Q
Let’s get started by talking about what you consider your greatest circulation and database challenges.
Bonnie Preace, circulation manager, REI Our biggest channel is retail, our second biggest channel is Internet, and our third biggest channel is telephone, and I do not get source-coded sales, period. So I do a lot of testing to determine what was the incremental amount of sales driven because of the catalog going in homes. Not having the wonderful source code information I used to have is by far my biggest challenge.
Mark Mackaman, vice president of operations, Computer Gear Our biggest database challenge is continuing to grow and find names that work. We are niche oriented, so our challenge is trying to find out of the big pool of computer users those who are computer enthusiasts.
Barb Lund, merchandise and marketing director, Museum Tour Museum Tour is predominantly a fourth-quarter book. It’s tough to find good spring help, and we’ve tried several things to keep busy and keep those people employed in spring. We find the school market to be a nice piece of bread and butter for us, but targeting them is difficult at times because they come in under [purchase orders] and you don’t know what school they are coming from.
Karen Rueter, circulation analyst, Griot’s Garage Griot’s Garage is an upscale seller of car-care-oriented products. A big challenge, obviously, is rising handling costs and finding ways to keep response rates, average order size, and dollars per book level while trying some other ways to keep the cost per order down.
David Batchelder, IT manager, Raintree Nursery Our biggest issues at Raintree Nursery are like everyone else’s. We have a match-back problem with trying to find out which lists provide our best customers and to keep expenses down because catalog expenses are going through the roof.
Sam Benowitz, owner, Raintree Nursery What we’re trying to do is maximize our profit margin and make the best use of our facilities, and to reduce our catalog expenditures by getting a bigger percentage of orders. We are not doing a whole lot of prospecting. We do a little bit, and we’re also doing some PR [to attract business].
According to Multichannel Merchant’s most recent Benchmark Report on Marketing [February 2005 issue], house file response decreased 17% during the past two years, to less than 5%. Has anybody here experienced this, and if so, what are you doing to counteract it?
Lund Older segments of our house file of course decrease, but that would be in line with our product offering. We try to offset that by adding some unique products for the family [in addition to items for younger children]. It has given us a little more stickiness, but we try to keep the products in the same educational vein.
Rueter I’m looking at one year to the next and I’m seeing dollars per book being maintained, and then I start looking at my percent response, and that’s where I’m starting to see the degradation. Are we forcing the situation where we’re going after dollars per book and suddenly we’re seeing percent response start to shuffle away? And as the response comes back, there’s the long-term effect of the names that you gained from the [database] co-ops that converted into your house file. How have they diluted the house file? Are you starting to dial away from what your core customer wants?
We’re looking at this and trying to understand that this in fact is happening. So should we pull back on prospecting and such a heavy reliance on the cooperative databases and give it a rest?
Are you saying that the co-op databases have introduced into your house file one-time buyers and people who aren’t Griot’s core audience?
Rueter That may be part of it. They could have come in and made a large purchase that bumped up the dollars per book, and I don’t have enough of a long-term history to go back and see what was the lifetime value of that particular source. That’s the danger of watching just one particular metric. You have to watch the top line and the bottom line at the same time, which is kind of a tough way to go. As far as it looked for the past couple of years, dollars per book was right on the money, but then I started to look at that percentage response. It’s bigger, but is it better?
Mark, have you found that with computer enthusiasts?
Mackaman We’re still a relatively small catalog and still growing. Our [database] cell sizes are relatively small for us to take a definitive look at this. But I’d have to say we have not really seen a falloff at all. In fact, we’ve probably gone the opposite way with being more responsive.
We started out doing trade shows and a number of other things, and we sold a lot to the computer enthusiasts direct. By participating in the co-ops we’re getting a lot more of the women buying the gifts for the enthusiasts. Our house file used to be two-thirds male, one-third female, and now it’s 50-50, so we have seen the importance of the gift buyer buying for somebody else. Now we really have two customers — the gift buyer and the end-user enthusiast — so how do you target things?
Rueter: Have you been adapting the merchandising to match what is happening?
Mackaman We have made some changes, yes. As you probably would have guessed there are not that many female computer enthusiasts. There’s some, and we do get requests for female stuff [such as women’s T-shirts] and we tried that, but we can’t get that to work.
Lund Do you find the average order is higher on the female side or the male side?
Mackaman It’s hard to tell, because you really need to know why they are buying. We have not really looked at it from male/female. We’ve done some things to increase average order size, and it’s gone up $15-$20 over the past four or five years. That’s a huge jump for us.
How have you done that?
