Google Ads Academy: Powered by AI
My name is Kate Adams and as a Performance Leader at Google I'd like to talk to you about one of the most promising and rapidly evolving technologies of our time. AI. I'm lucky enough that in my work I get the opportunity to explore new ways in which our technology can help companies maximize their profits while saving time every day.
And it has always been fascinating to me to see how AI helped us get much better at that. When I started at Google searches were typed into the Google Search bar but they could still be matched to relevant, organical page results despite there being spelling mistakes in the search term. Thanks to AI and all the way to today where Bard Google's experimental, conversational service is reimagining how people interact with computers in a helpful, responsible, and intuitive way. But at the same time, I've also witnessed how certain things in my personal life have become so much more easy thanks to these leaps and bounds in technology.
For example, I love taking pictures of my kids especially on holidays. But as any parent will know getting that perfect shot can just take a lot of time. And that's why I really love the Magic Eraser on my Pixel phone.
It automatically erases unwanted objects from my photos or in this case, my husband. It allows me to capture these priceless family moments quicker and focus on what's most important. Digging my kids out from under a pile of sand.
And I'm sure that there are many more examples that you can think of too. Examples when you weren't even aware that you were using AI but were simply making your life easier or better by using auto suggestions to write that email much faster or to drive home via the most energy efficient route on Google Maps. All of these moments are examples of tasks that would previously require a lot of manual work but have now thanks to AI become effortless. The good thing is if you are using Google products you have probably already been using AI for quite a while.
Doing so successfully in your businesses too will help uncover great potential. And that is exactly what I want to talk to you about. I'm confident you'll get some inspiration from the next few minutes where I will share with you on the one hand how AI has shaped and improved Google Search as we know it today.
And on the other hand, I hope to give you a few concrete ideas of what you could be doing to embrace this technological evolution even quicker and more successfully. When it comes to Google Search, AI creates a lot of new opportunities. Like finding answers to open questions doing visual searches, and using Multisearch. Let me start with answering open questions. Thanks to AI, it is now possible to give better answers to questions that aren't definitive such as is Brussels or Amsterdam the best city for expats? Answering these open questions requires a more fundamental way of understanding the topic and being able to distil multiple perspectives to then provide easy to digest information.
Google's development of large language models is instrumental in helping compose construct and summarize this information. The progress on larger and more powerful language models heavily influenced by the Transformer paper that Google published in 2017 has been one of the most exciting areas of machine learning research over the last decade. When it comes to Visual Search it's fair to say that generative AI goes far beyond language and text and it makes search much more intuitive than ever before. One of the most natural ways that people absorb information is visually. Demonstrated also by the fact that people now use Google Lens more than 10 billion times a month we expect your camera to become increasingly complimentary to your keyboard when you seek to explore the world around you and get things done. How easy is it to take a picture of that great pair of blue shoes that your friend is wearing and find out where you can buy a pair for yourself? And then Multisearch, it's one of these exciting new opportunities that AI provides us with and that I already see being used by many youngsters.
Searching for information, using multiple inputs like a picture and text at the same time. It opens up an entirely new way to express yourself but also, to help you not only find those great blue shoes your friend is wearing but by adding the word green to your search you can find them straight away in your favourite colour. It's clear that AI is rapidly changing the behaviour of both consumer and businesses. A very recent McKinsey study found that the biggest reported revenue effects of AI are found in marketing and sales. And when I look at the companies that I speak to it's those embracing our AI-powered tools that see the best business results.
And I'd like to illustrate this with two examples. Smart Language Technology and Text-to-Speech. I already mentioned how large language models are built to deeply understand human interactions. This same technology has supercharged Broad Match a feature that has been around since the dawn of Google Ads. But more recently, thanks to AI improvements it can now help advertisers reach even more people whilst easing the manual campaign management work. So how does it work? Let's take a highly specific search like where can I meet singles in Stockholm? This search term is now able to match with the Broad Match keyword dating website.
Thanks to the recent advancements in large language models AI can tell the two are related even though none of the words in the search term and the keyword are the same. Dating platform Meetic uses Broad Match now for this purpose and saw 70% more registrations at a stable cost per acquisition. But how does this work technically? First, we train the language model through billions of pieces of text to learn what sequences of words make sense to really comprehend the language. We will train it so it will understand all the different variations and meanings of a word or phrase. Only once that fundamental foundation is in place and this intelligence will continue to learn and evolve will we assign it a specific task such as matching the right Ads to those search terms thanks to Broad Match now knowing this contextual relevance You don't need a specific keyword strategy anymore.
