Google > Case Studies > Sellpoints moves beyond last-click attribution to double performance of non-brand paid search and triple performance of display

Sellpoints moves beyond last-click attribution to double performance of non-brand paid search and triple performance of display

Customer Company Size

SME

Region

America

Country

United States

Product

Google Analytics
AdWords

Tech Stack

Multi-Channel Funnels
Position-based model
Time-decay model

Implementation Scale

Enterprise-wide Deployment

Impact Metrics

Cost Savings
Revenue Growth

Technology Category

Analytics & Modeling - Real Time Analytics

Applicable Industries

Retail

Applicable Functions

Sales & Marketing

Use Cases

Predictive Quality Analytics
Demand Planning & Forecasting

Services

Data Science Services

About The Customer

Sellpoints is an expert in online selling, founded in 2000 and based in Emeryville, CA. The company aims to help its global clients make informed marketing decisions and drive performance improvements. They wanted to obtain a more accurate understanding of users’ paths to conversion. The agency understood that those advertisers who give all the credit for a sale only to the final click along the conversion path aren’t getting a true reflection of reality. They realized the need for a more comprehensive analysis that takes into account the entire conversion path, from the initial search to the final click.

The Challenge

Sellpoints, an expert in online selling, wanted to help one of its global clients make informed marketing decisions and drive performance improvements. The agency aimed to obtain a more accurate understanding of users’ paths to conversion. They realized that attributing all the credit for a sale to the final click along the conversion path doesn't provide a true reflection of reality. Conversion paths are complicated; a user may begin with a search using generic terms, see and click on a display ad, perform another search later using narrower branded keywords, click on a paid search ad and eventually convert after a wide variety of online actions. A last-click analysis ignores all of these actions except the final one.

The Solution

Sellpoints used Google's range of tools to move beyond the last click. The agency used Multi-Channel Funnels, a feature of Google Analytics, which can also incorporate Display impressions, to show how users convert across a range of online channels. It also applied insights into customer behavior to create models tailored specifically to its client, using a combination of a position-based model and a time-decay model. In valuing each step along the conversion path, Sellpoints’ analysis turned up a number of important discoveries. It determined that in a last-click model, 30% of conversions were attributed to brand keywords. Using its combination of position-based and time-decay models, Sellpoints was able to demonstrate that 40% of these in fact had started on a non-brand interaction.

Operational Impact

Highlighted brand and non-brand interaction – 40% of conversions attributed to brand terms in last-click model had actually started on non-brand interactions
Improved display measurement – while last-click model reported display CPA as eight times higher than goal CPA, other attribution models showed that display CPA in fact outperformed goal CPA
Refined marketing spend according to findings – doubled non-brand paid search performance and tripled display performance

Quantitative Benefit

40% of conversions attributed to brand terms in last-click model had actually started on non-brand interactions
Display CPA outperformed goal CPA, contrary to last-click model's report
Doubled non-brand paid search performance
Tripled display performance