As an SEA marketer you increasingly feel like a kid in a candy store: you see potential bid adjustments all around you, but what are the most interesting for you and your customer? And how to use them? We take a quick look at the most prominent:
Device Bid Adjustment
New, newer, newest! We begin immediately with the latest bid adjustment that AdWords offers us: the device bid adjustment. The newest kid in town allows you (as the name might already suggest) to adjust your bids according to the device type on which the user is doing its query. It was of course already possible to change your bidding when you browse via smartphone, but now you can do the same for desktop / laptop and tablets.
Where do you start? Of course you can choose to create specific campaigns for each of the three device types, but to me it seems to lead to an overload of work. Keep in mind that you not only spend time in the set-up (once), but also with the fact that your time you spend managing your campaigns will double/triple.
Demographics For Search Ads Bid Adjustments
Demographics For Search Ads, or DFSA to keep it simple, is the original basis of this article. It's obvious why many marketing managers get enthusiastic with the discovery of this feature: at last there is a touch point in the digital search story what they can match with the traditional offline campaigns.
Important to know is that a lot of your data will still fall in the 'unknown' category (until now only people with an account linked to Google are classified). Another good tip is to keep your bid adjustment to the specific gender and age categories set at + 0%, otherwise you will collect no data for these categories.
RLSA Bid Adjustment
Here you’ll add a bid adjustment to your remarketing lists for Search Ads. For RLSA this is usually interesting as your RLSA Audiences consist of people who have already chosen for your brand in the past and therefore are more likely to do so again. That means that these people are more interesting for your brand. That in turn means that you can pay more for a click for one of these people as it is more likely that you will generate conversions out of these clicks compared to a click from someone who is not yet part of a you RLSA Audiences.
Location Bid Adjustments
Sometimes it is interesting to examine at which (geographic) locations you get the best results. This is also a passionate plea to not just set your targeting on the entire country, but to be more specific (example at county level). You will soon find out that there are indeed differences between different areas.
A small example is the Brussels region. For a number of customers we saw that the number of clicks in Brussels was very high compared to other locations, while our conversion volume for the same Brussels ratio was again quite low compared to other locations. This is probably because many people quickly look something up during a 5 minutes break at work and then when they come home they effectively place their order (=> hello RLSA).
Ad Scheduling
The last (but not least) bid adjustments which I discuss here is concerning ad scheduling.
By default, your AdWords ads appear 24/24, 7/7. When you do an analysis of your results (you can segment your data based on time of day or day of the week for example), you will be able to determine quickly that your campaigns are not equally performing at any given time. It is of course crucial to be present as much as possible at those times of the day when people decide to proceed with a purchase.
The best thing you can do here is to create a mapping of your peak times during a full week (over a longer period in order to exclude the impact of accidents/bugs). Next you can raise bids on those times when you generate the most conversions.
What’s next?
I already listed some adjustments above. Some of these will make your results evolve positively, others might not have the desired effect. The important thing is that you're trying to test new features each time. If it does not, no problem, but at least you know for sure you’re not missing out on additional leads/conversions. Furthermore, it may be that a certain bid adjustment for a given customer may have no effect, while the same bid adjustment sometimes works wonders with another customer. Make sure that you run your test long enough so that you can draw an informed conclusion from your collected data.
How do you decide when you use which bid adjustment? All this depends on your client. What are the KPIs you work with?
A brief example: at one of our clients we work with call-only ads. It is logical that we allow these ads to appear only when the call center of our client is operational. Therefore, we apply Ad Scheduling here.
Another example is when a customer has a clear idea of who its clients are. Thanks to DFSA we can respond to this by increasing our bid adjustment for the demographic segments the customer gives us.
Another question to ask yourself: how big should I set the bid adjustment?
Keep in mind that several adjustments are all accumulated together. Your original bid of € 1.00 for instance can rise very quickly when you have multiple bid adjustments. This means that you sometimes may have spent your daily budget faster than expected (=> hello ad scheduling).
Generally speaking, I first make an analysis of the collected data after which I then compare the average score. If for a certain segment/time of day/location the score is X% above the average, I start with a positive bid adjustment of (X/2)%. I then let the bid adjustment run and reassess the situation. For negative adjustments I do almost exactly the same but here I used a bit more caution and choose for an adjustment of (X/4)% (we are keen to not miss potential conversions/leads). This method allows you to spread your budget better in order to reach better results overall. On the other hand, this way you also limit the risk when your adjustment does not lead to the expected results.
If there is one thing you need to remember from this article it's that you should always test new features. Do you feel unsure? Limit the test to a part of your campaign, or set a lower bid adjustment. After a short period you can evaluate the results and decide whether or not to discontinue your test or (hopefully) continue and even expand.
Author: Vincent Saelen
