42KB
The Direct Mail Solution — Chapter-09: Tracking Your Mail Campaign
- Source: /library/The Direct Mail Solution/source-file.pdf
- PDF pages: 164–187
- Pages without text: none
143
Tracking Your
Mail Campaign
K, SO YOU JUST SPENT $3,000 ON A DIRECT-MAIL CAMPAIGN. THE
pieces are out the door, and you’re finished with the whole
process. Now it’s back to business, just waiting for all those
customers to flood into your store or call you on the phone. There’s
nothing more to do with the mailing, right?
Wrong.
Now comes one of the most critical parts of the entire campaign—
assessing the results so you know:
Was it worth it to spend the $3,000 in the first place?
Of the two versions of sales pieces you tested, which one did
better?
Which of the two mailing lists you sent to did better?
O
CHAPTER
9
144
THE DIRECT-MAIL SOLUTION
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
We’re talking about tracking. Tracking your mail campaigns isn’t
about the past. It’s about the future. It’s about understanding every
element of the mailing you just sent out so that you can refine everything
you do for future mailings. That’s the only way to systematically get
better results as you grow your business.
I can’t tell you how many new clients didn’t know the results
of their direct-mail campaigns before I started working with them.
They had a sense that orders picked up after a mailing, but they
didn’t have any hard data. And maybe they tried several different
pieces over the years, but they didn’t really know which one did
better.
A few years ago I was contacted by a well-known marketer that
targeted dentists. The company had tried direct mail a number of times
but said it was no longer working.
I asked for all the data relating to the five most recent direct-mail
campaigns. After I dug in, I found that it had a few segments that were
not only working but working extremely well. The company just had
to focus on those segments and drop the ones that didn’t respond.
It was thrilled when I showed them how it could increase its profits.
Unfortunately, it is not uncommon for companies to miss the critical
information that tracking could reveal.
Most people are lazy when it comes to tracking their marketing
campaigns. Frankly, I don’t get it. Why would you spend thousands of
dollars on something and then make very little effort to find out if it
was successful?
Sometimes a mailing works, sometimes it doesn’t. But if you don’t
know why, you can’t use the infor mation to create more effective
campaigns. You could waste thousands of dollars on future cam paigns,
using the same methods and lists that don’t work. And even more
frustrating, you could fail to use methods and lists that you’ve already
proven to yield the best results—but you haven’t done the simple math
to see it.
THE DIRECT-MAIL SOLUTION
145
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
A Scientific Approach
When you track direct mail, or any marketing campaign, you need to
become a master of logic. No emotions involved! We collect the data
and crunch the numbers, and up comes the answer we need to continue
building on our success.
Think of every piece of mail you send out as a seed with the
potential to grow you a new customer!
Now, if you were an agricultural researcher who was being paid to
grow more abundant crops of carrots, you would be very careful about
what you did with your seeds. You would keep very careful records of
how the seeds from one batch did compared to seeds from another
batch. You’d measure how your seeds did in different kinds of soil or
what kind of watering schedule yielded the best results.
You would do all that because it was your job, and your livelihood
depended on finding the very best way to grow carrots.
You need to be just as careful about following up on the mailings
you send out. You have to know what works and what doesn’t when it
comes to building your customer base. Accurately tracking the results
of your mail campaign plays a critical role in determining the success of
your future marketing efforts.
You can never be sloppy with your mailing results. You spend a
great deal of money to send out your sales material. You want to learn
as much as possible from each campaign so that you can maximize the
effectiveness of each mailing.
Careful tracking shows you exactly what works and what doesn’t. If
you do your tracking properly, and follow up on what you learn, your
business will blossom.
It All Starts with a Mailing Code
If you’ve been in business for a while, you already have customers
coming into your store, calling you, or visiting your website regularly.
146
THE DIRECT-MAIL SOLUTION
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
Now you’ve sent out a direct-mail campaign, and you want to know
whether it increased those numbers. You could figure out your average
number of calls and orders for each month for the past year, two years,
etc. and then base your answer on the increase from your average. I talk
to businesses all the time that track promotions this way. But, how
accurate would that be?
