B-to-B Cataloging Every Cataloger's Guide to Analysis
By George Hague
Editor's note: This is the first of a two-part series on analytics and measurement. This month's column focuses on circulation essentials.
Response analysis sets direct marketers in their own class. Of all the marketing professions, we have the best opportunity to be precise in our response analysis and predictive in our forecasting.
Whether the analysis examines circulation efficiency or product profitability, direct marketers pull the numbers together to help guide their companies with strategic plans and forecasts projected from actual results.
For marketing analysis to be effective — for both circulation and merchandising — we first must look at three factors.
1. Understand your raw data. Direct marketers must always question the raw numbers. With raw data in hand, look through each column heading on your spreadsheet with your IT person. Do you understand what's being reported in every column?
Ask specific, exacting questions: Are returns already removed from these sales? Are you sure that shipping revenue isn't included in the product sales number?
If your questions don't cause your IT staff to double-check its queries, you're probably not asking specific enough questions.
This also is a good time to make sure everyone is on the same page with your definitions of housefile segmentation. It's not uncommon for a programmer to misunderstand that frequency and monetary criteria apply to the life of the customer, not to a 12-month or other predefined period of time.
2. Update your data. When you begin your actual data analysis, make sure you have the most recent, complete data.
For example, it's not uncommon for raw data to be pulled, spreadsheets to be built and numbers to be flowed in, only to wait for several weeks or months before planning the next season starts. By that time, your data will be out of date.
Use the most recent data for any analysis that directly will affect your planning. Building spreadsheets with lookup formulae allows you to update your spreadsheets quickly and easily. By using lookup formulae, you can update a prebuilt spreadsheet in minutes.
3. Organize data. Pull all your data into one spreadsheet. Have your keycodes, descriptions of your file segments, quantities and promotional costs, and other metrics collected and organized.
Raw Data Metrics
Here are the raw data metrics to be included in your spreadsheet.
- Keycode for each segment. Including your keycode for each segment is vital to allow you to automatically update your spreadsheets with lookup formulae.
Descriptions of each segment. Clearly describe each segment by recency, frequency and monetary value (RFM) criteria. Your keycode may include a structure that explains this data, but it'll be easier on everyone, including yourself, to have an actual description.
Quantity mailed per keycode. This is the total number of pieces mailed per keycode or segment mailed.
Offer. List any offer associated with any list or segment.
Number of orders per keycode. This ideally is the gross number of orders received. If your IT system automatically subtracts returns, that's OK. Just know that to analyze your returns, you'll need a separate report.
Gross sales per keycode. This number should represent your total product sales per keycode. You don't want shipping revenue included in this number.
Cost of goods sold per keycode. It's not uncommon for a company's reports to exclude this number. If your response reports don't include it, ask your IT department if it's available. You may be pleasantly surprised. If it's not available, use an average based on gross sales. The average should be stated in your analysis report.
Discounts. Whether your discounts are offer driven, at the discretion of your order response staff or part of a volume-purchase structure, ensure that your discounts are counted.
Number of returns per keycode. Returns are a category of numbers that are handled differently by different systems. Having return categories by keycode is the ideal. However, this often isn't the case. If you have this metric available, make sure you know if the number represents the number of orders or the number of products.
Gross dollars of returns. Dollars refunded or credited for returns.
Cost of goods of returns. The cost of goods associated with your returns, if available.
Calculated metrics are developed and calculated based on the above-mentioned raw data.
- Promotional cost. Determine your promo cost per segment, including postage, printing, list processing, mailing and list rental, if applicable.
Response rate. On a segment-by-segment basis, divide your number of orders by the total number of pieces mailed.
Average order value. On a segment basis, divide your gross sales by your number of orders.
Contribution per segment. Contribution is short for "contribution to overhead." It's different than a fully loaded profit/loss number. In this article, I define contribution as dollars left after taking your gross sales and subtracting your promotional expenses, cost of goods sold and any applicable returns or discounts. Some catalogers take it a step further and subtract their variable fulfillment cost, which is how much it costs you to fulfill an order after taking into account your shipping revenue. If you subtract your variable fulfillment cost but have trouble getting an exact number, use an average. (Twelve percent of gross sales is the catalog industry average.)
Contribution per order. Take your contribution-per-segment number and divide it by your number of orders.
Sales per book. Divide your gross sales per segment by the quantity mailed per segment.
Segment Your Segments
Your main segmentation categories are housefile segments and rented lists. Within these categories, further segment your lists by:
- House buyers. Segment your house buyers based on recency. For example, if your RFM segmentation is based on year of purchase, list all your 2006 buyer segments together with a subtotal for each column. Segment and subtotal down the list of 2005 buyers, 2004 buyers, etc.
House prospects. It may be helpful to segment and group prospects together for analysis. Run subtotals for all your columns. If you decide not to group your house prospects separately, include them in the appropriate year of the above housefile groups.
Rented cooperative database lists. Generally, segment and subtotal each collection of co-op lists separately from your other rented lists. You'll find this helpful when determining your continuations and discussing your results with your various co-op brokers.
Standard list rentals. Think in terms of subgroups that will aid in analysis. For example, if you rent from several brokers, some catalogers subtotal their lists by broker. These categories help them discuss results for continuations and new tests. Sometimes it helps to segment rented lists by product purchasers versus subscribers or requesters. If there's no logical subdivision, group them together.
Let the Analysis Begin
When you build your spreadsheet this way, your numbers tell you a story. Look at your house buyer segments by recency of purchase. A quick scan will tell you the most recent buyers perform significantly better than your two- or three-year buyers.
Are your older buyers still profitable? Look at the contribution per order metric. Should you test reactivation strategies?
Beyond their analysis value, these spreadsheets are a vital tool for strategic planning and forecasting. When you build spreadsheets to this detail over the course of a year, you come away with a complete set of response data that is invaluable for next year's planning.
In part two, I'll discuss merchandising analytics and how to use your findings to support your merchandising and product development teams.
George Hague is senior marketing strategist at J. Schmid & Assoc., a catalog consulting firm in Mission, Kan. Reach him at firstname.lastname@example.org or (913) 236-8988.