What is Smart Bidding?
Smart Bidding can briefly be described as automated bid management based on conversion data. It is based on the technology of machine learning and uses numerous user signals starting with the search, through interactions to the conclusion of the customer journey. In this way, individual conversion performance models are generated and used to make the best use of the marketing budget.
Why is Smart Bidding important or perhaps even necessary?
Before Smart Bidding, all bids were still maintained manually. In the monitoring process it was necessary to evaluate whether the currently selected bid could meet or support the company’s goals. With constantly growing keyword numbers and the newly gained freedom to bid on different levels such as time, place, device, demographics, target groups, topics – the list does not end here – the scope and effort of this monitoring is rapidly increasing.
Since the available support cannot keep up with the day-to-day business, the operative work takes up more and more space and thus makes the unfolding of strategic developments more difficult.
In addition, it must be taken into account that probably nobody – and we hope nobody! – can or must deal with it 24 hours a day, every day, the whole year. However, since the Internet never sleeps and the performance of a keyword or a product should be evaluated differently, even outside of monitoring, there is a need.
And Smart Bidding meets this need! The question of whether automated bid management is now important or necessary becomes superfluous with the increasing number of new features and possibilities in online marketing. The real question is therefore not so much whether, but rather which Smart Bidder is the right one for me?
Google Smart Bidding
Smart Bidding from Google has actually been with us for several years. Many people probably still know the conversion optimization tool, which had to make way for today’s smart bidding. Today, after about 3 years, various Smart Bidding systems are available, all of which are subject to a predefined business objective. For example, there are Smart Bidders available who want to maximize conversions on a budget target or who strive for a profitability target as in the case of target ROAS or target CPA.
However, all Smart Bidders are designed for specific campaign types. However, the available Smart Bidders become more and more flexible over time and therefore this exclusivity also gradually disappears. For example, a Smart Bidder like the Target-ROAS can also be used in the display, but on the Target-CPA for shopping campaigns we still have to be patient.
Not only in the choice of campaign types, but also in terms of fine-tuning, Google was able to deliver. While with fully automated systems such as the Target-ROAS you could only watch, today a minimum and a maximum bid can be set for the entire strategy according to which the smart bidder should orientate himself.
The emphasis here, however, is on the word “one” minimum and “one” maximum bid for the “entire” strategy and this brings us to the problem of this topic. The need for automation is there, but what about the willingness to give up the reins almost entirely?
The problem with Google Smart Bidding
We already know the advantages of Google Smart Bidding, but what are the disadvantages? Google is known for providing a lot of information, but also for wrapping a lot of information in a black box.
Targeted interventions or fine-tuning are only possible in a few cases, and in some cases can only be enforced through complex campaign structures or other dependencies through other Google or third-party systems. Small sub-strategies such as the promotion of a product or product series, the general highlighting of a specific brand in the product range or seasonalities can quickly become major challenges.
Many advertisers probably wouldn’t see these reasons as a problem because automation in general is supposed to take the work off your hands, but does it have to be the whole thing? After all, what happens if the company’s goals are met, but not to the extent or with the framework conditions that one had imagined or for business reasons.
For those who don’t want to miss the smart bidding train without giving up the reins completely, but also for those who don’t want to deal with bid management and are looking for alternatives, we can help.
Our Smart Bidding Solution: BiddingLab
The Smart Bidding System from Sunlab fulfills the requirements of full automation on the one hand, but also enables teamwork between man and machine on the other hand.
Account or campaign goals can be set holistically, just like with Google Smart Bidding, and if you wish, you can leave it at that.
But it does not have to! Based on a tree structure, further strategies can be added to the main strategy, which can even pursue other business objectives without being completely excluded. For example, the lower and upper limits of the CPC range to be used can be set differently for different strategies or an aggressive bid management can be selected to force visibility on the 1st page or even into the top position.
Further, these strategies can be timed in a roadmap. The default setting – as probably with all Smart Bidders – starts with the day of activation and works with an open end until the strategy is deactivated again. However, this can be delayed if you wish. In addition, specific days can be selected in the roadmap if you want to suspend the bidder on these days, or from the other perspective, if you still want to continue to use primarily manual bid management and then only call in the bidder for support on specific days, this would also be possible. With the latter, however, it must be borne in mind – and this also applies to all smart bidders – that this variant involves a significantly higher risk if the data basis is too small for short-term use.
In addition to the roadmap and the strategies, however, further requests for adaptation can be set up via rules. The choice of the level of impact is particularly diverse here. For example, in shopping campaigns, rules can be applied specifically to IDs, brand, product groups, user-defined labels and other metadata. If not all applicable elements of the selected effect level should be affected, a selection can also be created to define very specific performance groups. The available rule actions here can be a push via multiplication, the overriding of higher-level goals or even visibility forcing as on the strategy level.
But “From great strength comes great responsibility” is the saying and this is also true here. Even if the BiddingLab gives you these freedoms, you should consider how useful it is to invite numerous different conductors into the orchestra over a long period of time. Ideally, these rules should only be used for a short period of time, for example for a sale event lasting a few days. Overall, the more harmonious the rule objectives are, the better it is for the overall marketing objective.
So it is necessary to find the so-called golden mean of automation and manual direction.