And this is even the case if people are searching for your products or services in a different language. This is making it one of the biggest bets for Search advertising today especially when it gets combined with our AI-fuelled bidding technology. The second example I'd like to talk about is Text-to-Speech. When it comes to running a successful video campaign elements like music, sound effects and the voiceover contribute significantly to the results.
Google's Text-to-Speech technology allows advertisers to easily add voiceovers to their ads on YouTube amplifying the effect on their marketing results. Voices can be tailored to represent distinct personas or speaking styles that then allow you to cast just the right voice for your ad creative. For example, if your brand exudes luxury relaxation or warmth a voice with a smooth baritone or easy demeanour could be a great fit. Making every picture breathe the TV brings you closer to reality.
On the other hand, if your brand aims to connect with a younger audience a youthful, informal and upbeat voice might be the right choice. You can use this technology to represent your brand authentically by selecting a voice that embodies your brand's personality and really resonates with your audience. These are just a few examples that illustrate how AI is at the heart of many of Google solutions. But how will you stay ahead of the curve? Google's AI-powered tools are broadly available and in order to get the best out of them you will need to do three things. First, know how to set the correct marketing goals for the AI-powered technology to optimize towards.
If your business goal is to maximize net profit but your marketing KPIs are set up to maximize sales value then you are not set up for success. AI algorithms are very good at optimizing towards one specific goal. So, you need to choose your strategy wisely. Number two, capture value and activate the required data only in a privacy safe manner. Think inductive, using data that you own like your financial data or that it is declared or clearly consented by your consumers and build up from there.
Versus thinking deductive, where you are capturing every single data point you can and working back from there. As you have heard in my previous examples in Search and our advertising solutions AI algorithms are designed to learn from and make predictions based on data inputs. But the quality and the accuracy of these predictions depend on the quality and relevance of the data that is fed into the algorithms by marketers. In this privacy first world, your first party data provides the best competitive advantage for your brand. This data is unique and relevant to only your business enabling you to express what is most valuable to your business and finding opportunities for growth.
And thirdly, invest in developing core capabilities within your marketing team. You will rely much more on people with great skills in marketing strategies data management, creative development and measurement and testing expertise than you have ever done before. Therefore, we recommend that you invest time in getting your teams comfortable with a learning mindset. They will need to upskill in these areas themselves whilst simultaneously learning that they have to embrace that way of working in an AI powered marketing world.
It will require constant measurement and testing along the way to generate those high-quality data inputs that will in turn help improve your business results using AI. So, when it comes to boosting business results, AI has a lot to offer. And the AI-powered tools that Google already offers today will help you stay ahead of the curve. And what's exciting is that it's only early days. There are so many opportunities that are yet to be uncovered from Visual Search to fully AI-generated ads. We never imagined we'd be where we are now five years ago.
So just imagine where we might be in another five years. We will continue to improve and innovate our AI-powered solutions to ensure that you can fully and most importantly safely benefit from the latest technology. We're committed to developing AI responsibly. In 2018, Google was one of the first companies to publish a set of AI principles and we recently updated these. Thank you so much for joining me today. If you are walking away from this AI fuelled episode with any further questions then know that we are here to help you.
I can't wait to hear all of your success stories now and in the future. If you would like to learn more about Broad Match and the technology behind it please go to thinkwithgoogle.com and check out the latest article that was published there. Hello everyone and welcome to Google Ads Academy on Air. My name is Ailbhe Kelly and I'm a Regional Product Lead here in Google EMEA.
And I specialize in Google's AI Powered Smart Bidding and Value Bidding products. So, I'm delighted to be here today to talk to you about how to set up Value-Based Bidding for success by using Qualified Value Bidding. I've worked on Smart Bidding strategies for businesses of all types and sizes and with agencies too over the past five years. So, I hope regardless of where you're tuning in from today you'll learn something new and useful. So first of all before getting into the nuts and bolts of Qualified Value Bidding let's take a quick step back to understand why Value Bidding in the first place.
