Recognising a user across channels can be a challenge. But if you’re able to do so, you can answer some interesting questions:

• What part of my offline sales can I attribute to online orientation behaviour?
• How many online orientations result in an offline purchase?

For cross-channel reports, it’s important to collect the right data. If you do so you can answer the two questions with ease. The most important question here is: how do you identify users across channels?

The Cross Channel challenge

We’ve noticed the percentage of users that orient online and purchase offline has been on a steady rise over the past few years for our clients. Right now, companies often collect online data and offline data, but they aren’t directly connecting the two:

A simplified cross-channel journey.

In this simplified journey, we have an online visit that results in an orientation event. Afer that, we have an offline purchase journey that results in an offline sale. Without the connection, it’s hard to tell how online website behaviour attributes to offline sales. But if you do connect them, you’ll be able to directly attribute offline sales to online campaign sources. Wouldn’t that be great?

Solving the challenge with a customer identifier

Luckily, there’s a solution to our problems: the customer identifier. This is a value that allows you to recognise a user both online and offline. Potential values for identifiers are:

• Customer ID
• Email address
• Home address
• Postcode
• Phone number

By adding a customer identifier, our view of the cross-channel behaviour will be improved:

The improved cross-channel journey.

In the example journey, the customer identifier allows us to match online behaviour to offline purchases.

Selecting the right customer identifier for your business

If you’re looking for a customer identifier for your website or client, there are two main categories to choose from:

1 Broad identifiers

These values are suited for products that have a long buying cycle and aren’t bought that often by users. Think about bigger expenses, e.g. a kitchen, a bed or a fridge. Values like a postcode work well here, as there’s only a low chance that two people from the same street or neighbourhood will buy a product within a small time range.

2 Narrow identifiers

These values are suited for products that have a short buying cycle and are often bought by users. Some example products are clothing or groceries. With these products, there’s an increased chance that two people from the same area will buy the product within a short time span. Here, values like email address, phone number or full address come into play.

Keep in mind that the narrower the identifier is, the higher the chance is that it contains Personal Identifiable Information (PII). This raises privacy questions and not all web tracking tools allow you to collect PII, e.g. Google Analytics.

The importance of the customer identifier collection moment

A good place to look for your identifier is your offline sales data. It is often harder to change the data collected with these sales than to change the data collected on your website. So if you find a good identifier in your offline data, I suggest using that value for your online data collection as well.

From the moment that you know what your customer identifier will be, it’s time to start collecting that value online. The moment that you collect it is of great importance. Collecting the value early on in the orientation process will greatly increase the amount of data you can use. So try to collect your identifier as early as possible on your website to get the biggest pool of users for your cross-channel analysis.

Analysing the data

When you start analysing the data there are some important things to keep in mind:

• When matching online behaviour to offline sales, make sure that the online orientation happened before the offline purchase.
• Analyse the time between the orientations and the purchases and select a maximum time gap. If the time gap between the offline purchase and online orientation exceeds this threshold, don’t link them.

If you take this into account, you can start generating interesting metrics:

• Percentage of matched offline revenue: matched offline revenue / total offline revenue.
• Percentage of online orientations that resulted in an offline sale: matched online orientations / total online orientations.
• Average orientation value: matched offline revenue / matched online orientations.

These three metrics will give you an understanding of how your online website visits attribute to your offline sales.

Starting a customer identifier project

At Greenhouse Group, we analyse the impact of online orientation on offline sales in two steps:

• Manual analysis to see if there is an effect. If we see an effect, we continue to step 2.
• Automated analysis to get the insights on a daily basis.

The first step helps us and our client understand the cross-channel impact: is there an effect, and if so, how big is the effect? In this step, we manually connect the identifiers of the online data and offline data. The second step automates the analysis and allows us to use cross-channel data in our day-to-day performance analysis.

Conclusion

The cross-channel challenge can be solved with customer identifiers. All you have to do is select the right identifier for your business and start collecting it both online and offline. The data will help you understand how your online website attributes to offline purchases. The question is: what identifier will you use?