18/06/2024 10:08 AM

wildflowercafetahoe

Piece of That Fashion

Everything You Ever Wanted to Know About Matchbacks (and More)

Everything You Ever Wanted to Know About Matchbacks (and More)

Not all direct marketers are familiar with matchbacks and why they’re important. I thought it might be helpful to explain the matchback process and why it’s a critical tool for evaluating results and developing mail plans.

Matchbacks are the process of having your order file “matched back” against your recent mail files in order to give credit to the proper source or key code on a list-by-list, segment-by-segment basis. This applies to the housefile and to prospect lists. Without a matchback, a direct marketer has little way of knowing the origin of the orders they receive. This is because customers placing orders as a result of receiving a catalog commonly place their order online, rather than via phone where the call-center representative can try and determine the source of the order. It’s not uncommon today to track only 10 percent to 30 percent of your orders to a specific key code without the aid of a matchback. This is why the most recent trend is to rely 100 percent on the matchback to track results, abandoning the idea of trying to track to the source at the time of order entirely. Since you can only track 10 percent up front, why make the effort to try at all?

This is why matchbacks are a way of life for catalogers today. They’re automatic and routine. Matchbacks are the only way to tell what source generated an order and which key codes should be given credit for the sale. If you don’t run matchbacks, you’re flying blind.

How Matchbacks Are Done

The matchback process compares name, address and customer number from your orders file to the same on your mail files using similar logic as a merge/purge. All matched order records are reported on within the list/segment from which they were mailed. Only orders which match to the mail file, have an order date after the in-home start of the mailing and within your match window will be reported on as a match. A match window is a predetermined date range applied in the matchback process which covers the period which your mailing is considered active. After the matchback is run, you’ll receive a report which typically includes key performance indicators; source or key code, segment description, total circulation, total orders, total revenue, response rate, average order value and dollars per book.

Typical data needed to do a matchback include:

  1. Mail files from appropriate time frame.
  2. A listing of all key codes included in each mailing (often referred to as a List of Lists, or final counts by key code).
  3. Order records for the appropriate time frame. Desired fields include name/address, customer number, source code (if tracking up front), order date and order amount (product demand $).
  4. Other key codes if you’re tracking sales to a source at the time of order (e.g., email campaigns, affiliate marketing, bounce backs, initial catalog request, etc.).

There are some considerations to keep in mind when reviewing the results of your matchback. The results will be allocated to the most recent active mailing which occurred prior to the order date and within the match window. If there’s no match, the order activity might be allowed to match to an earlier mailing if it falls within your match window. If your business is highly seasonal, this will skew results to the mailing which was in home prior to and closest to the season peak. Therefore, the results of each mailing preceding a seasonal peak should be considered individually and weighed together in order to account for possible skewing.

Date ranges are important to accurate matchback reporting (aka your match window). How many weeks of order data should be allowed to match back? This is influenced by the format and number of pages of the mail piece, how often you mail your housefile, expiration dates (offer or mail piece), etc. A postcard mailing should have a narrower match window than an 84-page catalog, for example, because people tend to keep catalogs longer than postcards. Your service bureau can look at your sales data and help you determine the best match window to use for your business.

How frequent you should run a matchback really depends on how often you mail and how you’re going to apply the results and/or data to your marketing/circulation strategy. I feel a matchback should be run at least twice per year, depending on your degree of seasonality and how much testing you’re doing. For example, if your business if based on a fall/holiday season followed up with a spring season, doing a matchback twice a year when the season ends is probably adequate.

However, if you’re running a test and want to get an early read on the results, that would warrant an additional match. Some service bureaus offer the ability to matchback weekly. It’s nice to have the ability to matchback often, but probably not cost justified for most. Doing matchbacks with the same service bureau that is doing your merge/purge work is most common since it already has your mail files.

The cost to run a matchback is in the range of $2.50 to $4.00 per thousand with minimum fee levels (the quantity is based on the circulation of mailings included in the match, plus the number of records in the order file being matched). The base price often decreases if the matchback is done as part of a housefile update. Subscription prices which provide for the ability to matchback online weekly range from $2,000 to $4,000/month and up. This option is more expensive and more time consuming to manage for the mailer.

*If you have mailings that were not 100 percent complete at the time of the prior match, considering including that mailing again in the next match to make sure you have final figures. The additional cost is often justified.

In addition to gaining a better understanding of past mailings, matchback results can be used to mail at a detailed segment level. You can also expand your current segmentation beyond RFM (recency, frequency and monetary) to RFMC by adding channel of influence into your segmentation structure. If possible, your service bureau should apply channel of influence after the match is run. It’s more important that channel reflect order influence rather than just how the order was placed. Often internet-only buyers’ results aren’t as strong as catalog-influenced buyers (for comparable RFMs). The performance from internet-only segments might not justify receiving a catalog, or might only justify one or two mailings per year, for example. This is especially true for 13-plus-month, one-time internet-only buyers. It will be important to segment channel of influence out separately and read results before applying a contact strategy using this information.

Keep in mind that assigning the internet portion fairly is difficult. For example, if a customer purchases a product from the internet, they become an internet buyer. They will receive a catalog in the mail. They may make a second purchase from the internet. When a matchback is done, this customer will always be considered a catalog buyer after their initial purchase because they were mailed a catalog with a housefile key code identifier. While the catalog remains the largest driver of traffic to the internet, this does make it difficult for online marketing managers to verify that they’re growing their business and remaining on budget since, in this case, the credit is going to the catalog. This factor will impact the repurchase stats for the e-commerce business.

Matchbacks are necessary to have a comprehensive understanding of your results, which support planning, including contact strategy. Matchbacks are not perfect. However, if you keep in mind some of the influences discussed in this article and remember that the relative performance reported for each segment is reliable, then you can proceed with confidence knowing that the matchback results drive sound decisions.

Stephen R. Lett is the founder and chairman of Lett Direct, Inc., a catalog consulting firm specializing in circulation planning, forecasting, digital marketing, and analysis since 1995.