March 11, 2003
How To

Quick Tutorial: How to Track Email ROI Metrics - New Solutions & Tips

SUMMARY: How can you tell if your email program is truly working?
There is more involved than just counting immediate clicks. This
Quick Tutorial includes tips on:


- Landing page costs

- The bane of email order tracking: Multichannels

- Start measuring lifetime value

- How to find out if you are overmailing your list

- Always ask for \"favorite\" email account changes
At first glance, the ROI of email might seem pretty darn simple.
You compare your campaign costs to your sales and that's that.

According to direct response marketing math and finance expert
Robert Weinberg of RW Consulting, email is dangerous because it
seems so simple... when it's really not.

Normally marketers pay $1200 to attend Robert's seminar on the
topic, but we called him up and cribbed some notes for you:

1. Ignoring landing page costs
2. The bane of email order tracking: multichannels
3. Start measuring lifetime value
4. How to find out if you are overmailing your list
5. Always ask for "favorite" email account changes

-> Ignoring landing page costs

Many marketers just put the cost of the broadcast in their costs
line and forget to include costs related to collecting the order
such as your landing page. Your campaign ROI figures are not
accurate.

On the other hand, some marketers become hyper-aware of landing
page costs and try to "save money" by only investing in a single
landing page.

It is a false economy.

Aside from your list and your subject line, your landing page is
the most critical factor in the success of your campaign. As
such, you should be budgeting for at least one test-version of
your page whenever you do a major campaign.

The more you can learn about how landing page design factors
affect your conversion rates the better.

(B2B marketers are the worst offenders. Many either ignore a
landing page altogether and send traffic to their main home page,
or they do not bother tracking their landing page abandonment
rate. Even if you can not afford to test your page, does not mean
you should not track it to compare results by list and effort.)


-> The bane of email order tracking: Multichannels

Unless you are marketing purely through email, and no one could
have heard of your product or your site any other way (even via
search engines), you can never be absolutely sure that your
campaign response was 100% due to the campaign itself.

This matters because if you are budgeting and/or forecasting
sales based on email results, you have to take other channel
factors into account. A few examples:

Example A: What if someone, who knew of you through your other
marketing efforts and was about to order anyway, got your
email and used it to order?

Solution: Consider adding a question to your order form that
asks "How did you hear about us?"

Atomz's Marketer Seth Brenzel does this on his registration
forms and learned to his surprise that although email
"generated" the click, many respondents decided to click
because they'd heard about his company due to PR efforts in
the past. Result: He is continuing email, but he is also
increasing his PR budget.

Example B: What if someone got your email and ordered through
another channel, such as phone or catalog?

Solution: Consider adding a blatant 'Offer Code' to your
creative that all recipients must use to get the offer.
Anyone clicking on your link will see the offer code filled in
for them by default on the order screen. Anyone else will need
to hand-enter your offer code or mention it to your call
center rep.

Example C: What if someone got your email, but instead of
clicking on your link, typed in your URL by hand and did not
include the tracking data? (This may be a particular problem
with Hotmail addresses for several reasons.)

Solution: If the recipient is on your house list already, you
should be tracking all the email you are sending them as it is.
Therefore, no matter what channel they order in, you should be
able to assign some of the credit for that order to the email
campaign you sent them.

A second solution: Promote an easy-to-type URL in your email
creative. This means you can not solely rely on HTML hotlinks
that hide a URL completely (such as the words "click here"
underlined).

Also, remember that most people *stop* typing at the ".com"
even if there is more to your URL because they know the ".com"
will get them to a live site, so why type more? Therefore,
instead of putting your special landing page code after the
".com," put it at the very start of your URL. Example:

Wrong way = http://www.marketingsherpa.com/inspiration
Right way = http://inspiration.marketingsherpa.com


Example D: What if a email clicker does not buy on their first
visit to your site, but returns later independently (or due to
a different campaign) and ends up purchasing then?

Many shoppers need to see your site repeated times before
they trust it enough to purchase, the repeat visits are part
of the buying cycle and should be credited toward making the
sale.

Solution: Cookies.


-> Start measuring lifetime value

Too many email campaigns are only measured by the immediately
resulting sales. If your campaign is to an outside list, or to
your own in-house list of non-buyers, you should be tracking
"lifetime value" on that account.

Weinberg notes that "lifetime" sounds more formidable than it
really is. "It just means longer term value. I'll take three-
to-four years to track lifetime."

You will want to track lifetime by the original list the name came
from, the type of offer (sweeps often have lower values than
other respondents), and by all channels you market in.

Do you know if buyers acquired from email are more or less
valuable to you than buyers you get from other channels? You
should.

Also, Weinberg suggests you measure multichannel buyer value vs
email-only people. Many studies indicate multichannel buyers are
your most valuable.

BTW: Weinberg says he is heard some anecdotal evidence that
suggests the lifetime value of customers acquired through pop-ups
may not be as high as other channels. Very interesting.


-> How to find out if you are overmailing your list

Almost every emailer we have ever met is worried about this. How
much mail is too much? Will frequent mailings to the same name
over time crush response? Are you slowly killing your list?

Weinberg suggests this solution:

Split off at least one or two test cells from your list. They
need to be big enough that their resulting responses to your
offers are statistically reliable. They need to contain the same
mix of types of buyers and sources, so no other factor would skew
results.

Then as he says, "Group A gets minimal email, Group B more email,
and bombard the hell out of Group C."

Critical: Do not just do this for a week or two. You must track
response from the groups over a longer period of time to
determine true results. This length will vary depending on how
many mailings your least-mailed Group gets because you need
enough data to see a pattern.

If your least mailed Group gets monthly mailings, Weinberg says
you may need to run the test for up to a year. If frequency
is higher, you may be able to get viable results from three-six
months of mailings.

However, consumers' response to email is changing in general
these days as people get more and more inundated. You may
need to run this test every year or two to see if there have been
changes.

One fascinating note: Weinberg expects that you will find
different frequencies work better for different types of buyers.

"That top tier love you to tears. They can't get enough of you.
You may think you're mailing them too much, when you might be
better off sending them more!"

The answer to your question: Which frequency is best? It may be
several different ones. Welcome to the thrills of database
marketing.


-> Always ask for "favorite" email account changes

As you know, the average American has more than two email
accounts. Some are work vs. home, some are summer home vs.
winter home (lucky them), and some are "junk" vs. "stuff I want."

Plus, people's email accounts are changing constantly. More than
30% will switch accounts, and probably not think to tell you.

Weinberg advises that you ask the consumer if the email
address you have on record for them is the best one. Ask this
question every time someone orders with you (no matter what
channel) and every time they enter their account on your site.

You do not have to be obnoxious about it. A simple link that
reads, "Is this still your favorite email account? Click to
change it here" next to the spot on your Web site where they
enter their email is sufficient.

We can not think of any site that currently does this overtly,
and yet everyone worries terribly about the cleanliness of their
email lists!

The cleaner your list the better response you will get to
campaigns. The better chance you have of avoiding counting
a returning customer with a changed email address as an entirely
new customer. That would screw up your lifetime value math.

http://www.rwconsulting.net/

Improve Your Marketing

Join our thousands of weekly case study readers.

Enter your email below to receive MarketingSherpa news, updates, and promotions:

Note: Already a subscriber? Want to add a subscription?
Click Here to Manage Subscriptions