Case study · Path analysis

Fixing leaky website journeys for a telecom eCommerce team

Clickstream path analysis that found where visitors dropped off, raising conversion and satisfaction.

TelecomMarketing & Customer

The challenge

The eCommerce group of a telecom client was struggling with low conversion rates and low customer satisfaction on its website.

Used customer transaction and interaction data captured in Webtrends.
Built a custom algorithm in SAS to identify the most common paths visitors took.
Captured time spent on each page and customer satisfaction at each stage of the flow.

The client restructured its website based on the findings. The most common leakage paths were plugged and dead ends removed.

Customer satisfaction and conversion rates improved, and the analysis surfaced new upsell and cross-sell opportunities.

Event-level clickstream from Adobe Analytics or GA4 modelled in the lakehouse.
Markov-chain journey models that measure how much each step contributes to conversion.
An LLM that summarises open-text feedback alongside drop-off points, so teams see why visitors leave, not only where.

This describes our current toolkit for the same problem. It is not part of the original engagement.

Churn segments and models that showed a telecom client why customers were leaving, and where to act first.

Sentiment classification across social platforms so the care team could reach unhappy customers faster.

Time-series forecasts for 2,000+ SKUs that cut both stock-outs and overstock.

Tell us the decision, the data you have and the timeline. We will come back with how we would approach it.