Do You Experience Banner Burnout?
By Mark Idzik
Do your banner ads suffer burnout?
How long does it take for your banners to start dropping
in effectiveness? Are you using more than one banner?
Is it animated? With click-thru rates running less than
1% industry wide, how do you make banners work?
Analyzing banner ad effectiveness is quite interesting,
and can be very revealing.
We're in the process of running a campaign for a client
across a larger banner network using several rotating
banners and the results are very interesting.
This client is in the computer business, so targeting
computer or internet oriented sites was our first
thought, although we decided on doing a run-of-network
(RON - across all site categories) to test it for the
first time.
Surprisingly enough, the category that received a whopping
4.5% click-through was "Family, Children, Pets & Education"
sites. Computer and Internet related sites ran approximately
2.5%, still good, but not nearly the response from the
top category.
The ad worked well for about 10 days, and then started
dipping. These 10 days represented about 300,000
impressions.
The banner did not have animation, was 3K in size, and
resembled a Windows 95 system message.
We've found that by alternating several banners and testing
their effectiveness, we could pull out and replace those
that weren't working, and get a very impressive overall
click-through rate.
Knowing the above, here's what we recommend to maximize
banner effectiveness and minimize the burnout:
- create a fast loading banner.
- if appropriate, use a Windows system message theme.
- test several banners on a quality banner network
like Flycast or Burst - run tests on LinkExchange
if you are on a tight budget.
- run the banners run-of-network (RON - across all
categories) to test and find out what categories
work for you.
- spend between $3-$6 CPM for your tests RON. Selecting
categories will increase the costs significantly.
- review your results weekly and replace or add new
banners when you see a dip in performance.
- test, test, test!
Finding out whether your campaign will work will depend
greatly on many other factors, including what product
or service you market, whether you are creating a
mailing list or selling directly from the banner, doing
a survey, or a variety of other strategies.
The most important thing we've learned is that you must
test, and continue testing. What works for someone else,
may or may not work for you. You can research other's
results (as we mentioned above) and use that as a basis
for your campaign.
Good luck and get rid of that burnout!
Peace.
Back to Library
