Introduction to Paid Media for Marketing
Social media gets a lot of attention because it’s free and easy to get started with. But the downside is that you have to spend significant time and effort to build a following and you might not see results for a while. Paid ads are the other side of the coin: instead of paying with your time, you pay with your wallet. And ads start working for you immediately.
Knowing how to use both techniques means you can balance your investments of time and money. To help you get started, I’ll walk you through a few ad systems I used recently.
This is the granddaddy of them all. AdWords lets you buy short text ads that consist of a headline, a message and a URL (i.e. the landing page). You target these ads against keywords, like “houses for sale in scottsdale“. When someone does a search for those keywords on Google, your ad might get shown. Also, Google might show your ads on websites that are part of the Google Display Network, if the website content matches your keywords.
Google uses an auction system to determine which ads to show, and in which order. Two factors are in play here. First, you bid an amount that you are willing to pay for a single click on one of your ads (the cost per click, or CPC). Second, Google automatically determines a quality score for your ad-based on several factors, including the text in your ad and your landing page.
When Google needs to decide which ads to show for a keyword, it ranks ads targeting that keyword based on a combination of the bid and the quality score. The more you bid and the higher quality your ad is, the more likely your ad is to show (and the higher the position it will show in).
It’s important to choose your keywords wisely. There’s a lot of competition for keywords like “san diego real estate“. Targeting against more specific keywords means lower costs and possibly, higher quality scores.
Facebook lets you advertise an external website, your Facebook page, or a particular story on your page. You can specify an image, a headline and some text for your ad (unless you’re promoting a story, in which case Facebook just shows that story’s headline).
In a lot of ways, Facebook’s ad targeting options are a lot richer than Google’s. You can target based on location, age, relationship, education and workplace. Most importantly, you can target based on a person’s interests, which Facebook figures out based on information on their Timeline and things they’ve Liked. These interests can be things like “sports cars” or “National Association of Realtors”.
As with Google, Facebook’s ad system is auction-based, so you’ll need to specify a bid. And like Google, Facebook’s auction takes into account both the bid and ad quality.
Twitter is still experimenting with their ad products (Promoted Tweets, Promoted Accounts and Promoted Trends). They’re currently only available in a beta that requires a minimum expenditure of $5000 a month.
If, like me, that’s too much for you, you can use a system called BuySellAds to buy sponsored tweets. It’s basically a directory of Twitter accounts with huge followings, who are willing to tweet a message you specify for a fee that varies depending on the account. The tricky part is finding an account with a following that matches the audience you’re trying to target.
Measure, Evaluate and Experiment
It’s important to measure the performance of your ads to figure out if they’re worth what you’re spending. Both Google and Facebook provide reports on the click-through rates and costs of your ads. BuySellAds doesn’t provide any reporting, so you’ll need to use your own analytics to figure out how many visits you got from your BuySellAds tweet.
In my case, I found that Google and Facebook were pretty effective and sent me significant traffic. Of the two, Google was 40% cheaper. BuySellAds didn’t work very well for me at all; I got very little traffic from the tweet I paid for. This data helped guide my next set of investments.
Finally, it’s important to experiment with your ads. Different wording or a new image can dramatically affect your ad’s performance. But be careful, because it’s really easy to draw conclusions that are not statistically significant. Unless you have a big budget, I would recommend only testing two variations of an ad at a time, and use the Split Test Calculator to determine whether the trend you’re seeing is really statistically significant.
Derek Cheng works on abodograph, an app to help realtors market themselves with beautiful photos. Follow him on Twitter and check out the abodograph blog.