessions that drop off. This reveals what successful visitors do differently.

Zone visibility — Examines whether the conversion element (button, form, link) is positioned where visitors actually scroll. A CTA below the drop-off point cannot convert visitors who never see it.

Channel performance — Shows how different traffic sources perform at each step. Paid traffic behaving differently from organic often indicates a messaging mismatch.

Journey analysis

After individual steps complete, journey analysis examines cross-step patterns:

Channel attribution — Traces how visitors arrive (first touch) and where they convert (last touch). This reveals assisted conversions and journey complexity.

Cohort divergence — Identifies the step where converter and drop-off behaviour diverges most significantly—your primary bottleneck.

Off-path exploration — Tracks pages visitors view outside the main funnel. Pages with high conversion lift are helpful; pages where visitors exit are friction points.

What should I do next?

Priority levels indicate when to act:

Reading the results

Priority improvements

Each recommendation includes:

Cohort insights

This section compares converter behaviour against drop-off behaviour. Key differences reveal what successful visitors do that unsuccessful visitors do not. Early engagement signals from the first moments on the page often predict conversion success.

Zone visibility

This section examines whether visitors can see the conversion element. If the CTA sits at 65% of page height but drop-offs exit at 50%, most visitors never reach it. The recommendation provides specific repositioning guidance.

Channel insights

This section identifies best and worst performing traffic sources. Large differences between channels (organic at 14% click rate versus paid at 2%) indicate messaging or targeting problems rather than page issues.

Adding recommendations to Kanban

6.     Find the recommendation to action

7.     Click Add to Kanban

8.     Select the project board

9.     Task creates with all details pre-filled

The task includes the step name, page URL, current state, recommended change, justification, and priority—everything needed to brief a developer or copywriter.

Using filters

Conversion path

Select the conversion path to analyse. Each path represents a sequence of steps leading to a specific conversion goal.

Date range

Longer date ranges provide more data but take longer to process. For paths with low traffic, longer ranges produce more reliable insights.

Troubleshooting

Analysis takes too long or times out

Switch to a 7-day date range and retry. High data volumes can exceed processing limits.

Insufficient data messages

The step or channel lacks enough sessions for confident analysis. Expand the date range or wait for more traffic. Insights marked "directional" have limited confidence but indicate patterns worth monitoring.

Recommendations seem too basic

When competitive gaps are small or the page already follows best practices, recommendations focus on refinements rather than fixes. This indicates a well-optimised funnel.

Step shows as failed

The page content may be inaccessible or the step has no data for the selected period. Verify the page URL loads correctly and that tracking is active.

FAQ

How long does analysis take?

Each step takes 10–30 seconds. A three-step path with journey analysis completes in 1–2 minutes.

Can I run analysis on a single step?

Analysis runs on the complete path. Individual step results appear progressively, allowing review before journey analysis completes.

What does "visibility gap" mean?

Visibility gap measures the distance between where the CTA sits and where drop-off visitors stop scrolling. A 15% gap means the CTA is 15 percentage points below the typical exit point—most drop-offs never see it.

Why do converter and drop-off numbers not match my totals?

Analysis uses session-level metrics. A single visitor may have multiple sessions, some converting and some not. The analysis examines session behaviour, not visitor behaviour.

How does the AI select what to analyse?

SERP360 prioritises large behavioural gaps between cohorts, visibility problems, channel performance variance, and comparison against target conversion rates. The analysis surfaces the most impactful issues.

Can I ask follow-up questions about the analysis?

After analysis completes, a chat interface appears. Ask Tia about specific findings, request clarification, or explore alternative recommendations.

Getting help

Contact support for tracking or data issues