Sure, you could try to estimate your mailing results by looking at
the increase from your past sales vol ume, but there could be a number
of reasons why orders picked up, like changes in the economy. How do
you test the direct result of your mailing?
You need to make sure you have a system in place to track each
mail campaign. There are two things you can do to make tracking really
easy:
- Use a dedicated phone line and/or URL/landing page. That way you know that 100 percent of the leads/orders came in from your direct-mail campaign.
- Put a tracking code or coupon on the mailing piece. The next step is to collect that information from every customer or order that comes in. Whether it’s printed on the coupon or the order form, or an operator requests it when people phone in, get that code every time. Track the Number of Leads/Orders Having codes in place is the first step. Then you have to be consistent in collecting them for every lead or sale that comes in response to your campaign. When the campaign is finished, you will, of course, want to know how many leads or orders you received—but there’s a great deal of other specific information that will be very valuable. Not only do you need to know the overall response from the mailing but you also need to know how many leads or sales were in response to each mailing list, each segment, and each sales piece, if you are using more than one.
THE DIRECT-MAIL SOLUTION
147
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
If you are testing the effectiveness of a new sales piece vs. the piece
you’re currently using—your control piece—you need to know which
one performed better and which lists it performed better on.
For example, I have a client who is in the financial/business
opportunity niche. The lists I select for this client are either buyers of
financial newsletters and courses or individuals who have purchased
something relating to starting their own home-based business. For
one of our mailings, I tested two sales pieces for the client. I mailed
a 16-page letter and a 32-page magalog. The letter was plain, with
emphasis on the copy. The magalog looked more like a magazine, with
pictures and sidebars.
The letter outperformed the magalog on the financial lists, but
the magalog outperformed the letter to the business opportunity lists.
Accurate tracking showed me what type of sales piece worked better in
each niche. The results allowed me to be more targeted for each list type
so I could get the best response.
From then on, I mailed the 16-page letter to just the financial
lists and the 32-page magalog to just the business opportunity lists.
Using the two formats in their respective niches has proved to be a very
successful approach for this client.
This is a great example of how tracking the performance of each
sales piece for each list can help you maximize your profit potential.
List Segmentation
In order to properly track each version of the sales piece and each list,
it’s critical that you do a good job of list segmentation, which is the
process of systematically splitting up lists into smaller units to keep
track of your different parameters. That’s just a fancy way of saying you
have to use basic logic to keep all your variables straight.
For example, let’s say you have five lists with 10,000 names each,
and you want to mail two sales pieces, a 16-page letter and a 32-page
magalog. Here’s one way you can effectively segment your lists:
148
THE DIRECT-MAIL SOLUTION
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
16-Page Letter List Segmentation
List
Number of Names
Code Assigned
List #1
5,000
M1000A
List #2
5,000
M1001A
List #3
5,000
M1002A
List #4
5,000
M1003A
List #5
5,000
M1004A
32-Page Magalog List Segmentation
List
Number of Names
Code Assigned
List #1
5,000
M1000B
List #2
5,000
M1001B
List #3
5,000
M1002B
List #4
5,000
M1003B
List #5
5,000
M1004B
In this example, you have two formats that you use on five different
lists. That gives you 10 possibilities. You assign a code to each of them.
In the Code Assigned column, the Letter “M” represents your offer
(product/service). The number (1000 to 1004) represents the list. And,
the last letter (A or B) represents the version of the sales piece (or price,
etc.) you are testing. That code is printed on each of the pieces you send
out—either as part of the address label or directly on the piece itself.
When orders start coming in, your next step is to determine which
of the 10 groups the buyer was in. If buyers return a coupon or order
form, you would have the buyer write the code on the form; if the order
form is the overleaf of the address panel with the printed code, it will
automatically be included. If customers call in their order, the operator
just asks for the code.
You would keep track of how many orders came in from each
group. After a certain number of days or weeks have passed, you can
easily see which of the 10 possibilities brought the best results.
THE DIRECT-MAIL SOLUTION
149
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
Was the Mailing Worth It?
