Branded campaigns in paid search are often the most profitable and successful PPC campaigns a business can create. It’s no wonder why this is; branded campaigns receive the benefit of all prior branding work, so if you’re not seeing success through brand campaigns, it’s because you haven’t yet built your brand. But there is a hidden problem that creeps into both branded and non-branded campaigns alike – dishonest attribution. In this post, we’re going to explore how you can properly plan out your branded and non-branded negative keyword strategy so you can make your campaigns a little more honest.
If you scour your search terms reports in AdWords, you may find some interesting things. If you’re running branded keywords, you may be receiving clicks (and who knows how many impressions) from non-branded keywords. And if you’re running non-branded campaigns, you’ll often find branded queries coming through on your non-branded keyword selections. Often, with further investigation, you’ll find that your data is less reliable than it should be.
Managing PPC comes down to the numbers. Good PPC managers make nearly every ongoing decision in PPC management based on the results of the campaigns they’re running. If the data that is being created is not being fully honest, more budget may be put at risk than is necessary for the intended result. An example: if I’m running a non-branded campaign on the keyword “widgets”, and I see that we have 20 conversions at a $30 CPA (assuming this CPA is profitable), I will be inclined to increase my bids and budget for this campaign. If, however, 10 of these conversions came through on queries that include my brand, say “Digital Third Coast Widgets”, then this may drastically change the CPA outlook. In the chart above, we can see the results of our non-branded campaign. We have a 10% conversion rate, and we may decide that we’d like to put more of our advertising budget in this campaign and we may increase our bids accordingly. However, once we break down the results and bucket them by search intent – brand vs. non-brand – the outlook completely changes. Suddenly we’re paying 11 times as much for a non-branded conversion than most basic reports in AdWords would lead us to believe. Perhaps worse, if I increased my bid based on these results, and my conversion rate on non-branded terms does not improve, I may just end up spending a lot more money without really affecting the results I need.
Unless you are running exact match keywords, your keywords have the opportunity to trigger ads for many different keyword combinations. In our example, our ads for the keyword “widgets” will be eligible to display for keywords like “free widgets”, “the best widgets” and even branded keywords like “Digital Third Coast widgets”. Further, this can happen even if we have a dedicated branded campaign. In this case, the dishonest attribution is a function of how ad rank works in AdWords. Google selects the ad with the highest Ad Rank in an account for each ad auction. Ad Rank is determined by the equation Ad Rank = Bid * Quality Score. Based on this equation, sometimes, because of Quality Score, a branded term can win an auction for a non-branded query, while other times, a non-brand term can win for a branded query due to a high bid. It all depends on how the Ad Rank in that particular auction shakes out. *This really only happens when you have a broad match branded term, more on this later.
This separation of brand and non-brand is classic AdWords and PPC best practice. We should always be doing this. Always. But as we saw above, even this move is not enough to thwart all undesirable matches.
The best way to ensure your brand campaign remains honest is to anchor your brand terms. This is best done by avoiding Google’s default match type, broad match. Broad match has its uses, but if you absolutely know what your intention is with a campaign, its use is ill advised. Broad match makes it possible for brand terms to show up for non-branded query, and it happens with some regularity if the right bids are in place for branded terms. At the very least, instead use broad match modifier on your brand terms. You can also use phrase and exact match if you have a very good idea of what types of searches people are performing when looking for your brand online.
Now, if you want to keep your non-branded budget focused on non-branded queries, simply add your brand terms to your negative keyword lists in your non-branded campaigns. If there are common variations or misspellings, these should be added as well. Moving forward, your non-branded ads should never appear for a branded query.
The biggest advantage is that you’ve now made your AdWords account more honest. But it doesn’t end there. The real point of choosing a keyword and grouping it in a certain ad group within a campaign is to trigger a specific ad when that keyword is queried, since you may want to deliver a specific message to people based on their query. Now you will have a better format for more effective testing and you will ensure you’re getting the right message to the right person at the right time.
Once we do our due diligence and set up our campaigns as outlined above, there is still another consideration. Just because we have made our campaigns more honest does not mean we succeeded in making them more competitive. In general, non-branded keywords require a greater bid than branded terms will to achieve a high ad rank. If it’s common in your industry to see queries that include both your brand name and a highly competitive keyword, we’ll have to rethink the approach slightly. For the widgets example, if “widgets” is a competitive term, bidding on a branded keyword like “Digital Third Coast” might not be enough to grab the high spot for a search query like “Digital Third Coast widgets”. Sure, our quality score will be very high, but our bid might not be enough to secure the position we want. In these cases, it’s recommended to add additional ad groups in your branded campaign so that you can bid differently for the different variations of how your brand is searched.
In this example, it may require a larger bid to defend your brand when a competitive term accompanies your brand name in a user’s search query. Doing so will still yield a better CPA than letting your non-branded term pick up this search. Plus, your ad will likely be much more relevant since you’ll better be able to cater the ad to this class of query.
Setting up this groundwork will not guarantee success. There are many other factors that will dictate the profitability and long-term success of your PPC account. But what you will have accomplished is setting up a solid foundation that should allow for better data collection and easier identification of areas of opportunity. If you can spend less time digging through reports to understand the true value of search queries, you can spend more time on improving other areas of your campaign.