Getting the Big Picture With Web Metrics

Mar 10, 2008 10:54 PM  By

If you had told retailers twenty years ago that they would have the technology to follow everyone around their stores and record everything they looked at or put in their cart, they would have laughed at you. But that level of customer insight is exactly what Internet shopping has created.

But if merchants can’t integrate Web metrics data with their other channels, they only have part of their customer’s picture. Marketers must consolidate prospect and customer data, combine all the operational, attitudinal and transactional data they hold, and link it up with related campaign history.

That can form the basis for smarter decisions about what message should be sent to whom, when and through which channel. Marketers can then work out how much they should invest to get the return they need, whichever channel it comes from.

Not only does this mean that the marketing spend is optimized, based on an understanding of who is most likely to respond, but it also means that traditional and digital channels can be used in combination. The returns are greatest when companies can use their channels in the same way that customers do, switching between them as appropriate, to ensure there is a continuous rapport with customers and prospects.

As businesses capture more customer behavior than ever before, they can struggle to separate signal from noise. At the highest level, companies want to know what kind of response rate they should be achieving, whether that’s clickthroughs from an e-mail or sales conversions on the Website.

They want to be able to benchmark their campaigns, so that they can work out the likely return on investment for their marketing activities. They’re still feeling their way online, trying to work out how much traffic their Website deserves and how much will make it profitable.

The latest marketing analytics tools provide the kind of in-depth analysis that marketers need to obtain a full picture of customer behavior online. But you can’t look at the web in isolation, as many merchants did at the dawn of the Internet, when they created a new silo of customer data for online purchases. And as we know, there’s a world of difference between winning a brand new customer and rediscovering a customer who’s migrated from your phone or catalog channel.

When you’re overwhelmed by data and don’t know where to start, they call it ‘analysis paralysis.’ So you need to carve your data into meaningful chunks. You could, for example, get hung up on which corner of the screen gets the best response, but few companies would attempt that outside of a controlled focus group. For most, the Web page is the basic unit of measurement, and customer behavior is analyzed by department, product or product category.

That information can be combined with what you already know about customers from other channels. If customers have only bought books from you before, but spend a lot of time browsing your CD catalog online, that’s valuable intelligence. It might be time to start inserting the music flyer with your books catalog next time you mail them. For products with a long sales lead time, it’s essential to know what customers are thinking about well in advance and the best clue is to know what they are reading.

One thing that companies often miss is the opportunity to enhance and verify data online. Users hate lengthy forms, but they don’t mind providing the odd answer now and then if it will improve the service you deliver. ‘Additional voluntary questions’ (or AVQs) can be presented when users log in – usually just once per day – and the answers can be used to enrich the user’s profile.

Indeed, some companies get their most valuable data this way. One client is a major newspaper publisher that segments its audience according to how often customers buy its daily paper. How does it know? It asks them an optional question about it. That information is virtually impossible to find out any other way; even if a customer always bought their paper at the same shop, it’s unlikely the seller would know their name or address.

Curt Bloom is managing director, International of smartFOCUS, a supplier of enterprise marketing management software for direct and database marketing, headquartered in Bristol, UK, with offices in the United States and continental Europe.