Offer Testing: The Best Way to Start Display Network Advertising
Google’s display network is massive, encompassing more than 4 billion daily page views, 700 million monthly users, and reaching more than 80% of the online audience.
Yet, with all this inventory, many marketers fail with display marketing.
The major reasons are they either are trying to reach the incorrect goals or they don’t understand how all the targeting options work.
In today’s column, I’m going to walk through how to get started with display advertising so you can avoid some of the most common mistakes that cause marketers to fail with the network.
Most marketers just use Google’s default option for display – add some keywords and see what happens. Often, these are a list of the account’s search keywords. This is a mistake.
For search, Google really cares about your match types for both positive and negative keywords. On display, Google ignores match types. They will use negative keywords to help them with the ad placement, but if you just import thousands of negatives from the search campaigns, often Google will not do a good job serving your ad. Usually, you want just a few keywords in an ad group; and there will be times you will not even use negative keywords for display, which is a cardinal sin for your search campaigns.
In the past, Google favored a theme based approach to choosing display keywords. Over the past few months, they have been giving a lot more emphasis to individual keywords. This has made it easier to take your search keywords and run them on display, but its still not a perfect system.
So, a good starting place for display keywords is to take the broader search keywords (the 2-3 word variations, not the 6 word combinations) and duplicate them in a display only campaign. If you have a few negative keywords you really want to use – feel free- but do not import your thousands of negatives into the display campaign.
The biggest problem with most search ad groups is that there is no granular organization between the ads and the keywords. This is just as true on display. When you are creating display ad groups, first determine the ad and landing page for each ad group. After that is determined, then use keywords that match both the ad and the landing page. If a keyword doesn’t match both, it needs to be in a new ad group. This segmentation is good for both search and display.
However, if you just target your display ads based upon keywords, not matter how good the keywords are, then you might find your ad on a large variety of sites. Some might bring you good results, others will fail, but you don’t really know if it’s the site’s traffic or the offer – your ad and landing page.
AdWords offers a lot of targeting options for display. One of them is known as placement targeting. With placement targeting, you can choose an ad placement on a site within the Google Display Network and only show ads on that particular site or ad slot.
Using placement targeting takes the guess work out of the main keyword problem: Is it the site or the offer that isn’t converting?
If you use a research tool, such as doubleclick Ad Planner to find placements where your ideal customer spends time online, then you can ensure your ads are only being shown to your top prospects.
The largest issue with placement targeting is some of the best sites have a large variety of traffic and customers. Targeting NYTimes.com or About.com will lead to a tremendous amount of impressions, many of which are not valuable for your offer. With placement targeting, you can target a section of some sites, such as the business section on the New York Times; however, even that section has so many daily page views that most companies can’t afford to run an ad on every page view.
Therefore, what we really want to do is to choose placements where our audience spends time, but also layer-in a keyword filter.
This is where the real magic of display targeting comes into play. With flexible reach, you can easily combine multiple targeting methods together so that you ad is only shown when a user hits certain targeting combinations. This options is still fairly new; but it will one day be the default option.
With flexible reach, you can set your targeting so that the ad is only display if a user is on a placement you choose and the article matches your keyword selection.
With the combination of display keywords and placements, you should be certain that the placement is not the problem if you don’t receive conversions – it’s a hand picked placement.
If you have segmented and chosen your top keywords, then the keywords should not be the problem. However, always check this by examining the actual URLs of the pages where your ads are being shown. If they are not appropriate, then you should refine the keyword list.
So now we’ve taken the variables of placement and keywords out of the equation. Therefore, if the offer doesn’t convert – its probably the offer.
With the display network, you are not always reaching a user who is far down in the buying cycle. In some cases you will reach users far in the buy process, in other cases you will not. Therefore, you need to test your offers between hard conversion activities, such as sales, versus soft conversion activities, such as white paper download and email subscriptions, so you can find what is going to work best for your display objectives.
Please note, as you become more sophisticated with display and add additional targeting options, such as remarketing, then you may be running different offers by targeting type or placement. Before you can get to that point, you need to make sure that you can find the simplest of combinations that will allow you to attract new customers using display targeting.
The display network offers a tremendous amount of targeting options, control, and inventory. However, if you don’t find an offer that works across the display network, then you will end up wasting money and come to the conclusion that the display network doesn’t work.
This is rarely the case. It does happen on occasion that display advertising doesn’t work for a company. However, most companies can find success with display, if they first find an offer that works.
The best way to find what offer will work is to take out the ad serving variables, such as site quality and user interest. By using flexible reach and only targeting articles that match your keywords on high quality sites, you can remove the variables. Once those are removed, what you are left with is the offer. Then, by doing some simple offer testing, you too can find success with display advertising.