We know that regardless of what type of business you represent not every customer brings the same value to you. And if customers all bring a different value to your company then why would you invest the same amount for each? And that's where Value Bidding comes in. Value Bidding helps advertisers maximize revenue and Return on Investment by differentiating customers based on their business value and bidding towards higher value customers. Here, when we say business value, it's most commonly defined as purchase value. But it can be profit, perspective value or even lifetime value. And if we differentiate our customers we can then use Value-Based Bidding to bid dynamically and spend our marketing budget effectively bidding more for those who have a higher value for our business and less for those who have a lower value.
And this isn't just theoretical impact. In fact, we've seen the results in numbers too. We now know that when you switch from target CPA or Maximize Conversions to Value Bidding you can actually expect an average uplift of 14% in conversion value.
And that's at a similar ROAS. Value Bidding may be traditionally more associated with online retailers but it is actually relevant for every single business model. So, we know that whether you're working with lead generation e-commerce or omnichannel business Value Bidding will be a strategic step for you to differentiate your customers and grow revenue. What do we mean by Qualified Value Bidding? Here you can think of it as setting the correct foundations and really solid measurement in place for effective and sustainable Value Bidding. So here are the three main ways to achieve Qualified Value Bidding. And the method you choose will depend on your business strategy and tracking capabilities.
The first option here is dynamic values. Value Bidding works best with reliable accurate data. That's actual revenue-based values updated frequently and for each conversion. And this would be the most common setup for an e-commerce business tracking towards purchase or transaction value. But it could also work for a Lead-Gen business for example, an insurance provider with different varying costs. Static value tracking is option two here.
It allows you to optimize towards two or more different conversion actions each with a different static value. An example could be a dental clinic with two conversions A lead form bookings and phone call bookings each with a different static value. Note here that using one single static value will not actually differentiate by sale or user and thus, is not effective Value Bidding. And the third option here is Conversion Value Rules.
An example here could be if you're a university or an educational course provider. And you know that students in Paris are worth twice as much as students in Marseille. So, you can add a layer to your existing values to differentiate by location. And we'll go into all of this in a little bit more detail later.
So, we've learned all about the why behind Value Bidding and we've also learned the main methods to implement Qualified Value Bidding. So now let's talk about the best practices. You might be thinking right now that there are a lot of events and conversions you could track for your business. But the most important thing to remember here is that you should optimize towards the conversion that is closest to your revenue generating event. And in order to decide which one to choose here are the key points to keep in mind. The first one is ability to import or track.
Make sure you're choosing a conversion action that you can actually import daily or track daily. The second point here is to avoid multiple actions along different stages in the funnel. This would conflate your values inaccurately and it can lead to low Value-Based Bidding performance.
The third point here is volume requirements. So, we should aim for a conversion action that has 15 conversions at an account level but ensure not to analyse performance until you have at least 50 conversions. Final point here is to align with business objectives. Always ensure that your campaign strategy and campaign bidding strategy is directly aligned with the business goals. So, you know already that it's crucial to connect your campaign objectives with your business goals so that you can make the most out of your search investment.
So, whether you're tuning in today as a CMO or CEO maybe you're a campaign manager practitioner or an agency partner you should ensure that your performance goals are actually directly in line with the underlying business objective. Your or your client's business goal is to improve revenue commercial margin, or profitability. Good, now that we've uncovered that let's make sure that we align with those metrics in our campaigns. So, let's have a look at the path to effective Qualified Value Bidding. Today, you may be somewhere at the beginning of this journey here tracking only the conversion itself and not the value it has for your business. This basic conversion tracking is what you might use for target CPA or Maximize Conversion strategies.
As a next step, you may transition to static values as a medium-term solution. But the North Star goal for you as a business should be to always prioritize dynamic values if possible. If you have dynamic values available or accessible to your business right now you can actually switch to Dynamic Value Tracking right away. As mentioned, the preferred option is to use dynamic values and it is often the most straightforward. You can implement this with Google Tag Manager or Google Analytics if that's your preferred tracking method. So once the tag itself is set up you should create transaction specific values for your event.
But optimizing towards revenue is not always easy especially if advertisers are not able to track dynamic values in Google Ads. You could be a real estate company with a pretty long conversion delay and no definitive way to track the dynamic value yet not until the sale is closed a few months later. So, if you're in that situation learning how to leverage Value-Based Bidding with static values will be really crucial as a next step. So, let's take a look at how we would implement static values. The first step will be to map out your Lead to Sale journey.
