That 17% return is for my monthly postcard d-mail. I sent out from 1000-2000 a month per store, with consistent returns in the high teens, and if I have a reputation here for anything it should be that I am really anal about numbers. I am pretty much mercenary in tracking my marketing numbers.
My monthly mailer comes directly from the POS database. I take my 30-60-90 and occasionally 120-150 day address with no orders and mail them a post card. The highest response by far is the 30-59 day no order group. Then it tapers off. I also graduate my specials depending on the group. Split testing showed early on that $off coupons pulled a few % better than free item deals in 3 of my 4 markets, so that is what I go with.
A normal mailer will consist of:
$2 off for new customers
$3 off for 30-59 day no orders
$5 off for 60-89 day no orders
$6 off for 90-120 day no orders
By the time someone gets 3 post cards 3 months in a row without ordering, they are probably never going to order again. I leave them in the database until 210 days with no orders, then I purge my inactive addresses. The reasons for that are many. The address might have had the grand kids over, they might have moved, they might have died, they found a hair in the pizza and didn’t call and complain, who knows. If they do order after 8 months of no orders, I would like to make sure they get a thank you post card anyway, so erasing them from the database for inactivity also triggers the thank you post card when (if) they order again. I have seen operators who brag about a 10 year customer list, I only shake my head. In terms of a good database, inactive data is a killer.
Split testing (Multivariate_testing) is something that the internet (google specifically) really helped perfect. It works just as well with database mailers as it does for google ad words. It boils down to sending out equal numbers of slightly different coupons.
For example, you have 400 customers who haven’t ordered in the last 30-60 days. Splitting them randomly and sending out 200 with a $3.00 off coupon and 200 with coupon for free bread sticks (or cheese bread, or any item with a $3 value), then measuring and comparing the results. You can hone your testing and realize a few % increase after only a few tests. When you are sending out a few thousand post cards a month, a 1-2% increase in redemption is huge.
Back to door hanging, that was one pillar of my front end marketing. I also used carrier route mailers (I consistently failed with these, and gave up after testing 4000 direct mail letters in 2 markets over 3 months with negative ROI each and every time I sent them out). Some of my effective marketing efforts are newspaper inserts and magazine ads. I had some marginal success with newspapers, but nothing consistently good.
In all of my testing and results frequency trumped reach each and every time. Door hanging is the shot gun approach. Blast an area with marketing materials. The only ways I know to increase your ROI there is to increase the frequency. Hitting an area 3 times in 10 days keeps it at the top of the mind. Funny thing is, in competitive areas, if you are the third pizza place to hit a customer in 10 days, it has nearly the same effect as if you had put your own fliers on their door each time. 3 Hits in 10 days gets people thinking about pizza.
Other ways to increase the redemption, are to attach coupons. People like coupons. They like to see something they can pull off and use to get a perceived deal. Do coupons create a never ending spiral? Yes, and it is tough to tame the coupon monster when you let it out, but if you don’t, your competition is going to anyway. No sense in looking a gift horse in the mouth.
The bottom line is, figure out what you want. The only ways to make more money are:
- Increase the # of customers you have
- Increase the frequency they buy from you
- Increase the average order.
#1 is front end sales, #2 is back end sales, and #3 is upselling. MacDonald’s aggressively upsells for a reason, they are in business to sell you super-sized cokes and fries, not to be concerned if asking you if you would like that super-sized is going to tick someone off.