Trying out new advertising opportunities is always a good idea – you never know what works until you give it a shot. The question is, how long does it take to determine if advertising works or if it’s a waste of money?
To answer this question, you’ll need to do two things.
First, you need to ask yourself what you expect the ad to accomplish. Second, you need a system to measure the performance of your ad relative to your goals. With your measurements and some simple math, you can determine if your ad needs refinement, or if the medium doesn’t work.
What Do You Want Your Ads to Accomplish?
Typical goals for advertising include:
- Increased brand awareness and recall
- Increased product awareness and recall
- Website visitors
- Information requests (coupon offer requests, dealer locator requests, rebate form downloads)
- Phone calls
- Online sales
- Store visits and offline sales
Obviously, every advertisement could do all of the above. But most ads are designed to do one or two things.
For example, this ad from Apple piqued the curiosity of TV viewers and increased brand awareness and recall for a product and company that was new to many in 1997.
The goal of a branding type ad like the one above is to get the consumer thinking about the brand. The effectiveness of this type of ad is typically determined by measuring recall – asking people if they can remember seeing the ad or not.
Ads that are focused on a result, on the other hand, can be evaluated by how well they accomplish the stated goal.
If your ad prominently features a special price on a popular product, and shows a phone number to call, it will be effective if the phone rings with people asking about the popular product.
Obviously, there’s a lot more to advertising than this quick explanation. But the big takeaway should be that you can’t evaluate an ad without defining goals and expectations.
Measuring Activity Associated With An Advertisement
As challenging as it can be to design and produce a great ad, it’s often more challenging to measure the ad’s results. In fact, attributing advertising with scientific certainty is very difficult.
Still, there are some things that can be done to try and approximate an ad’s effectiveness:
- Put a unique tracking phone number in the ad. HostedNumbers offers dirt cheap toll-free phone numbers you can put in each of your advertisements. Some people will call the number they see in the ad, and some won’t. But the tracking number will help give you some idea of what the ad is doing (or not doing).
- Use a custom web address. It takes a little bit of work, but your development team can create either a custom domain name that redirects to your website (with tracking info attached), or a ‘fake’ URL on your website that redirects and attaches tracking.
- Ask people to use a code when they buy. If the ad is focused on increasing transactions, a special discount code that’s specific to the ad is a great tracking tool. This works best for ads that people see or here (TV, YouTube, or Radio), but can be used in any ad.
- Put a unique offer in your ad. One of the oldest – and simplest – tricks is to put a unique offer in the ad that isn’t advertised anywhere else. The offer should be substantially different from your normal day-to-day specials…that way, you know the ad is responsible.
For more info about tracking URLs, check out this article about tracking part sales to back to online videos. There’s a section about tracking URLs that should be pretty helpful.
Measuring advertising results requires both planing and diligence, but the payoff is huge. Many of our clients have used these processes to eliminate ads that didn’t work, saving thousands of dollars in advertising costs every month.
How Long Does It Take To Test An Ad or Ad Medium?
Once you know what your ads are supposed to accomplish – and you have a system for measuring their effectiveness – you can use some relatively simple math to determine if your ads are working and/or if the ad medium is worthwhile.
1. If you’re trying out a new ad medium, start by testing an offer that you know to be effective. If you know that a “10% Off Everything!” offer always works well, that’s probably your test offer.
2. If you’re trying out a new medium, any activity at all is a good sign. Every ad medium has its’ own quirks. Ad creative that works great on Facebook can bomb in a Google display campaign, and vice versa. If you’re testing a new medium, all you’re looking for is some proof of activity (at least to start).
3.Test a few different ad ideas. A good ad test will have 2 or 3 different concepts, and 3 to 5 different layouts. It’s best to prepare all of these ads before your campaign begins, so you can test them all in succession, refining as you go.
4. Each ad needs between 1,000 and 5,000 impressions to have a chance. In a perfect world, you’d be able to give each ad 5,000 looks before you assessed it. However, that might be cost prohibitive. If so, you can trim that back to 1,000 impressions per ad.
Now, it’s time for some simple math. If you’ve followed the process outlined above, you’ll have between 3,000 and 25,000 total impressions on the ad medium. At this point, you’re ready to do simple math.
Take the total number of positive results you’ve gotten from the best performing ad, sorted by type. For example, here’s the performance from one ad that had 1,500 impressions:
- 30 website visits
- 7 phone calls
- 3 sales
Take each of the metrics you’ve got and double them (generally speaking, doubling the performance of any particular ad is an attainable goal with time and effort). Can you live with the numbers after they’ve been doubled?
If the answer is ‘yes’, congrats! You’ve found a good advertising medium. If the answer is ‘no’, then it’s probably time to pull the plug.
Please note: The answer to the question “How long?” is “However long it takes to run your ad test.” If you’re testing magazine ads, it could take 3-6 months to test a simple set of ads. If you’re testing banner ads on a website, you could have the test done in 30 days.
What About Advertisers That Don’t Facilitate Ad Testing?
The ad testing process we’ve outlined requires quite a bit of effort, both on your part and the part of your advertiser. While most advertisers are willing to work with you on this sort of process, some will push back and say that you can’t change creative more than once a month, that you need to give your ads more time, etc.
While there is some room for argument about how long an ad should be ‘given a chance,’ there is no room for disagreement on the following statement: It’s your ad budget, and you can spend it wherever you want to. If an advertiser doesn’t want to work with you to run a test, it’s a simple matter to find another advertiser.
Ad Testing and Opportunity Cost
Another thing to keep in mind before you test any ad medium – what’s the opportunity cost of testing this advertising option versus another?
Let’s say you want to test a full page ad in a monthly magazine. Each ad will cost somewhere between $2,500 and $5,000, and the test will last 3 to 6 months. That’s a total test budget of $7,500 to $15,000. What else could you test for $7,500?
- Half a dozen online pay-per-click options could be tested for a total of $7,500 – Google AdWords, Bing AdCenter, Pinterest Ads, Amazon, AdBrite, etc.
- A YouTube video ad campaign can be easily tested for the $15,000 budget, and that includes $5000 for ad production. If you reduce the price of production, the test could be done for $7,500.
- A cable TV ad campaign that includes both 30 second ads running for several weeks plus a segment in a show like All Girls Garage (there are other shows too – this is just an example)
- A campaign on a streaming radio service like Spotify or Pandora can be tested for $15,000 or so, only that might limit you to a handful of markets
The not so subtle point here: print isn’t a great ad medium. It’s expensive and hard to test. Caveat emptor.
Testing ads is the nuts and bolts of marketing, and we wouldn’t be doing our job if we didn’t mention that ad management is one of our services. We frequently run ad tests, which includes coming up with different ad concepts and layouts, setting up tracking, and measuring.
However you test your ads, diligence is key. Good luck!