That means mapping out the path your customers take from the moment they first engage with your ad right to the stage of paying for your product or service. Then you can calculate the value of each stage simply by multiplying by the average conversion rate from one stage to the next. As illustrated here. Choose which conversion action as close as possible to the final event that you wish to optimize towards. And remember that you should have more than one action in order to have effective value differentiation.
And both of these actions should be at the same stage in the funnel. Adding the static value into Google Ads is actually really easy. So, after you've created the conversion action itself you simply go in to Edit the Conversion Settings and choose, Add the Same Value for Each Conversion. Tou input your value and it's ready to go. So that means if you have a conversion action ready in Google Ads right now and a value in mind you can edit it right away now.
Now let's talk about Conversion Value Rules. This is a really exciting new feature that was just released last year and it allows you to add an additional layer of value differentiation based on location, device, or audience. You could be, for example, a London-based bank who knows that local Londoners spend twice as much over time as those living outside of London.
Or you could be a Lead-Gen locksmith service who knows that mobile users are worth on average, 50 year or more than users converting on other devices. So, if you know this as a business you can simply add a Conversion Value Rules layer in the Google Ads platform itself that will increase the value accordingly. And this effectively trains the Google Ads algorithm to differentiate users based on this criteria and bid dynamically for more valuable customers.
And the best thing is you can do all this without any developer resources. Now, let's talk about offline conversions. We know that not all of your customers convert online and for some businesses, the majority actually, convert offline. For example, if you're a university course provider you might find that your prospective students browse around on your website download your prospectus and have a call with your team before converting offline a few weeks or maybe even months later.
If I'm going to be investing a lot into my education I'm going to spend some time and research before paying. So how can we train the Google Ads algorithm on conversions that have not happened on your website but instead happened offline? The answer lies in Offline Conversion Tracking. And in a nutshell, it works like this. The user would click onto your ad and onto your website. They will browse around your website and fill out the form.
When they submit the form, they become a lead in your system. But a unique Google Click ID is captured in that form and that is sent to your CRM alongside the lead's information. Then when the lead status changes after your sales or marketing team has a conversation with the customer that information is uploaded back into Google Ads using a Manual or Automatic Scheduled Upload.
Then Google Ads AI trains on this data and it knows which ad clicks actually resulted in offline conversions. But using standard OCI is not the only way to implement Offline Conversion Tracking. In fact, we now actually have a solution that is much quicker to implement and it's called Enhanced Conversions for Leads. This feature allows you to use hashed first party user provided data from your website lead forms for offline lead measurement. And let's take a look at what the implementation steps would look like for Enhanced Conversions for Leads.
So, the first step is you would create the offline conversion action in Google Ads using the source track conversion from clicks. The second step is that you would verify that the Google Tag is installed. Configure the Google Tag and that allows you to send data from your website to linked Google product destinations. You can click the switch to Turn This Feature On to start collecting form events. Then step three, the Google Tag will capture the hashed emails and or phone numbers automatically. Once all this data has been captured we can then start importing it back into Google Ads.
So, the final step here would be to upload the data back to Google via manual or automatic methods. So, you can see here that the process for Enhanced Conversions for Leads is actually much quicker. But there may be some reasons that you cannot implement EC for Leads.
For example, you could have longer conversion windows or your business policy might not allow you to send hashed PII data to Google Ads. So do make sure to work directly with your account manager on choosing the right setup for your business. As discussed already, Qualified Value Bidding requires a differentiation of values.
So, make sure that as you import these conversions back into Google Ads regardless of whether you are using standard OCI or Enhanced Conversions for Leads ensure that you're importing values alongside it. Again, always prioritize dynamic values here if possible. And once you have imported your dynamic value data back into Google Ads you're ready to start using Value-Based Bidding and optimize your campaigns towards your real business value. So, whether you are ready to use static tracking dynamic values or Conversion Value Rules we hope that you've learned something new today and are ready to start growing your business with Qualified Value Bidding.
Thanks for watching, and I hope to see you next time. Hi everyone, my name is Michelle. I'm an EMEA Retail Lead, here at Google. Some might remember me from the Ads Academy Program last year.