When tracking the profitability of a direct-mail campaign, you need to
make sure that you track all the costs associated with the mailing. That
means tracking the cost for each of the following: printing, list rental,
data processing, mail processing, and postage. I consider all these costs
the “hard costs” associated with the campaign.
If you are directly selling a product, you’ll also want to make sure
you take out the product cost for each order fulfilled. For example, I
did a mailing for a large pet supply store. It wanted to promote a free
bag of dog food to attract new customers. The dog food cost $15. The
calculations in the mail-tracking report accounted for each order. Even
though the new customer received a free bag of dog food, most went on
to spend another $35 in the store. I counted $20 in the tracking report
for every order received to help determine the actual return.
Taking out the costs associated with the mailing and the product
fulfillment cost will help show you the true profit/loss for each mailing.
You need these accurate numbers so you can determine your cost per
acquisition (CPA) or your cost per inquiry (CPI).
List History
Keeping track of the list history is important for knowing which lists to
order in the future. By keeping a list history, you track the performance
of one specific list over the course of all the times you’ve mailed to it.
Let’s say you own a car wash and you want to increase the number
of clients each month. You decide to offer a 20 percent discount to all
new customers. You rent a list of the most recent local “new movers”
(people who have recently moved into your area). You find that the
campaign is a huge success. The list is updated monthly so you decide
to rent the most recent names each month and send them your 20
percent off offer.
When you do this, you’ll want to keep track of the “new movers”
list history. How did it do when you mailed it in January? How about
150
THE DIRECT-MAIL SOLUTION
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
February? March? April? May? June? And so on. Watch the response
rate. Do you see a lift in response when you mail it in one month vs.
another? Do you see a decline in response the more frequently you mail
it? Does it give your business a boost in your slow season? Is it worth
mailing each month?
Keeping track of list history helps you identify which lists perform
consistently well time and time again, which lists are on a downward
trend, and whether there are any lists that work better in one season
than others.
A few years ago I was mailing for a financial publisher who
was selling a course on how to trade commodities. The direct-mail
campaigns were doing very well, and I was looking for ways to expand
into new niches.
I decided to test a list of farmers, thinking, “They grow corn,
soybeans, wheat, etc., so they should be interested in commodities.” It
sounded like a reasonable test.
So I researched all the farming lists on the market, and I came up
with what I thought was the best list. I mailed it and sure enough, the
results were outstanding. I continued to mail farming lists for about
three more mailings. Each mailing I took one or two additional farming
lists until I was mailing about five or six.
They were all working extremely well when—out of the blue—all the
farming lists stopped working. I couldn’t believe it. Why?
I decided to keep testing in hopes of figuring it out. I started
retesting the lists one at a time every few months. Then, once again, they
started working again.
By tracking the list history, I was able to figure out exactly why the
farming lists stopped working and then started working again. Can you
guess what the answer to the mystery was?
It turned out the lists stopped producing orders during the harvest
season. How many days a week does a farmer work during the harvest
season? Eight days a week! They are always busy! They don’t have time
to read mail.
THE DIRECT-MAIL SOLUTION
151
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
The list history showed me that the farming lists gave a great
response during the off-season but performed terribly during the
harvest season. I would have never known this had I not tracked the
history of the list. List history reveals details that help you know the
best times to mail to specific lists and when to avoid mailing.
Dig Even Deeper into the Lists You’re Buying
There’s a great deal more you can find out by studying how your list
performs. I always compare the list to how it per formed last time, and
the time before that, and the time before that, as far back as I have data
for it. I monitor my complete history of that list.
Looking at the complete history of the list allows me to see how the
list is trending.
Is the list still performing well, and are the results consistent?
Is the list trending up, and are the results getting better each
time?
Is the list trending down, and are the results getting worse?
The only way you can see this is by examining the history of the list.
So, in order to see a trend, you have to mail the same list over and over.
By tracking this history, you can see how a mailing list, and the names
included on it, can and will change.
For example, it’s not uncommon for me to mail a list with first
quarter update, and then a few months later take the same list with a
second quarter update. (The quarter represents the time frame when
the list owner acquired the prospects on the list. Most lists are updated
four times each year.)
