Advertising a small business is too expensive to do haphazardly. If you’re going to invest money into either print, radio, TV, or online advertising you need see a return on your investment. Unfortunately, many business owners don’t know how to evaluate the success or failure of their ads. Ask them how well their advertising is working, and the best some owners can say is that they “think” their advertisements are working well, they don’t know which ads are working well, or they don’t know if any are working.
Don’t fall into this trap. To make the money you spend on ads pay off, you need to measure and evaluate your ad spend based on data.
Although there’s a lot of science involved in advertising, there’s also a lot of art. Many factors are hard to measure. For example, if somebody sees your ad, comes to your store and doesn’t buy anything, that builds brand awareness. They may come back later when they need what you sell. In that case, your advertising worked—just not immediately. There are plenty of situations where advertising is hard to measure but that doesn’t mean you should give up trying. The fact is, most advertising is measurable if you design your campaign correctly.
Before You Advertise…
To measure the effectiveness of your ad you need baseline data. What was your store or website traffic before you ran the ad? How were your sales? Did sales increase for the products or services you advertised? You will only know if you have past data.
Use a Coupon
Everybody wants to pay less. If you’re advertising a product or service, include a coupon. If you’re running multiple campaigns, make sure to put a code somewhere on the coupon so you know which campaign the customer responded to. This works for both online and print advertising. For radio and TV, you could give them a coupon code to mention at the store or to type in online. You could also offer an incentive. “Mention this ad and we’ll give you an extra 15% off.”
Use E-mail Software
Email marketing is one of the most cost-effective ways to advertise your products and services to existing customers and to prospects. By getting customers, prospects and casual shoppers to sign up for your mailing list, and then sending mailings on a regular basis, you give yourself repeated opportunities to get their business or their repeat purchases at minimal cost to you. One tactic to build your list: ask people to sign up for your mailing list to get a discount coupon or free information.
Don’t use your gmail or a personal account for email marketing. Individual email accounts don’t give you any way of measuring the results. Instead, use software like Constant Contact, Mailchimp, Aweber or one of the many others that provide metrics like open and click rates. This is the most effective way to track the success of your e-mail campaigns. As you get more familiar with these programs you’ll find plenty of other powerful features that you can use to better target your customers and improve the metrics.
Related Advertisement: Constant Contact Special Offer
Advertising is an art because what should work sometimes fails. Every advertising pro knows that they must make multiple versions of an ad and test them against each other. For example, if you have 30,000 people on an e-mail or mailing list, you might send 10% of the list ad #1 and 10% ad #2. Once you see which ad performed better, the other 80% will receive the winner of the split test.
You can split test with more than 2 ads, of course, but make sure you’re only testing one condition at a time until you gain experience with testing ads. If you change the art and copy, for example, you don’t know which of the changes resulted in a better or worse performance.
Once you do understand split testing and do enough volume to warrant it, you may benefit by using multivariate testing (ie, testing several variables in a single ad) to see which combination of elements best achieves the result you want to attain. Google Analytics Content Experiments is free to use, but takes some time to learn how to set up properly. There’s also commercial software such as Optimizely available to do multivariate testing.
Social media advertising has come a long way since its humble beginnings. Now you can see performance data in stunning detail, and you can target specific demographics. You can make changes to your ad early in the campaign if you see that it isn’t working, and you can split test different ideas and compare data for each.
But first, you must understand how to read the data. Before starting a social media marketing campaign, understand the platform and understand how to interpret the data. Don’t spend any money until you master those two items. Same is true with search engine advertising.
Set Your Website Up Correctly
Create custom landing pages for your ads. For instance, if you own a florist and sell flowers and plants, instead of making all of your ads go to your home page, have separate ads pointing to individual pages for holidays such as Easter or Mother’s Day. Have other ads featuring sympathy arrangements, or get-well plants and floral arrangements, each pointing to an appropriate page. You accomplish two goals this way: One, accommodating customers by sending them straight to what they were looking for, and two, allowing you to easily measure how well your ad got people to your site.
If you do want to send people to your home page from various ads, use tracking code in the ads so you can tell which ads actually delivered people to your site.
Tracking visitors to your site is an important metric but also tracking how they interact is equally as valuable. If you aren’t already using it, Google Analytics is a free tool that will help you better understand what people are seeing on your site and how they move through it.
Go Old School
Before the days of sophisticated measurement data, business owners would ask the simple question, “how did you hear about us?” Even in the 21st century, that’s a great question to ask. Not only will it give you valuable data, it’s an effective icebreaker that might result in building a relationship with that customer