I'm delighted to be here with you again this year. There have been lots of changes and challenges in the digital ecosystem over the last 12 months and it has left marketers asking themselves "How can we be sure to drive the most optimal performance with our investment with Google?" We believe that the answer is Performance Max. So last year we spoke at length during the Ads Academy Program about the migration from smart shopping campaigns to Performance Max, and I've no doubt that you're all very familiar with Performance Max now.
For many of you, performance Max is part of everyday digital advertising management. And in this session we'll review the best practices to unlock the full potential of your investment with Performance Max. So let's get started. We have some brilliant performance Max resources we'll share with you via the Ads Academy program you see on screen. These resources include an introduction to Performance Max, a detailed implementation guide, and a best practices guide. We'll go through a lot of the content in these resources today.
nonetheless, I think they're a great thing to have on hand if you have any questions down the line. So Performance Max is our best in class automated shopping campaign, but don't just take my word for it. The results definitely speak for themselves.
Performance Max with the GMC Feed delivers significantly more value for every Dollar, Euro or Pound spent. Performance Max is the newest way to serve across Google Ads inventory from a single campaign and ensures that your agency is always automatically buying the best placements to drive maximum conversion value for your customers. On average, when clients have moved from smart shopping campaigns to Performance Max, they saw an increase of 12% in conversion value. Not only that, but when clients move from standard or regular shopping campaigns to Performance Max, they see an average increase of 42% in conversion value. What's really important to remember is although Performance Max is heavily leveraging automation, there are still some key areas in which agencies can strategically steer Performance Max campaign performance.
First up, we have creatives. We all know that the good quality campaign inputs will result in strong campaign performance. So let's take a closer look.
Agencies can enhance Performance Max campaigns by taking strategic approaches to assets. Asset groups are where you manage your creative assets, including your text, your images, your videos, logos, and your final URLs that are either related to a theme or a common audience. Machine learning will then be able to pick and choose the best performing assets and auto-generate ad formats that are eligible to serve across all of our channels. Using ad strength, you're then going to be able to assess how complete your asset groups are. Some of you might be new to Performance Max, and you might be wondering what exactly is an asset group.
Uh, well an asset group is a collection of creatives that have a common theme or related to a target audience. The creatives in your asset groups are used to assemble all of your ads and build an inventory for all applicable ad formats for your advertising objective. I just want to call out here that it is important to draw a distinction between asset groups and listing groups.
Because they're commonly confused. The products selected and associated with an asset group are called listing groups, and they're very similar to product groups we used to have in smart shopping campaigns. To start, its recommended to have a general asset group, which includes all products you're targeting within the campaign. Consider using this asset group to highlight seasonal messaging or what makes your brand unique. You can then add up to 15 assets in a single asset group to maximize learnings from automation. And it's worth noting that you can definitely create more than one asset group if you need to.
So let's dive into it with an example. Let's imagine that you are a retailer who's selling winter sports clothes and accessories. You might have one asset group dedicated to your ski jackets that include your texts, your image, and your videos related to ski jackets. And you can use the jackets page on your website as the final URL to set up audience signals.
You might then have a second asset group dedicated to ski goggles, which includes texts, images and videos related to ski goggles. You can use the ski goggles page as your final URL and also set up audience signals. So there are three types of asset groups. The first one is the generic asset group.
This is all of the creatives that are relevant to the products that are targeted within that campaign. We recommend that every campaign has this asset group. Then you have the option to add category specific or event specific asset groups and related creatives as you require. Once you have your assets in place, you can start deep diving the performance, and you can do this via the Asset Performance ratings report. Within the campaign, you can monitor creative and asset performance here. Once you've reached the maximum number of assets, you can start replacing assets with a low rating.
But remember, don't delete assets without replacing them because it's always better to have more assets to make sure that you are maximizing your performance. So I'll leave you with some top tips to consider when creating assets for your Performance Max campaign. Generally speaking, I'd highly recommend you to use as many ad assets as possible. More assets mean more potential ad formats to create for different channels and users.
Use at least five versions of text assets and at least five versions of image assets. Okay, so that's everything that we need to know about creative. The next pillar which agencies can use to steer performance is insights.
Many of you are probably familiar with the insights tab within Performance Max campaigns, but I do find it's generally underutilized by agencies and by clients. And it's actually a good way to understand what's going on in your campaigns. So I'd highly recommend getting familiar with this page. We'll deep dive into some of the key insights available in the coming slides. Okay, so let's have a look at some of the insights. We have everything from search trends to consumer interest insights, to persona insights.
