June 26, 2008

Paid Search And Catalogs

So many of my Multichannel Forensics projects now include both referring URL information and catalog / e-mail promotional history.

When you have this type of information, you quickly notice that customers blend advertising strategies into a slurry of confusion that results in the same purchase the customer used to place with you fifteen years ago.

This caused our industry to dive head first into matchback analytics. We try so hard to allocate every order that happened in the past.

It might be time to view the future.

In other words, we can measure past relationships, modeling them to see what a customer might do in the future.

For instance, I notice that some customers use paid search and catalogs as a combined effort, then use paid search and e-mail as a combined effort, then use paid search, then simply purchase without the benefit of any advertising.

Identify these customers, mail fewer catalogs to them, and focus ad spend on customers who require various forms of marketing to place orders.

4 comments:

  1. Anonymous11:34 PM

    Kevin,
    Forgive the long comment but I wanted to address your recurring theme. In many of your posts you advocate fewer catalogs. By your own experience you dropped a catalog division and still showed a profit by successfully converting to online marketing and online stores. (Congratulations!)
    Bloomingdales recently dropped their catalog while Cornerstone announce financial difficulty this week. With catalog choice and rising fuel costs pressure is on for scaling back or dropping catalogs.
    So....where do you think catalogs shouldn't be dropped or scaled back? Certain regions of the country? Certain industries?
    From the vendor side, I don't have a dog in this fight, although the companies I've worked for have sold to catalogs, shopping channels, .com's, and retail stores.
    Now I usually only see overall sales, so I don't know if a catalog or a shopping channel has a higher percentage of online purchases. But here is how I would rank their promotional effects as I see them.

    1. Dept stores still are biggest draw of impulse purchasers. That may change, but a good promotion leads to high unit sales.
    2. Shopping networks are 5-10 minutes of focused on a particular item. That makes it so powerful,esepcially with right slot and length. However, the channels have high benchmarks so what would seem high number of units to a vendor doesn't always translate as "profitable" to the channel.
    3. Catalogs typically focus on a set number of skus. I think there is still customer loyalty there as there are many repeat buyers. The campaigns lead to a burst of unit sales that levels off after 4 weeks.
    4. While .com business is growing, the number of skus on any retailer website is astounding, especially when retailer chooses to set up as drop ship from vendor so little inventory carried. This limits impact of say a sale on category, like watches where there are over 1000 skus to choose from. While I will see a 1-5 day spike in category or holiday sales, overall the various websites are very slow sales rate and to a vendor in a drop ship program, not very profitable, yet(online is growing). You could argue that is due to retailers we deal with but it seems across the board with several .coms.
    Thanks again for thoughtful posts.
    K

    ReplyDelete
  2. Non-traditional segmentation schemes make it reasonably easy to see who should continue to get catalogs.

    In many cases, these folks are rural customers who order over the phone.

    You'll never see these folks with a traditional RFM segmentation scheme --- rural and urban folks are all blended together.

    ReplyDelete
  3. Kevin --

    Consider posting a case study or stats on the interaction of PPC and catalog. Could be a client, obfuscated, or a set of clients, aggregated. Would be very interesting -- perhaps even (w/ client permission, of cource) a co-project w/ RKG on a common client?
    Very important interesting stuff, do blog more on this!

    :)

    Alan

    ReplyDelete
  4. Ok, more information will be coming in the future!

    ReplyDelete