Predicting the results of newsletter direct mail. (DM Notebook).

"No one can predict the response to direct mail, and don't believe anyone who tells you that he or she can."

A useful truism, but not one that stops marketers and publishers from wanting to know, in advance, what sort of return they can expect on the next drop.

What you cannot know is whether a launch mailing for a new title is bringing prospects the word about a newsletter they hadn't realized they had always wanted. Perhaps the publishers of Louis Rukeyer's Wall Street had a good feeling that this one was going to be a winner when they launched, but you just can't predict if the world wants another newsletter on women's health or the very first title about the Welsh Corgi breeding business.

For an established title with a track record, however, you can make at least an informed guess about future response. Beyond economic conditions and any other variables beyond the control of the marketer, there are a number of factors you can control that do affect return.

Seasonality

I've seen a number of studies on seasonality, done by The DMA and others, and they track remarkably consistently. January is always the best month. The other months of the year show the following response rates (expressed as drop-out from the January rate of 100).

* February--5 percent

* October--10 percent

* August--15 percent

* November--20 percent

* December--20 percent

* September--20 percent

* July--25 percent

* March, April, May, and June--30 percent.

I've never seen a study that didn't include consumer mailings. For b-to-b direct mail, I suspect the entire pattern is somewhat flattened. People work 12 months of the year. I suspect for business mailings December and August are worse than these stats show. The "holiday rush" isn't good for b-to-b offers, and there is a vacation dead zone, but the March-June period is better than this chart makes it look.

An old bromide had it that the best three days of the year for newsletter direct mail to drop are: the day after Christmas, the day after Labor Day, and (three years in four) the day after Election Day (the theory being in the Presidential year the public's attention is exhausted by the campaign).

A number of newsletter publishers I've worked with have found that while December 26 may be the best day of all to mail, March-April was also an excellent time to remail everything that worked on the previous mailing.

Other factors ranked

A longtime magazine marketing consultant, John Klingel, once gave me an analysis of additional factors affecting response on an established title. Is he right? He's the only one brave enough to try it.

Keeping things simple, Let's talk about a Labor Day mailing on a publication that pulled a 1.0 percent response in January. (Italics are my comments.)

* Seasonality--20 percent (This is right from the charts; I suspect a b-to-b mailing will do a bit better than this).

* List fatigue--30 percent (For the best lists for an established title, I'd be tempted to disregard this factor almost entirely; they can often be mailed effectively three or more times annually).

* Price increase or descrease--20 percent + or -. (For a b-to-b title I'd say a small increase that doesn't take you over a price point won't have much effect, but to get a 20 percent increase in response, a price savings or cut offer has to be big enough to be noticed. "Save $50,' on a $397 title probably won't do it).

* New package--10+ percent. (Hard on the vanity of consultants and copywriters that this might be the limit of their influence, but probably true).

* Premium offer-7.5 percent (If it's new. If you already are including a premium in the offer, as 80 percent of newsletter publishers do, a new or slightly different premium probably won't make this large a difference).

Does it work? Taking my assumptions, a Labor Day remail of a package which pulled 1.0 in January will get 0.8 percent response. Or, sexed up with a new package and a price savings offer, perhaps 1.1 percent. It certainly sounds reasonable.