Mackaman Merchandise selection. And being a lot more creative. We put four or five items together, and that moves the average price point up.
Bonnie, I see you nodding your head.
Preace We haven’t seen a decline at all on the circulation side. In the past couple of years we’ve had some terrific successes. We’ve certainly heard it from the industry. But we’ve always been happy to talk to our list broker/manager and say we’re not seeing it. REI had a great couple of years.
Sam, have you seen a deterioration of response in your house file?
Benowitz We’re lucky to have customers who are going to be gardeners until they die. But we are seeing changes. Over the years, we used to have people buying big trees, and now a bigger percentage of our customers own smaller and smaller properties, so they don’t want to plant a big tree and have it take up their entire yard but rather they would plant 100 smaller plants.
You mentioned, Sam, that Raintree Nursery isn’t doing much prospecting. What about the rest of you?
Preace We are growing our prospecting. We have changed the way we are prospecting, and in a way it’s gotten a lot more complex. We’re prospecting based on whether our customer is in a retail trade area or out of a retail trade area right now. An out-of-trade-area customer isn’t going to perform as well [as one who lives near a store], but we still have to be able to acquire that customer. So we’re judging them by a different cost per new customer than the cost to acquire an in-trade customer.
Why do out-of-trade prospects and customers perform less well?
Preace They don’t have the brand awareness. So I’ll take a select that’s been working really well for me [for in-trade-area customers] and if I go out of trade I’ll have to tighten that select. Certainly they are responding, but at direct response rates vs. [the higher] retail rates. It’s been a big challenge but very rewarding to us to understand the difference between the two customers. When we were clustering them together we weren’t making the best decisions possible.
Lund I work with three co-op databases, and that’s been working really well. I don’t work them against each other. But I do control what I give them. I try to buy lists that are not saturated in the market. There are some new mailers that do not exchange, rent, or otherwise, so I do try to rent names of buyers just to keep it growing. In general, I think the co-ops are doing really well. The one thing I am doing next year is tightening up my segmentation of the co-ops, and I am ordering lists of our e-mail buyers with the same parameters.
So you’re buying the postal addresses of e-mail buyers?
Lund When I get a list from Abacus, say, I’ll take segment one, $50 order plus e-mail, presence of children, whatever, with buying habits online. Basically we are asking for four or five parameters within that list. We tighten it as much as possible. It costs more to do that. But the return on that investment is tenfold.
Preace I start out as tight as possible and loosen them as quickly as I can. So if my tight list is working phenomenally I’ll start taking those parameters off.
Benowitz We are a small company. We used to get a list of 12-month buyers without being more specific as to how much they purchased.
Lund Abacus will offer me a $50-plus [select], and if I want a higher dollar order [value], I’ll lose a lot of the rollout segment, but I only want to roll it out so far anyway. So I might [tighten the select] to $100-plus to keep it high end, but we’ll keep some $50-plus. That works really well.
Mackaman You also have to be a little careful about the economy… at least for us. When the economy turned down we had to tighten up the selects, and now it appears as we come back out that we can relax some of the selects and experiment with marketing again.
Beyond list rentals and the co-ops, has anyone tried any other prospecting methods, such as inserts or e-mail lists?
Preace At REI because of our three channels we are acquiring customers every which way, and that’s part of my challenge. I’m only responsible for the small percentage of new customers who are being acquired via the direct mail piece. We’ve also got the radio advertising campaigns going on, we’ve got free-standing inserts [FSIs] in newspapers — we’re actually testing FSIs that would normally go in a newspaper and putting those in the mail and seeing how they perform in the mail. We’re acquiring customers online, search, affiliates, all different ways, and the biggest challenge is saying which of those [new customers] are mine. And I honestly work hard to try to figure that out.
We only take credit for the incremental customers, so the way we do it is match-backs. We go ahead to Experian, our service bureau. They hold on to the acquisition mail files for us, and we will send them three weeks’ worth of orders and say, Were any of these people on that mail file? And if they were I’ll take credit for that.
I know a lot of the companies sell match-back tools, and they are really, really expensive. I don’t use a tool at all. I simply say, Will you match these orders to this mail file? It’s much less expensive.
We keep mentioning the co-op databases. How many of the co-ops do you use, and why? And are you seeing an oversaturation of names?
Preace I am superhappy with the co-op databases right now. I use three: Abacus, Z-24, and NextAction, which is the newest one for us. It’s a lot of work to maintain your co-ops, keeping everything going. But I kind of laugh and ask myself, Why am I not in all of them? They are really working for me. The price is right. Some of my lists that are coming in on rental are getting more expensive, especially the niche lists.