The first quarter update may get a marginal response rate, while
the second quarter may get an outstanding response rate. If you
mailed the same piece to both lists, it may leave some scratching
their heads and asking, why did the response jump up like that?
Not that an increase in response rate is bad . . . but the same could
152
THE DIRECT-MAIL SOLUTION
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
happen in the other direction, too, and you’d have a significant drop
in response.
So why does this happen? Most marketers change the way they offer
their products over time. They change price points, sales copy, format,
order options, and the way they acquire customers. This changes who
lands on their mailing list that they then rent to you.
In one set of list orders, you may get buyers who paid $97 for a
specific product or service. In the next set of list orders, you may get
a group of buy ers who paid only $9.95 for the same product. Which
group of buyers will respond better to your offer? Most likely the high-
priced buyer.
The constant change in offers by the list owners directly affects your
list rental because you may be mailing to a different kind of buyer with
each mailing, even though you are taking the same list.
If you are mailing that same list or group of lists over and over,
ask your list broker to send an inquiry to the list owner and ask if the
offer, sales package, price point, or anything else has changed since you
last rented the list. If there are no changes, then you shouldn’t see a
significant change in results. If the list has changed, then you’ll want
to retest the list again rather than assume your results will turn out the
same. Dig in and research your lists, just as you keep tabs on every other
aspect of your mailings.
Lifetime Value
I know I keep harping on lifetime value, but once again I want to look
at it, this time from a slightly different angle. It is essential to calculate
the lifetime value of the customer in each mailing. This provides a
quick way to determine how profitable your mailing will be. Let’s say
your offer is for an auto repair shop. You offer a discounted oil change
for $15.95. You send out a small mailing to 500 prospects. The mailing
costs you $250, and you gain 10 new customers. Initially, the numbers
do not look good:
THE DIRECT-MAIL SOLUTION
153
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
[FOR] 10 Customers × $15 = $150 gross
With a cost of $250 to send the mailing out, it looks like you’re
losing $100. Was this mailing worth it?
Now look at the numbers after taking into account the lifetime
value of the average customer . . .
Let’s say you’ve determined that average customers who come in for
their first oil change will come back for other auto repair services and
spend at least $295 with you sometime in the next 12 months. Knowing the
lifetime value will help show you the true profit/loss for your mailing. In
this case, your 10 new customers are actually worth $2,950 (10 customers ×
$295 average LTV) instead of $150. This means your mailing wouldn’t lose
$100, but instead make you $2,750 over the next 12 months.
If you look only at the return on the initial purchase, you will
conclude that your campaign costs outweigh the number of customers
who responded. But, if you are willing to look at the campaign with the
big picture in mind, lifetime value included, then you may come to a
very different conclusion.
Keep in mind, you must track the lifetime value of your customers
for all your direct-mail campaigns.
Individual Mailing Totals
Similar to list history, it’s important to track individual mailing totals.
Don’t just look at the final mailing total one time, and then never
refer back to it again. Add the results of all the individual mailings to a
special spreadsheet, and track the history of all the mailing totals.
Tracking the mailing totals from each mailing in one easy-to-read
spreadsheet has helped me find specific dates throughout the year that
are very good mail dates as well as dates that I know to avoid.
I encourage many of my clients to mail at the end of December so
that the mail arrives in the mailbox around the first of January. This
is the time of year when people are making goals and decisions for the
154
THE DIRECT-MAIL SOLUTION
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
new year. This is when people are considering buying a new car, signing
up for a health club, starting investing and planning for retirement,
purchasing more health insurance, whitening their teeth, and so on.
This is a very good time for most offers.
Also, during certain seasons I encourage my clients to avoid
mailing. For example, I always try to stay away from mailing in April.
Tax time is a tough time of year for most families, and many are not in
the buying mood.
Knowing your best mail dates will help you know when to increase
mail quantities. And, knowing which mail dates bring you the lowest
response rates will help you know when to decrease mail quantities or
not mail at all. Keep in mind that the more years of mail history you
have to look at, the more valid your results will be.
By comparing mailing totals you may also find that when you mail
a certain quantity of pieces, the overall response drops significantly
(possibly because you have to dig deeper into older, less responsive lists).
