Leveraging marketing attribution for ecommerce - SwiftERM

Leveraging Marketing Attribution for Ecommerce. Just because ecommerce has become reasonably ubiquitous does not mean that in-store retail shopping is dead. In many cases, ecommerce brands are opening brick-and-mortar locations. Companies such as Five Below, Burlington and Sephora have opened and continue to open physical retail locations at a surprising rate. They realise what Apple discovered several years ago: some products sell best when customers can touch and feel the merchandise — and because of that, using cross channel attribution to track consumer behavior online and offline is a necessity.

Why Ecommerce is Expanding to Bricks-and-Mortar

Many traditional retail brands have indeed either fallen on hard times or gone out of business. One reason for the demise of certain stores is the rise of ecommerce, but it’s also a result of poor decision-making in light of changing consumer behaviour.

Attribution for Ecommerce

While there are pros and cons to catering to consumers in an ecommerce and in-store environment, marketers must take a comprehensive approach to track behaviour. Consumers will engage in online and offline shopping and don’t feel the need to distinguish between them. If a consumer browses a website and eventually visits a store to make a purchase, it’s up to the retailer to understand and track that customer’s journey.

Many retailers have found that consumers also purchase online but pick up in-store. Intelligent brands are creating another reality that combines online and in-store experiences as a single journey. When done right, marketers can more easily track consumer behavior and measure how well their online and offline campaigns are doing.

Leveraging Marketing Attribution for Ecommerce

To properly track and measure online and offline behaviour, brands can use marketing attribution. Based on the different models available, attribution can work in both an online and offline environment. There are single-touch and multi-touch attribution models where online and offline behavior can be tracked using either approach.

The first-click attribution model gives credit for a sale to an initial engagement a customer has with a brand. This touch point is easier to measure if there is an online component, as it becomes a challenge to properly track a purchase that an in-mall ad may have influenced.

The last-click attribution model is used similarly as the first-click, giving credit for a sale to the customer’s final engagement before purchase. Again, it’s easier to track this online because the same challenge exists for properly tracking this offline.

Whether or not customers have one or multiple engagements, the above two models only track a single touch point.

Multi-touch models are easier to implement when tracking online and offline behavior. This approach can take into account multiple engagements a customer has when potentially combining online and offline shopping. Loyalty programs, credit cards or RFID information can be used to measure whether customers engaged in online behaviour, engaged in an in-store pickup, or sometimes visit a store to handle merchandise that is then purchased online.

The U-shaped, or position-based, model assigns credit to the first and last engagement a prospect has during the buying journey. In this model, 40% of the sales credit is given to the first and last engagements equally, while 20% is divided evenly between all other engagements. This can work very well when there is engagement in both environments. For example, there is an online coupon for an in-store purchase.

The Linear model assigns credit to all touch points equally, no matter how many there are, where they took place or where they fall on the continuum. The linear model does not identify which engagement was most responsible for the sale, but it can offer insight into a brand’s overall online and offline strategy.

The Time Decay attribution model focuses on engagements that occur leading up to the sale. The closer a touch point is to the sale, the more credit that touch point receives. With this model, retailers can see if online momentum drives offline purchases or vice versa.

The Custom or Data-Driven attribution model offers perhaps the most insight of any other model. This approach assigns different weights to individual engagements based on the unique characteristics of a brand’s sales cycle. While this model is the most complex to configure, it likely provides the most insight, especially when tracking online and offline attribution.

An example using the Custom model for cross channel attribution is tracking an in-store announcement for an online survey that then leads to an online purchase.

Which Attribution Model is Best?

There are different reasons why companies use one attribution model over another. For brands measuring both online and offline attribution, a multi-touch model is more insightful than single-touch models. Of the multi-touch models, the Custom model is ideal, as it provides the most insight regarding your customer’s buying journey as they move from online to offline and back online.

Whichever model you choose, you’re ultimately trying to get to the heart of what influences your customer to buy from you. When you obtain this knowledge through attribution, you can then assign more resources to those engagements that are most relevant and lead to increased sales.

 

We hope you enjoyed this article, intended to help improve our client’s profitability. It reflects the care SwiftERM offer. If you haven’t already done so, then please enjoy a FREE month’s trial of our predictive personalisation software on your site, and let us know what you think. Register for the free trial, or call us on 0207 998 3901 (US 260 410 3747), or book a call with us  Zoom ID 964 515 7464 .

Other articles of interest below:
(Index to all articles here)

Retailers evolving their tech stack
Live-streaming stressing real-time aspects
Predictive personalisation software to use now