I don’t feel there’s saturation. I’ve got these three different co-op databases; their models are working differently, so they are pulling different names, and they are all continuing to work.
Why do you think that each of them is working differently for you?
Preace NextAction is looking at the item-level data, so that’s why they are pulling differently than Abacus and Z-24. Why Abacus and Z-24 are continuing to work side by side I don’t know why. They have different models, and that’s probably a co-op’s goal: They know what their competitors are doing, so they think, We’ve got to act differently. For me it’s been good.
Lund We use three co-ops right now, and I just stick with three. I think that their modeling is very different, but on the other hand I’ve tried item modeling, and that does not work for us. The price points are too low, and the conversion on it doesn’t work for me. They are a lot more cost-effective, co-ops. There’s a lot they manage for you.
How do you determine how much of your prospecting is going to be from co-ops vs. list rental and exchanges? Is there some magic formula?
Preace I’ve watched the co-ops continue to perform. I used to think that if one little thing happens to me in a co-op it could change my results quickly, and I haven’t seen that. I’ve had two years of fantastic results, and I’m like, Okay, loosen the strings a little bit. But there’s no formula. It’s more like my gut says, Okay, let a little more through.
Is the majority of your prospecting through the co-ops?
Preace No.
Mackaman We’re going full tilt. Whatever works. We tested several co-ops three years ago and couldn’t get any one of them to work above breakeven. But we’re going to retest this fall. For instance, NextAction hadn’t started [item-level modeling] at the time and was just like everyone else at the time.
Lund Question for you, Mark: How much is too much to test on your segment?
Mackaman We’ve been testing 5,000-10,000 names into a segment, and if it works then we roll out into a segment for the next year. We’ve been doing that all along. We’ve been testing Segment Twos and have been getting them to work on many of the models. Again, we’re still in this acquisition mode, so we’re still growing.
Lund Do you go as deep in Segment Three and Four?
Preace No. Because I’m so tight in my circulation with them I haven’t tested that deep. With the success I’ve had I probably could.
Lund I only go as far down as Segment Two. I’ll try a new model before I go past Two.
Benowitz We just go to One. We’ve done Two before, but we don’t need that many names. We have all the names we need in the areas that we want.
Preace You guys are in a great position.
Barb, I know you said you had issues with match-backs.
Lund The match-backs that Abacus did last year were great, and I have another co-op working on it this year. Anyway, what I wanted in a match-back was pretty simple. I want to know where my Web buyers came from, and I want to know who those unknowns are. That’s really all I want to know. [The service provider] can [perform match-backs] to such a degree that there’s more information than I ever want to know.
Can you give an example?
Lund They can do it on a household level. They are taking your address and saying, Okay, if you live on 123 Main St. and that drop was two weeks ago, we’re going to attribute you in that drop. Well, if I didn’t have key codes, I would need that. But I have key codes. So I know where you’re coming from. All I really want is simple: How do I apply this across the board so that I know where my Internet buyers are coming from?
Are they giving you that?
Lund Well, they will. They’ve done all this other stuff that I don’t need. The point is they can do it.
But once you know that a person came from the Web and wasn’t necessarily from the catalog can you then ascertain how he came to your Website?
Lund If it wasn’t a co-op name or a list rental name, then no. We optimize and everything else that we need to do, but our percentage of truly unique Web buyers is minimal. Very minimal.
Rueter We were mailing in quick succession for a while, and there’s a little bit of a challenge in that [regarding match-backs]. It’s a cause-and-effect kind of thing. You have to stick to your rules. During this period of time, you look at the mail tape. Then the next mail tape hits, but someone will purchase from the previous book. Ultimately there’s going to be a certain percent that are unknowns. But you settle in and use basic assumptions so that you can move forward.
Are you using a service provider to perform your match-backs?
Rueter Our service bureau is doing it.
Mackaman How many people are doing it after each mailing? We were doing it after the end of the season.
Preace We are doing it after every mailing. Our rationale is that it would be the same price for me either way because we are paying on a per-input cost. My problem is that we are in the mail so often. I have a catalog that’s in-home every four weeks, so I just have to take the four-week time period and apply it to this catalog.
Mackaman Do you mail to all the same people each mailing?
Preace Not all the same people, but I do lose that incremental person who’s out there [buying from a previous mailing].
Is anyone doing any unusual methods of prospecting?
Lund There’s a product coming into our catalog next year that happens to be from a manufacturer that runs a national ad campaign. So I exchange space [in my catalog] for ads in Smithsonian and Newsweek. That’s one thing.