Looking at one mailing total in comparison to other mailing totals will
give you a quick reference to your overall mailing performance year after
year.
If you’d like to see a great example of a mail-tracking report, you can
go to www.TheDirectMailSolution.com and download a free example.
You Need a Great Database
All this powerful information about the results of your mailings lies
in your database/order management system. In order to properly track
your mail campaigns and calculate lifetime value numbers, you need to
have easy access to your customer database and be able to manipulate it
to find the numbers you need. It is critical to have an efficient database
that can help you keep track of your customers and allow you to know
what and when they are buying.
New customers must be tracked individually to see whether they
continue buying other products or if they just remain a one-time buyer.
THE DIRECT-MAIL SOLUTION
155
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
How Do You Determine the Best
Months to Mail?
There is only one way to know for sure: test. Let’s say you
are a landscape/yard maintenance business owner. You
should run a test mailing in every season to see which months
are most responsive and which months are not.
Start with testing a mailing in February, May, August, and
November. This is the middle month in each quarter and cov-
ers each of the four seasons.
Month
Quantity Mailed
February
5,000
May
5,000
August
5,000
November
5,000
In most cases, you will find that two of the mailings per-
form much better than the other two. The next step would
be to mail larger quantities in the months that had a higher
response rate and smaller quantities in the months that had a
lower response rate.
You’ll also want to expand your mailing by testing other
months. You’ll test smaller quantities in the months that fall
in the quarters that didn’t perform as well in your initial test,
and larger quantities in months falling in the quarters that
performed better.
If February and August were your best months, then the fol-
lowing year’s mailings may look like:
156
THE DIRECT-MAIL SOLUTION
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
Best Months to Mail, continued
Month
Quantity Mailed
February
7,500 (higher quantity)
March
5,000 (new test)
May
2,500 (lower quantity)
June
2,500 (new test)
August
7,500 (higher quantity)
September
5,000 (new test)
November
2,500 (lower quantity)
December
2,500 (new test)
Keep increasing the mailing quantity for the months that
your mailings do well in and decrease the quantities that your
mailings do poorly in. You’ll come to a point where there are
some months that you will not mail in at all and other months
where you aggressively mail.
Your results will tell you what months it is safe to mail in, but
you must be sure to compare results side by side. You’ll also
want to make sure your offer and sales piece are consistent
between each mailing or your testing results will be off. For
example, you can’t compare a postcard mailed in one month
with a letter mailed in a different month.
If you want to test different formats or offers, you can split-
test each individual mailing. Mail one piece in each of the
mailings that is the same (from one mailing to the next), and
mail a separate version that tests a different variable in each
mailing.
THE DIRECT-MAIL SOLUTION
157
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
If they do go on to buy other products, you must keep a close eye on
what products/services they are purchasing. Are they buying videos,
books, accessories, or a variety of items? Knowing every detail about
the buyer will allow you to know where to direct your efforts when it
comes to creating and offering new products or adding and perfecting
the old ones.
Not only does your database need to give you convenient access to
all the information about your customers, it also needs to allow you to
accurately track your back-end campaigns as well.
Accurate tracking gives a significant advantage when evaluating
campaign results by showing what works and what doesn’t. If accurate
tracking measures are not in place, you may be sending sales pieces to
your prospects for a product they are not likely to purchase. Or you may
have a product that is generating so much response that you should
continue offering that same product—and perhaps develop variations
of it for additional back-end campaigns.
Best Months to Mail, continued
I’ve spent a number of private consulting days reviewing and
analyzing testing variables and mailing seasons. The results
are invaluable. It helps trim the fat, focus marketing efforts,
and can change the way you do business. All making you
much more profitable!
You have to test to make sure you know when the best time
of the year is for you to mail. Even though some of your mail-
ings may fail, it’s worth sending them out so you can be more
effective in your future marketing efforts. A little loss now will
help you reap huge rewards in the future!
