For our latest Google Marketing Platform webinar, we spoke to Google’s Mariana Valdez about smart bidding. Here, we’re sharing the insights.
SB_cogs2.gif
Show the right message to the right customer at the right moment.
You’ll learn how to:
- Expand reach and scale
- Free up bandwidth for more strategic tasks
- Achieve higher ROI
Consumers today are more curious, demanding and impatient than ever before. Their journeys are complex and made up of multiple intent-rich moments, presenting an opportunity for us as marketers. As machine learning evolves, we can better connect with customers and respond to user intent effectively by showing the right message, to the right customer at the right moment.
Learn how Google smart bidding solutions make this possible by helping you work faster (expand reach & scale), smarter (free up bandwidth for more strategic tasks), and win more (achieving higher ROI).
Google Smart Bidding: Be a Marketer of the Future
3 Goals:
- Machine learning
- The consumer journey
- Smart bidding
It takes a lot of manual work to improve ad relevance, quality score, conversion volume and impression share. Many advertisers struggle, but many thrive. Google has learnt from the top-performing search advertisers and knows what sets them apart.
Machine Learning
Google knows that what sets these advertisers apart is their use of machine learning. Because when you combine the creative smarts and strategic wisdom of a human and the cognitive and raw processing power of artificial intelligence, you’ll optimise performance.
This learning software is called a neural network and it very loosely mimics a human brain. It’s made up of millions or billions of neurons, which are simply small computational units that each do a simple computation and then pass the info to connected neurons. When these neurons are connected as a large network, it’s able to recognise and learn fairly complex patterns in the data.
With “deep” learning, the neurons are arranged in layers. Each layer learns patterns from the layer below, learning patterns of patterns of patterns. This means the highest layers can learn more abstract and complex patterns. This could be such as how to identify a cat or dog or even what a party looks like. This is why deep learning is so popular.
How does it work? See for yourself. Search for “dogs” in your Google Photos library to find your favourite furry friends and moments. Even if you haven’t captioned any photos, Google can find the ones that have the object, animal or action you’re looking for.
Here’s a video on how a machine learning algorithm uses data to optimise its performance over time. This clip is from a talk with the CEO of Google DeepMind, Demis Hassabis, in which he explains how DeepMind trained a machine to play Space Invaders.
First, you’ll notice the algorithm makes a few mistakes as it learns. With each game, it gets better and better, until finally, it knows every strategy and dominates the game. It’s this ability to learn and improve over time that makes artificial intelligence such an important part of Google Ads.
The Consumer Journey
Mobile has drastically changed consumer behaviour and users have never been so connected. With the answer to almost any question within just a few taps, users are curious and researching everything.
People are searching for answers about small, seemingly mundane things. Users are research-obsessed. Because they can know everything, they want to know everything. Every decision they make needs to be informed.
In 2015 – 2017, Google saw mobile searches for “shower curtains” and “best toothbrushes” more than double. Marketers used to define high and low consideration categories, with cars and appliances high and men’s underwear and deodorant low. Now, consumers dictate the level of consideration. Every product can now be high consideration, raising both the opportunity and the stakes for marketers across categories.
Where to Buy
Google has also seen a rise of 85 per cent of mobile searches for “where to buy (product)”. Consumers know their phone can find products near them and let them know whether they’re in stock.
Users expect assistance at every moment. They are more demanding and impatient than ever. They expect high-quality, relevant answers and information and they expect to receive the information quickly.
The Customer Journey No Longer Follows a Linear Path
No two customer journeys are exactly alike. Most journeys don’t resemble a funnel at all. Due to the growth of mobile, the customer journey is also growing and evolving. So, what does that mean for marketers?
- This gives rise to complex consumer journeys with multiple intent-rich moments across multiple touchpoints across networks, platforms and devices.
- It’s necessary to understand these fragmented journeys and how to engage with users throughout all these moments of intent. But doing so manually would be extremely challenging.
- Thanks to machine learning, we can now understand and act on consumer intent in real-time.
So how can we understand and act on consumer intent in real-time?
Smart Bidding
So we’ve talked about machine learning and how it works, how consumer journeys have become more complex and how we need the help of technology to make sense of all the information across those complex journeys.
So why is bidding a challenge to begin with?
- If you don’t bid efficiently, you may miss valuable conversions
- The “right” bid can be a moving target that’s hard to reach at scale
- Identifying the right uplift for each signal for each keyword and bid would require an army
- 10 years ago, bidding was pretty straightforward. You simply chose keywords and specified bid based on the likelihood of that keyword to convert on a last click, same session, desktop basis.
- Then, mobile devices came along, which created more signals that marketers needed to account for, including device type, time and location.
- Today, we can track more of these signals than ever before. While this means more complexity, it also means more opportunity.
The key to masterful bidding is to adjust your bids based on each user’s unique combination of signals. Asserting all of these manually for every auction is impossible.
Machine learning helps drive the best performance for your ads through automation. The Google Ads Smart Bidding algorithm is constantly learning based on real-time feedback and adjusting to improve performance.
Machine learning, through smart bidding can help you to answer these important questions:
- How do I find my ideal audience?
- Google uses machine learning to find users who aren’t your customers but should be, helping advertisers expand their reach to people likely to convert.
- In short, Google is able to effectively capture intent through its Audience solutions.
- How can I measure performance?
- Google uses machine learning to understand the true drivers of conversion and assign the right value with data-driven attribution.
- How much should I bid per auction?
- With smart bidding, Google factors in countless signals to determine the optimal bid for each moment, so you can acquire more customers and eliminate wasteful spending.
- What message do I show my audience?
-
- With machine learning, Google uses the understanding of consumer intent and data to choose the right creative to show for every moment.
Google Smart Bidding Can Help You…
Work Faster
With Google’s Machine Learning capabilities you can rapidly analyse millions of signals and proactively set real-time auction-level adjustments.
Google’s Machine Learning algorithms can analyse up to 70 million signals within 100 milliseconds, but the number of signals processed for Smart Bidding adjustments varies per auction.
Work Smarter
Automate routine tasks like bidding and creatives to free up more bandwidth for strategic thinking and focus on tactics that move the needle. Imagine what you could do for your account with 5-10 more hours a week!
There’s no need to set bid adjustments when using smart bidding – this is done automatically.
Win More
Win more by driving growth while achieving a higher ROI when using a fully automated strategy.
- Advertisers have seen an average uplift of 30% increase in conversions at a similar CPA when using the Target CPA bid strategy on Search & Display
- Advertisers have seen an average 50% increase in conversions when using responsive display ads with image ads
These strategies help you achieve an improved performance goal without surpassing the financial limits you’ve set.
THANK YOU!
Huge thank you to Mariana Valdez for the insights! We may get her to speak on this topic with us again when we get back to in-person events for our Google Marketing Platform Sydney meetups.
Join us for our next Google Marketing Platform events:
