|
We need to be able to intelligently discuss the differences between the following:
• Visitor
• Unique visitor
• Return visitor
• Qualified visitor
• Stale visitor
• User
• Customer
Visitor
Somebody comes to your Web site. Pages have been served. Links have been clicked. You
have a visitor. Buy why is that person there? No way to tell. At least you can calculate how
costly it was to get the visitor there now that search engines like GoTo (now Overture.com)
are selling placement:
Ameriquest Mortgage (www.ameriquestmortgage.com) knows exactly how much it costs to
get a pair of eyeballs to its site from GoTo.com. It willingly paid $2.72 for
each visitor. It now has a baseline. What would it cost to run a direct mail campaign to drive
traffic to its site? If Ameriquest Mortgage sends 100,000 postcards, each costing 10¢ for
paper and printing and 20¢ for postage, it would have to get an 11 percent response rate to
get the same value.
On average, if you take the total cost of advertising in the home mortgage industry and divide
it by the number of loan applications the mortgage companies receive for their trouble, it
costs about $250 per application. Or, the companies could cut a deal with Sherri Neasham at
FinanCenter.com for an order of magnitude less than that. You want a loan application sent
to you electronically? That'll be $25, please.
The cost of acquisition has nothing to do with branding. We are not measuring whether
people feel better about your company because of your animated banner ads and interstitial
interruptions or if they feel worse. Cost of acquisition has nothing to do with sales .... We are
measuring only the number of people who show up at your Web site. It is the cost of acquiring
somebody's attention, the acquisition of their eyeballs.
You could stand on a street corner with fliers that say, "Come to our store and we'll hand you
$5." What would happen? People would walk in, they'd get their $5, and some would spend it
there. The next day, you hand out fliers that offer $2. The next day, $4. In time, you get a clear
idea of what it costs to get people to show up. If you're good at tracking, you get an idea of
how much it costs to get them to show up, look around, buy something-and come back and
spend more another time.
What matters then is not how many people show up, but how many of the right people show
up. Are they qualified leads? Are they serious potential buyers? Are they actual buyers? The
surefire way to Web success is to track from that very first click all the way through to the
sale and beyond. All in good time. Let's not get ahead of ourselves.
At the moment, you know that you've spent $X and got Y people to show up. At least you think
you know. The site you buy banner ad space from says you got 2,082 clickthroughs yesterday.
Your referrer logs show that you got only 1,765. What's wrong with this picture? Nothing -
it's quite possible that both are correct. Your banner was clicked 2,082 times, but only 1,264
made it all the way to your landing page; 317 people clicked by mistake or waited too long
for your page to load and hit the Back button. Happens all the time. From Sterne, World Wide
Web Marketing, Third Edition, John Wiley & Sons, Inc., 1995
So you got somebody to the landing page-now what? Now it's time to determine if it's
somebody you've seen before.
Unique Visitor
What makes this visitor different from that one? Your server logs show that two visitors came
from two IP addresses. But if they come to you from the same AOL gateway, or the same
corporate firewall opening, then they look identical to the server.
When two visitors reach your home page from the same gateway, your log file tells you that
the server sent the same page to the same IP address twice. Is this one visitor who hit the
Refresh button? Maybe.
If they both click on the same link, perhaps the Products button, you still can't be sure. Then
one clicks for Product A, and the other for Product B. Got 'em! Right? Not so fast. So far, we
could have one person clicking the Refresh button on the home page, then the Product button,
then the Product A button, and then the Back button before looking at Product B. Remember,
your server doesn't tell you about the Back button unless you drop in a little Java code to
watch for it.
In the end, without cookies, you'll never know if that's one person or two. You may try to
design your site such that a visitor can only follow one path at a time, revealing a second
visitor from the same gateway, but that won't account for a visitor right-clicking on a link and
opening a new browser. Thus, one visitor can look like two, or four, or ten. The solution?
Cookies.
Return Visitor
Placing a cookie on a visitor's computer is the best way so far of telling one visitor from
another and knowing if the visitor has been to your site before. It identifies the visitor as
coming from a unique computer.
Drawbacks? Two big ones. The first is that a small portion of the population is
afraid/annoyed/put off by cookies and disables them. There's nothing you can do about that.
The second is that a growing number of corporate network managers don't allow cookies to
pass through the corporate firewall. Still, they're the best tool we've got at the moment.
Qualified Visitor
A suspect is somebody who shares characteristics with your current customers. A prospect is
somebody who has expressed interest in your products-perhaps by responding to a promotion.
A qualified prospect is one who has the need, the desire, and the means to make the buy.
Yes, some people buy things they don't need, so think of this in terms of psychology and
emotion. "I really need that purple pair of pants!" But just because somebody needs a root
canal, that doesn't mean they want it. Just because they need it and want it doesn't mean they
can afford it. So all three components are really necessary for a visitor to be classified as a
qualified prospect. Spending a little more time on qualified visitors than on unqualified
visitors is worth it because the qualified ones have expressed some deeper interest than the
rest.
WebTrends defines qualified visits this way: "Visits by customers who are considered
qualified as revenue generators. To qualify, a visitor must access specific pages on your Web
site that are specified by the system administrator."
A qualified visitor sees a certain number of pages or downloads the white paper or plays the
game. That identifies that visitor as a potential customer.
Stale Visitor
Frequency is nice, but recency means everything online. Qualified visitors eventually lose
their qualifications when they don't come back for a spell. How long? That depends on what
you're selling. A car buyer is different from an office supplies purchasing agent, who is
different from a refrigerator shopper, who is different from a chocolate consumer.
Just make sure that you have a designation in your database for those who previously
expressed interest and haven't done so of late. The astute marketing maven will send out an
inducement with a note that says "We miss you."
User
A visitor is a visitor. They come, they look, they may even become qualified if they stay long
enough and dig deep enough. But a user comes to your site repeatedly and for a reason.
Be very careful when buying sponsorship space on sites and talking about how many users
they have. One of the most amusing examples was Sixdegrees.com, which was built to help
people find friends of friends. In the long run, they discovered that friends of friends of
friends really didn't want to be found and the site folded up shop. But before they did, they
were caught in a controversy of how many users they really had.
One press release claimed 3 million users. Their publicist stood by that number in an
interview with Silicon Alley. But the marketing director, in a subsequent interview, said they
only had about 750,000 active users. In this case, "active" meant that they were still receiving
email from Sixdegrees.
Less than a month later, an email from Sixdegrees put the number at 462,000, while PC Data,
which monitors Web traffic, estimated 340,000 unique visitors. That was only 11 percent of
the Sixdegree's total registered users. Note the conflict between active and registered users.
Meanwhile, Media Metrix estimated 193,800 visitors and NetRating estimated 133,000.
Silicon Alley reported that "the 3 million users cited did not indicate the number of daily,
weekly, or even monthly users, as the release intimated. Instead, the number represents the
aggregate of site browsers who, whether they ever returned or not, had once filled out the
brief registration form and surfed the site in the three years since its launch."
The need for clarity is clear: A user is an identifiable individual (via cookie or logon) who
meets certain criteria of frequency and sustained activity. Measuring whether that activity is
sustained will help you keep your eye on your churn rate.
Churn
With goods in inventory, we refer to turnover. With users on a Web site, we talk in terms of
churn. Visitors are too ghostly for this term to apply. They come, they go, and we don't know
if they're new visitors or the same, loyal people coming back again and again.
In keeping track of your unique, identifiable users, you can calculate your churn rate. Why?
So you'll know where to spend your time. Are you signing up more people than you're losing?
Should you be spending more time on retention than acquisition?
Churn measures how much of your customer base rolls over during a given period of time.
Divide the number of users who fail to return in a given time period by the total number of
users at the end of the time period and you've got your baseline.
Christopher Knight provides a concise example of churn in his "Churning Out Marketing
Ideas to Keep Customers" article that appeared in the September 1999 issue of Boardwatch
magazine aimed at Internet service providers who wish to measure customer churn:
Say you have 2,000 subscribers on the first of the month. During the month you add 200 new
subscribers. You also lose 50 subscribers. At the end of the month, you have 2,150
subscribers. Your churn rate is 50 divided by 2,150, which equals 2.3 percent, which is your
churn ratio. Your growth rate for the month is 200 divided by 2,000, which equals a 10
percent growth rate. Annualized, this means (assuming you continue averaging the same
performance each month) your ISP has a 27.6 percent churn rate and a 120 percent annual
growth rate.
Knight was looking at customers. Fortunately, customers are much easier to count than all the
rest.
Customer
A customer is a visitor or a user who buys something.
The wonderful thing about customers is that you know so much about them. Name, address,
and even credit card information are willingly offered up to your database. Simple.
Churn rates apply to customers as well, of course. How many people make the bounce from
browser to buyer compared to the number of customers who don't come back? (Yes, I'm using
browser in place of visitor or user. The difference? None at all. I just like the alliteration.)
"But it's important to bring it up here, while we're talking about measuring the value of
advertising your Web site and the value of those who are attracted to your site. |
Disclaimer