158
THE DIRECT-MAIL SOLUTION
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
I once worked on a back-end campaign schedule that offered
about 11 different products over a period of 17 weeks. There were a
few products that it paid (very well) to offer more than one time to
the buyer. I received a 6 percent response on the first mailing for an
expensive video series, and a 1.5 percent response on the second mailing
for the same product. If proper tracking had not been in place, I would
never have known that it was profitable to mail this sales piece more
than once.
Test the Waters Before You Plunge In!
Direct-mail campaigns can be very expensive. And often you don’t really
know what will work and what won’t before you actually run one. It’s a
very good idea to start off by testing a small mailing before investing a
significant amount of money in a larger campaign.
Start with the right attitude; keep in mind it is just a test. Your
goal is not to make millions of dollars off the first campaign. The goal
should be to see if your sales copy, format, lists, and offer can generate
an acceptable response rate that allows you to expand.
Each mail campaign has a number of parameters, and getting just
one of them wrong could ruin the entire campaign. You have to be willing
to do more than one test and be committed to testing over and over. I
can assure you from the hundreds of millions of dollars I’ve seen made in
direct mail, it’s worth the extra effort and testing to make it work!
Even though I’m not a baseball fan (it moves too slow for me . . . sorry
baseball fans, I like basketball, soccer, football), I think it’s a great
example of how direct-mail testing works. The odds of you stepping
up to the plate and hitting a grand slam on your first try are very
slim. Your chances of a home run are a little better, but not much.
Ultimately what you want is a base hit. Once you’ve gotten a few base
hits, then you can try for a home run, and eventually a grand slam.
Keep in mind that there are numerous ways you can improve your
response rates, but you must have a starting point. Start small, and
slowly increase your mailing size.
THE DIRECT-MAIL SOLUTION
159
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
The largest mail campaign I’ve ever done was almost 6 million
pieces. It took me years of testing to get to the point where I felt
confident enough in the sales piece and the program to send out that
many pieces in one mailing.
The size of your first campaign should be based on the size of
your niche and the area you are mailing. If you are a large publishing
company that is offering a new product in a niche you’ve been in for a
long time, you may want to test as many as 50,000 pieces. But if you are
a small regional retail shop mailing into your neighborhood, you may
want to test only 1,000 pieces.
You need to test your copy, format, lists, and offer, and know
that they are all working before you start expanding your mailing
universe and invest significant amounts of money into a larger
campaign.
Getting the Most Out of Your Database
Your database holds a wealth of information that
you need to develop your business. The key is being
able to get that information out. And it hasn’t always been
that easy to do. I remember when I first started working in
the direct-mail business over 15 years ago, we would have to
use long complicated SQL sequences to put lists together.
(SQL stands for Structured Query Language—a programming
language designed specifically to handle data.) The process
could take hours. But today, with the right database solution,
you can pull out the data you want with a few clicks of your
mouse.
Here are examples of two features that your database solu-
tion should offer.
160
THE DIRECT-MAIL SOLUTION
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
Most Out of Your Database, continued
Easy In–Easy Out
A great database will let you enter the information for your
prospects and customers in one place rather than having to
enter different parts of the system to enter different kinds
of names. In conjunction with that, it gives you a very flex-
ible “tagging” (coding) system for your contacts. This allows
you to pull extremely targeted lists out of your house file. So
entry of data is easy, while pulling data back out is precise.
For example, if you are looking for direct-mail responders
who purchased from your test sales piece between March
15 and March 22, you should be able to easily form this list
within seconds.
When you are able to pull a targeted list out of your data-
base, you can expect an increase in response. That means
you can be more aggressive in your direct-mail campaigns by
sending a larger sales package or including more steps in the
sequence.
Automated Follow-Up
A great database allows you to automate your follow-up
sequence. You can set up automated follow-up campaigns
that allow you to send a specific follow-up letter, email, voice
broadcast, or live phone call to your prospect or customer on
a specific schedule.
For example, you could automate an email for everyone who
signs up for your monthly dry cleaning service club. It could
say something like, “Thanks for signing up . . . as a bonus for
joining our membership, I’ve put a special package in the mail
for you . . . it should arrive any day now.”
THE DIRECT-MAIL SOLUTION
161
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
Getting Better All the Time
You should always be working to get better at de signing and running
your direct-mail campaigns.
