Recently, I received a few questions about measuring advertising response so I thought I’d cover that a bit today. The measurement and use of the results you record is one of the most important things to do when advertising – at least once the ad has been created for a particular target market.
Question: Why can’t all ads produce a response?
Very, very few ads pull nothing, but I have heard second hand of a business that mailed 20,000 direct mail pieces and gotten nothing for their trouble.
However, as I hear it, their mail piece was poorly done and was mailed to anyone with a heartbeat, so they sorta “deserved” that result.
In any audience, there is a percentage of people ready to buy (and thus, your timing is good), another percentage thinking about it, and the rest in various modes of not caring, not being interested, caring but not having a need or want at this time, etc. The key is motivating the 2nd and last groups to buy.
Question: How do you eliminate the process of testing ads and culling the non-performing ones?
The key isnâ??t to eliminate it, but to always test what you’re doing so that you can make decisions based on information rather than gut feel.
If we mail 1000 pieces, we might mail 333 people one letter, 333 people another letter and 334 people another one. Next time we mail, we’ll know which is the best producer. After that, we might mail 500 of the winner and 500 of a new challenger. You should always be trying to beat the current best performing ad you have in each media for a particular type of prospect.
If we place 20 radio spots, we’d alternate 2 or 3 spots in each time slot we select so we know which one works in that time slot (ie: different audiences, assuming they should all be “target rich” audiences). As each day goes on, we might adjust the spots that play in a slot based on the response we’re getting.
Question: Isn’t ad testing a very expensive process?
Depends on how you do it. If you try to contact everyone with a heartbeat instead of focusing on a personal, contextually important message for that prospect group, it can be very expensive, not to mention seriously unproductive.
For example, you wouldnâ??t likely send the same mail piece to opera lovers that you would NASCAR fans, for obvious reasons. Sure, there will be some exceptions (people who like both NASCAR and opera), but you aren’t worried about the crossovers. You’re worried about completely missing the boat with your message to one group or another.
Question: So how do I measure response on a mailer, newspaper/magazine ad, radio ad, email or website?
I could go on for pages about the details of this, but the bottom line is to do at least one of two things so you can tell exactly which ad they are responding to:
- Create an offer that is specific to the ad.
Ever notice how TV ads ask you to ask for a specific operator, department or send you to a website that has what seems like random numbers in it? That’s why they do this – so they know which ad you are responding to. They want to know which time slot works and which ad works, among other things. A particular price, quantity or product name can also indicate which offer you chose (and thus, which marketing effort you responded to).
- Create a mechanism for contacting you that is unique.
A different phone number (800 numbers are easy to use for this, because they forward to another number). A different fax number. A different web address. Google Analytics codes on your URLs (in emails, for example). A different email address. A special page on your website. A department number, or a contact name, ie: “Ask for Harry”. You can give them some sort of reference that gives them access to a discount or bonus, such as “Tell em Tiger Woods sent you”.