Search trends help you identify emerging trends and growth opportunities. Then your Demand Forecast Insights help you to anticipate upcoming demand for the products you sell. You can review predicted start, peak and end dates to capture the surges in demand. Consumer interest Insights are going to then help you understand what consumers are interested in.
And finally we have Persona insights. So these are going to give you more detail and more information around the performance of your audiences. So the final three insights I'm going to cover today are Asset Insights, Auction Insights, and the new and improved Change history. So your asset insights will give you a bit more information about the assets that were used and details about the audiences that were paired with those assets. Secondly then, Auction Insights are going to help you understand a bit more about what's going on in the auction with your competitors and high performance is impacted as a result.
And last but not least, the change history. I think this is probably my favourite insight if I can say so. We have improved the change history within Performance Max and we're now able to give you some insights that highlight significant shifts, explain the root cause, and provide recommendations to address any issues. Pillar number three for agencies to steer performance is audiences. I would highly recommend adding audiences to your Performance Max campaign to influence automation by directly indicating which audiences are relevant to their business and have the highest likelihood of conversion.
The two ways in which you can leverage audiences within your performance Max campaigns to steer performance are number one, conversion value rules, and secondly audience signals. Okay, so last but by no means least, the final pillar that we're looking at today is measurement and first party data integration. Investment in first party data is really important. Enabling first party data tells Google Ads what your highest value sales look like, and gives insights into what works to attract those high value customers. In order to achieve higher value outcomes, which, for many of you, will be more sales or more profitable sales from your investment.
You need to teach the algorithm how to determine which interactions are driving actual economic value for your business. So that is all from me. Thank you for your time today. Next, you'll be hearing from my colleague Derek, who will bring you up to speed on everything you need to know on omnichannel strategy. Bye for now. Have a great day. Hello everyone and welcome to the omnichannel solutions section of the Ads Academy Program.
My name is Derick and I'm the EMEA Product Lead for omnichannel solutions. Today we'll go through driving value and growth across offline solutions. This section, we'll discuss how to measure offline metrics, connect the digital presence to store visits, and to understand the value of a store visits before going into bidding.
With omnichannel solutions being a fairly complex topic, this will be a brief, whistle-stop tour. Let's dive right in. When aiming to grow in-store sales, it's crucial to measure value through offline actions. Local actions activates reporting of location-specific clicks that occurred during or after a Google Ad click. These are available as default in the all conversions column in Google Ads. Local action bidding includes phone calls, driving directions, orders, menu visits, other engagements and website visits, to name a few.
But it's worth noting that local action bidding is only for Performance Max, and not for standalone search, shopping, display or YouTube. That's where store visits come in. Store visits is our simplest and most scalable form of offline measurement. Store visit information can be used across multiple channels to understand omnichannel performance throughout the funnel. When users click on your ads and they then visit your store, we attribute the store visits back to the ad click.
Simple as that. We do this in a private safe manner as we only report an aggregate and anonymized data to our advertisers. So store visit data cannot be tied back to individual a clicks impressions or people.
So if you've ever wondered how store visits are measured, this is the right place. We use 4 key pillars of store visit measurement technology. Starting with mapping technology.
We rely on the same technology that powers Google Maps to capture the dimensions and geometry of over 200 million buildings around the world. Data from Google Earth and Google Street View helps us to get the most up-to-date external views of where the buildings begin and where they end. And then measuring Wi-Fi strength also allow us to differentiate between different stores and their boundaries.
We then use server verification. So when our systems detect that potential visits have occurred, we add an extra layer of verification by surveying selected users about their store visits. Using a panel of over 5 million users in Google Opinion rewards program, we directly select people and which locations are visited, see how this checks out against our predictions, and then use this data to fine tune our machine learning models. From price perspective, as we've covered before, store visits are calculated based on an aggregate and anonymized data from our users who opt in to activate location history. This helps them get the most out of Google Maps, Google photos, and so much more. We do uphold the strict, conservative confidence thresholds for our data.
So you only see a report on store visits when we're sure the data meets the highest standard accuracy. Store visits data cannot be tied back to individual ad clicks, viewable impressions, or people. And then lastly, advanced machine learning.