Sure, one day you may send out a winning cam paign that has a huge
response. But that doesn’t guarantee that you’ll keep getting that same
Most Out of Your Database, continued
Then, you could have another follow-up email that would be
sent seven days after sign-up that says something like, “I’m
just checking to make sure you received your free bonus
gift.”
Follow-up sequences have proven to help boost response
rates for later campaigns. This boost in response helps
increase customer lifetime value.
Finding the Right Database
While the technology to create the perfect database is avail-
able, in my experience finding one really great database that
does it all can be difficult. Most databases either don’t have
all the features you need, they’re complicated to use and
require additional programming to set them up the way you
want, or they require the use of a second or third set of soft-
ware to make them do the tasks you require.
That’s why I was excited when, a few years ago, one of my
clients introduced me to Infusionsoft, a database company
that really understands the needs of direct-response com-
panies. Now I recommend it to all my clients who are in the
market for a complete database solution. For information
on this excellent software product, visit its website at www.
Infusionsoft.com.
162
THE DIRECT-MAIL SOLUTION
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
response if you keep doing the same thing. And it doesn’t mean you
can’t get an even bigger response by making some small change. If your
goal is to make more and more money, your attention should be on
creating better and better campaigns.
But you don’t want to make random changes. You have to go
about it scientifically so you know what your findings mean. Make
measurable changes, one at a time, and track whether there’s a differ-
ence in response rate. The numbers will tell you what works and what
doesn’t. Then use what you learn from each campaign to make the
next one even better.
What Kind of Changes Are We Talking About?
Really, it can be almost anything.
The sales copy. The offer (e.g., price, discount, free trial, bonus gift,
etc.). The format (letter in an envelope vs. a magalog). The color of the
ink. The way customers are asked to respond (order form vs. call-in only).
The possibilities are endless. And any of them can significantly
affect response rate. That means there’s lots of opportunity for refining
what you do. You probably have an idea of what may be impor tant.
Maybe you think lowering the price would make a difference, but you
don’t know for sure. (In fact, in some cases, I’ve seen a higher price point
bring in more sales. I know it seems counterintuitive, but the numbers
don’t lie.)
Select what you think may be the most important variables, and
design your tests around those.
Segmenting Your List
This is where you have to think logically.
First of all, you have to set up a comparison. So, if you have one
sales piece that you’re using as your regular piece (your control piece),
and you want to see if you do better or worse by changing the price of
your product, you want to set up a kind of “experiment.” That means
you send one group of prospects the control piece and another group
THE DIRECT-MAIL SOLUTION
163
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
of prospects—one that is a close match to the first group—the test
piece.
Second, you want to set up your test so that you can clearly see the
influence of each thing you’re testing. You don’t want to test two things
at the same time. If you change both the price and the format, and you
get a different response, you won’t know if it’s because of the different
price or the different format or both.
That doesn’t mean you can’t test for more than one variable with
one mailing. But to do that you have to properly “segment” your mailing
list. That means you have to systematically split up your big list into
smaller units that allow you to keep track of the different parameters.
In other words, you have to use basic logic to keep all your variables
straight, with none overlapping.
Suppose you have a control piece you’ve used for a while, and you
decide to test one different headline. You also decide to test one differ-
ent offer. Maybe the control piece offers a hand car wash and wax for
$179.95, and you want to see if $159.95 is a better price. So you’re
testing two prices, and you have to be able to look at their influence
independent of each other.
Earlier we looked at a simple example where we tested only one
vari able. Now let’s look at a more complicated example. In this case let’s
say you have five lists with 20,000 names each, and you want to mail
two different offers (the control piece offer and the new offer) with two
different headlines (the control piece headline with the new headline).
