Multichannel Analytics: How to Track the Effectiveness of Traffic Channels

Tracking advertising effectiveness is easy if users leave a request when they first visit the site. But if one visit isn’t enough for the user, then how can you track the effectiveness of traffic channels?

End-to-end analytics helps you analyze traffic sources – from ad clicks to purchases. The service collects data from advertising platforms, CRMs, and websites and combines the information in a single window. Any business with more than one source of traffic should use end-to-end analytics. It will help you understand which ads are getting bids, and which ones are wasting budget. Ineffective advertising should be optimized or disabled.

However, if you only track bids and sales from each traffic source, you can disable channels that do not bring leads and customers directly, but help customers learn about your products and services. End-to-end analytics data is complemented by multi-channel analytics.

What is Multi-channel Analytics

Before a purchase, a customer usually comes into contact with a business several times: they come to the site from advertising, go from social networks to read blog articles, and receive an email newsletter. This is especially common for businesses that offer products or services at high prices. As a result, the user may see your ads several times, and leave a purchase request after switching from a conditional free SEO-issue. This chain of user interactions is called a “multichannel sequence.” 

According to a survey by Online Market Intelligence (OMI), more than 70% of 18-24 year olds use multiple search channels: 58% cited at least two channels for searching for a product or service online, and one in five said more than three. 

All traffic sources are involved in attracting leads. The traffic channel may not lead directly to the request, but introduce the future customer to the product, this fact should be taken into account. Otherwise, you can mistakenly conclude that SEO is the best way to attract new customers and turn off campaigns that bring traffic to the site. Use multi-channel analytics to understand the effectiveness of all traffic channels, ad campaigns, ads, and keys.

So you need to track all user interactions. But how do you determine the value of each channel?

There are different models of customer interaction. For example, a company that wants to reach as many users as possible cares which traffic channel brought the visit to the site first. For a company that wants to shorten the customer’s path to purchase, it’s important to track the channels which led the customer to purchase.

Attribution models will help make a report tailored to the specifics of the purchase funnel of a particular business.

What are Attribution Models: What are They and How Can They be Used for Estimating Traffic Channels?

An attribution model is a rule used to distribute value among the traffic channels that participated in attracting customers.

First Click

The first click in the customer interaction chain is assigned 100% of the value; the others are ignored. For example, the user first came to the site from a targeted Facebook ad and then came to the site a few more times from other ads. When a user leaves a request, the Facebook advertising channel will get all the value; all other interactions with the user will not be taken into account.

This attribution model helps identify advertising channels that increase business visibility.

Last Click

In this case, 100% of the customer engagement value is attributed to the last click. A user can go to a company’s website multiple times, click on multiple ads, and at the end, go to the site from an organic lead and leave a request. The SEO channel will get all the engagement value.

This attribution model helps determine which traffic channel ultimately converts a user into a customer.

Last non-direct click

All 100% of the value is received by the last paid click—for example, targeted ads on Facebook or contextual ads on Google.

Linear model

All traffic channels in the click chain have the same value. For example, if a user came to the website from 4 traffic channels, each channel will get 25% of the value, and if a potential client interacted with the business 20 times – each channel will get 5% of the value.

Attribution model based on the age of interactions (“Time Decay”)

The last click that led to a purchase has more value, the first click has less value. For example, a user first came to the site from Yahoo ads, then came to the site two more times from social networks, until he bought the product after switching from Google Ads. The Yahoo ad channel gets 10% of the value and Google Ads gets 40%.

Attribution based on position (“U-Shape” model) 

The first and last click of the user gets 40% of the value, the remaining 20% is distributed evenly to all other sources of traffic. For example, a user first came to the website through a YouTube ad, then came two more times – from Facebook and Instagram ads, and finally he came through Google Ads and left a request. YouTube and Google Ads traffic channels will each get 40% of the value and the rest will share the remaining 20%. That is, both Facebook and Instagram will get 10% of the value each.

This attribution model will give more value to the first and last channel in the chain – the traffic source that introduced the user to the business, and the traffic source that drove the lead to a purchase.