The other thing that we’ve done is because we’re a co-op catalog, [merchants] have to buy space. Some of our vendors have their own catalog, so we exchange lists with them.
Is that working for you?
Lund Yes, because they are new names and are unique to us. They are not diluted or spoiled.
Preace Our biggest challenge is taking too many names on exchange and getting our balances out of whack. Certainly the Annual Catalog Conference is the place where we meet and talk about the balances. And more than ever I’m having to go into that show and be so prepared and say, Here’s what I need the rest of the year, and how many of these names will I be able to use, and when will I return to rental? Because that changes my plan dramatically. And the worst thing is that the end of the year is holiday, and if you’re not prepared, you get hit because a list turns to rental on you.
A critic of list exchanges would say that you’re missing out on list rental income by exchanging names. Is that an issue?
Preace The value of the exchange is that you can sit there and say, I would pay this much for it vs. I would get this much in income. It’s a wash at the end of the day. The value of the exchange is much more beneficial to me at the end of the day, and I can go deeper into the list because I didn’t have to pay for it.
Lund Mostly we deny the exchange if [the other mailer will] have a free offer or promotions. We do not have the luxury of free shipping or sales. We just don’t do it.
Rueter My names work well for someone else, but theirs don’t work for me.
Lund I was sitting on a balance, and I didn’t want to mail [because the list hadn’t been effective the first time]. I tested it again and got great response. My suggestion would be to test it again.
Has anyone tried e-mail lists?
Batchelder We’ve been exploring it on a list where they actually have to sign up. It’s hard to monitor. We try to use e-mail to borrow and test names through an e-newsletter and give them more information to keep them involved and move our merchandise quickly. We never used to buy lists.
Rueter We haven’t explored that in a couple of years because when they first came out they were really expensive. Has the price come down? Is it under $200/M?
Some, not all.
Rueter It’s a shotgun approach.
Let’s talk about reactivation campaigns.
Preace Our response has been great. But if you’re not ready to pay for them and treat them like prospects…you have to be ready to lose money on that name and look at the lifetime value.
How deep into your house file do you go?
Mackaman Until recently we’ve been mailing nine- or ten-year-old customers and still getting them to pay for themselves. We’re doing RFM, so even [inactive customers with a lifetime value of] $50 or above can still be profitable to mail to. This fall we’re going to run some Abacus optimization. We have a huge bunch of prospects that requested a catalog at some point, and we want to see whether or not they would still be interested.
Rueter We’re in the middle of a reactivation campaign right now. Those who order will get a free bottle of Speed Shine. What’s nice about this is it’s a proprietary product, we manufacture this, it’s a great product. So this works on two levels: to hook people into this product line and also to reactivate.
How far back do you go to reactivate, Karen?
Rueter There’s a drop-off after 24 months, just like everyone else has. We also have a seasonality factor. For instance, we use Father’s Day as a focal point for one of the books each year.
Benowitz Through Abacus, we’ve optimized our customers from about four years back or so. We had over 100,000 names, and they didn’t perform any better than the list we were already renting. And we had paid more to [optimize the names], so we decided it wasn’t worth it anymore to do it. We may try it again, though.
Regarding next year’s likely postal rate increase: Will it affect anyone’s circulation plans?
Rueter Has anyone tried prospecting with a reduced page count or merchandise selection?
Preace That’s what we do. Our prospects get a smaller book. We tested a few years ago.
Rueter We’re looking at testing that and studying what first-time buyers from different prospecting sources seem to be attracted to.
WHO’S WHO
Roundtable participants
Seattle What’s in a name? Plenty, if you’re a direct marketer. And what with the Internet becoming a vital source of new names — not to mention ever-increasing modeling options and the difficulty in determining the sources of online buyers — prospecting, reactivation, and overall circulation strategies are especially hot topics. So Multichannel Merchant headed west, to Seattle, to hear how those in the trenches are handling the challenges.
David Batchelder is IT manager and Sam Benowitz is the owner of Raintree Nursery, a Morton, WA-based horticultural merchant.
Barbara Lund is merchandise and marketing director of Museum Tour, a cooperative toy cataloger based in Milwaukie, OR.
Mark Mackaman is vice president of operations for Redmond, WA-based Computer Gear, which sells gadgets and gifts for computer enthusiasts.
Bonnie Preace is circulation manager of Recreational Equipment Inc. (REI), a cataloger/retailer of outdoor gear based in Kent, WA.
Karen Rueter is circulation analyst for Tacoma, WA-based Griot’s Garage, a merchant of automobile and garage accessories.
Multichannel Merchant editorial director Sherry Chiger and senior writer Mark Del Franco moderated.