Here’s one way you can ef fectively segment your lists:
Offer 1: Headline 1 List Segmentation (Control Piece)
List
Number of Names
Code Assigned
List #1
5,000
M1000AC
List #2
5,000
M1001AC
List #3
5,000
M1002AC
List #4
5,000
M1003AC
List #5
5,000
M1004AC
164
THE DIRECT-MAIL SOLUTION
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
Offer 1: Headline 2 List Segmentation (Test Piece 1)
List
Number of Names
Code Assigned
List #1
5,000
M1000AD
List #2
5,000
M1001AD
List #3
5,000
M1002AD
List #4
5,000
M1003AD
List #5
5,000
M1004AD
Offer 2: Headline 1 List Segmentation (Test Piece 2)
List
Number of Names
Code Assigned
List #1
5,000
M1000BC
List #2
5,000
M1001BC
List #3
5,000
M1002BC
List #4
5,000
M1003BC
List #5
5,000
M1004BC
Offer 2: Headline 2 List Segmentation (Test Piece 3)
List
Number of Names
Code Assigned
List #1
5,000
M1000BD
List #2
5,000
M1001BD
List #3
5,000
M1002BD
List #4
5,000
M1003BD
List #5
5,000
M1004BD
In this example, you have two offers and two headlines that you use
on five different lists. That gives you 20 possibili ties. You assign a code
to each of those possibilities.
In the Code Assigned column, the letter “M” rep resents your
product. The number (1000 to 1004) represents the list. The next to last
letter (A or B) represents the offer you’re testing. And the last letter (C or
D) represents the headline you’re testing. That code is printed on each
of the pieces you send out—either as part of the address label, coupon,
or directly on the piece itself.
THE DIRECT-MAIL SOLUTION
165
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
If the control piece gets the best response, you know that Headline
1 and Offer 1 can’t be beat by the changes you made. You can forget
Headline 2 and Offer 2 and try something else for the next mailing.
If test piece 3 outperforms the control piece, you know it’s good,
but is it because of Headline 2, Offer 2, or both? We look to our other
two pieces to tell us. If Test Piece 1 outperforms the Control Piece we
know Headline 2 worked (both pieces had the same offer and only
differed in the headline). If Test Piece 2 outperforms the control piece
we know that Offer 2 worked (both pieces had the same headline and
only differed in the offer). But, if Test Piece 1 bombed, we know that
Headline 2 did not work, and the reason Test Piece 3 worked was
because of the Offer 2 and not because of the headline. If you had only
tested Test Piece 3 against the control piece, you’d never know that.
But because you did things logically, you now have a new, better
control piece that uses Headline 1 with Offer 2. Next mailing you can
test another variable!
One last note on this: If the control piece is beaten by one of the
test pieces, depending on the response, you may want to retest the sales
pieces. Mail the control against the test piece that won. That way you
can safely confirm the results before rolling out with the new piece.
To make all this work worthwhile, you have to put a system in place
to track every response that comes in and which of the 20 groups they
fall in. Do this for a cer tain number of days or weeks, until you feel
most of the orders from that mailing have come in. At that point you
can easily see which of the 20 possibilities brought the best results—or
if there was no dif ference between the groups.
This is the data you need to ensure that your next mailing does
better than the one you just sent out. It’s the payoff of the entire
operation—and the key to bigger and bigger profits!
I hope you can now see how important it is to properly track your
mail campaigns. It doesn’t work to send things out willy-nilly. Become a
scientific researcher of your mailings, and you will get better and better
at it every time you send out a new campaign.
166
THE DIRECT-MAIL SOLUTION
CHAPTER 9 / TRACKING YOUR MAIL CAMPAIGN
OK, now you’ve got a plan in place to track your sales pieces as they
traverse the nation or just go across town. You’re almost ready to drop
the mail.
But not quite.
There’s a little cleanup work you should do on your mail list to get
the best results for your money. That’s what we’ll look at next.
You Need the Numbers
There’s only one limit on the number of variables
you can test in a mailing, and that’s the number of names
you have to mail to. Statistics lose their value if they’re based
on too few numbers. You have to start out with enough
names in each cell of your logic table to make your results
statistically accurate. I feel comfortable with a minimum of
1,000 names per cell, but 5,000 would be better if you are
doing a regional mailing. If you are mailing nationally, I’d like
to see the minimum test cell quantity at 7,500; 15,000 would
be even better.
If your lists aren’t big enough, you should limit the number of
variables you’re testing so you have an adequate number of
names in each cell.