how to get the most out of smart bidding
Smart bidding incorporates machine learning algorithms designed to adjust maximum cost per click (CPC) bids at auction level in line with a given prerequisite target. With the right approach, this tactic can bring automation to an often-labour-intensive area of paid media management while keeping performance on track.
Smart bidding is also useful when paired with less restricted match types, like broad match, as it can surface new keywords and opportunities by uncovering search queries that perform but haven’t been considered within the keywords. Alongside this, smart bidding enables enhanced performance through audience-led campaign types, such as display, video and discovery.
However, smart bidding is just one piece of the puzzle. To really make the most of it, you need high-quality data, and you need to know how to use it.
In this blog, we discuss:
the difference between smart bidding and manual bidding
why smart bidding needs data
if smart bidding is right for you
smart bidding and manual bidding: what’s the difference?
Smart bidding and manual bidding are common bid strategies used in PPC campaigns. As the name suggests, they differ greatly and have their own strengths and weaknesses. But first, what exactly are they?
smart bidding: A subset of automated bid strategies that use machine learning to optimise for conversions.
manual bidding: The more traditional approach, manual bidding is when the advertiser manages keyword bids on the Google Ads platform to make decisions and take action on their ads. This is done without any machine learning support and is usually guided by experience and PPC knowledge accompanied by time, patience and analysis.
Head of Paid Media, Tom Walkden, explains further:
“When we talk about manual CPC bidding, what we’re doing is dictating to Google the maximum we’re prepared to bid to get a click on a keyword. For many years, we would use manual CPCs combined with scripts and rules alongside continuous manual optimisation to drive the best results. But as things have become more competitive online, smart bidding has unlocked real-time bidding which has helped make optimisation easier and more effective.”
smart bidding and the need for data
When it comes to smart bidding, the more data, the better. The default signals used by smart bidding to make bids is controlled by the information that Google provides. To get the most out of smart bidding, it’s important to add additional data sets and signals to this.
If you don’t give the machine enough data to go off and you don’t allow it to learn based on more than Google’s default data, your results won’t be as effective as they could be.
This is where the need for high-quality data comes in:
custom intent audiences
Custom intent audiences are a Google Ads feature that allows businesses to control who sees their ads based on the user’s previous online activity. This allows you to reach warmer prospects who are already searching for what your business offers.
This type of data is more regularly used for display and video ads. However, the audience data is built upon search activity so we can create a custom intent audience based on what users are searching. This data can then be applied to our display, video and discovery campaigns and paired with a smart bidding strategy to make decisions on whether bids need to be lowered or increased based on whether someone has been searching for the product or service.
custom affinity audiences
Custom affinity audiences are based on what users are passionate about such as their habits and interests.
You can create your own custom affinity audience tailored by interest keywords, competitor URLs and apps. For example, a user may not have visited your website directly before or searched for your product or service, but they could have shown interest in a competitor. We can take this insight to start showing relevant ads across the Google Display Network to these users too.
First-party or customer data is data that your business has collected first-hand. In many cases, businesses don’t realise just how much first-party data they hold but need the support of an expert agency to improve the quality.
First-party data fits into the cookie-less future we’re facing with third-party cookies due to be removed from Google Chrome by the end of 2023. To help prepare for this update, it’s a good idea to start collecting customer data in a GDPR-friendly way now. This could include:
improving first-party data collection with data collection exercises such as competitions
uploading data for Google customer match
leveraging Facebook lookalike audiences
working with Experian to produce and analyse first-party data for Experian profiling
is smart bidding right for your strategy?
There are a number of benefits smart bidding can bring to your paid media strategy, these include:
advanced machine learning
Backed by powerful technology, smart bidding uses machine learning algorithms on a vast scale to help you make accurate decisions across your account on how bids can impact goals. These algorithms factor a whole host of parameters that can impact performance - more so than a single person or team would be capable of doing.
With smart bidding, you can factor a wide range of signals in your bid optimisations. These signals are attributes about a user or their context surrounding your ads at the time of an auction such as device, audience type, demographics, language and location.
flexible performance controls
Smart bidding allows you to set performance targets and customise settings based on your business goals so you’re always on track to reach your objectives. This can be done in a number of ways, for example:
optimise search bids to your selected attribution model
set performance targets specific to specific devices
Smart bidding provides tools that give deeper insight into bidding performance, allowing you to troubleshoot any issues as they arise. These include:
bid strategy reports: To help you understand how your smart bidding strategies are performing.
bid strategy statuses: To show you what’s happening with your strategies under the hood.
campaign experiments: To make it easy to test how well smart bidding is performing against your current bidding method.
simulators: Forecast how your ads may have performed in terms of key metrics such as cost, conversions, impressions and clicks when different CPA, ROAS targets or budgets are set.
alerts and notifications: Flag issues with conversion tracking and provide clear steps for fixing them.
no delays in changes
On Google, auctions happen in microseconds. Let’s say you’re updating your bids every six hours. A major change could happen in that time that you’re not aware of until you check your bids again which can result in lost opportunities and wasted budget.
This is where smart bidding blows manual bidding out of the water. Every search runs Google’s algorithm in the background to look at all the organic results it should be showing. It then looks at all the advertisers who are bidding on the keywords that have been triggered. The AdRank algorithm is used to determine your quality score and if this is a high score, your ad will be more likely to show in a prominent position. This is constantly working in the background to ensure you’re getting the best performance out of your budget.
To learn more about quality scores and how they can impact your ad performance, read one of our blogs on the topic:
smart bidding and your strategy
Smart bidding is often a welcome addition to paid media strategies. If you have a lot of historical data to use, want to reduce how much time you spend on campaign monitoring and have specific goals you want to reach, this tactic is made for you.
However, it’s important to remember that you know your business better than Google which is why you should provide as much valuable data as possible to support smart bidding and its learnings.
Working with a specialist agency is where the real value can be uncovered. If you want to know more about how we use smart bidding to drive even better performance for our clients, get in touch.