We use deep learning models to train large datasets to prioritize which location signals are the most predictive of true visits. This helps us eliminate the noise from an activity that may look like a store visit, but it isn't. Like someone passing through to get to the parking lot or an employee staying there for hours. This is particularly helpful for measuring store visits in store buildings or shopping malls.
So once we have store visit measurement set up, this way, we need to determine the value of a store visit to the business. Understanding the value of the store visit can be as simple as following this formula. Your in-store conversion rate times the in-store average order value.
So for example, if one in four people purchase when they are in the store and they spend on average $400 when they do visit the store, this comes down to 25% of the 400, which brings us to an average order value of 100 dollars. The important thing to remember is not to worry about calculating the absolute perfect store visit value. Just set a value, start with and optimize from there. Getting started with measuring store visits at a reasonable value is more important than waiting for perfection. Now that we've discussed offline measurement through store visits or local actions, we can finally jump into omnichannel bidding. To make this this section easy to digest, let's consider three factors as you set up your store visits bidding, including a bid strategy, what your target will be and your offline conversion value, either through store visits or store sales for customers that are eligible for dynamic offline conversion tracking.
So let's explore bid strategies. Here, you may decide if you want to optimize towards ROAS or maximize conversion value within a set budget. Before we dive into specific strategies, let's start from the beginning and truly understand the power of AI. In other words, our auction powered technology through smart bidding. The benefits include advanced machine learning that drives informed, accurate bidding decisions, unparalleled optimization frequency and precision that adjust bids for every auction. And a rich set of auction-time signals used to capture contextual relevance for every unique search.
In fact, measurement of the median uplift in conversion value from when moving from target CPA to target ROAS is 14%. So it really speaks for itself. And value bidding will ultimately give you better picture of your customer. We are also giving Google Ads a clear view of how much each customer is worth through your business. So for example, when driving online sales, you can maximize your products that drive the most profit and set a different value for each.
For offline sales, you can set an average value for a store visit and let smart bidding calculate the perfect bid across all channels, whether that's online or in store. When choosing your bid strategy in your campaigns, align this to your goals, whether that would be getting as much value as possible for a set budget or to drive revenue at a particular ROI. But as a reminder, a dynamic market can affect different businesses in different ways and may cause unprecedented change or unpredictability.
With these changes, some businesses are having to manage twice the complexity without the capacity. especially when it comes to media planning and strategy. So by opting into automation, for example, smart bidding with store visits/sales, this'll allow dynamic adjustments to favour signals most predictive of a conversion. Next, we'll explore target setting. Your target is the key to success to your bid strategy.
Google has the best in class machine learning technology, but that technology is only as good as the data you feed it. Setting your target to optimize towards the full value of online and in-store sales allows our technology to maximize your local sales. Target settings should not be a "set and forget" exercise.
You should set your targets to align dynamic marketing goals. It's okay if the target changes are larger or more frequent than usual, especially in dynamic markets. This won't negatively impact the bid strategies as our technology acts quickly to these target changes. If you're focused on capturing holiday demand, you should lower your target ROAS. Or if you're focused on efficiencies and happy with lower volume, then you should increase your target ROAS. If you're unsure where to start, you can just use your current target or choose maximize conversion value for the best strategy as a start.
So as a reminder for accurate measurement, it's key to optimize towards the full value of online and offline sales across channels. We don't recommend having a siloed approach to KPIs for target setting on online and offline as the fluidity between the two for shoppers is even more prominent now more than ever. So when it comes to calculating your targets, to calculate those targets in line with your historic data as a starting point when you add store visits/store sales to existing biddable conversions. So selecting a period of roughly 30 days outside of your current store visit attribution window will be key to ensure that your data is accurate.
Targets are the best lever to control bids and spend or optimize towards a new ROI goal. We have planning tools and simulators to help you target and value scenario planning. It's worth noting though that while store visits data is not broken out separately in these tools, they are still a great resource to help you adjust your targets with evolving goals and demand. And then lastly, let's explore offline value.
This is where you will consider what a store visit is worth to your business or use store sales to measure the value of an offline sale. Higher value will weigh more towards store visits over other conversions. It's critical that you incorporate offline metrics into KPIs for your digital budget to get a better reflection of total return on ad spend across the channels. Focusing on online conversions only doesn't paint the whole picture of the impact your team drives when it comes to digital. So consider what a store visit is worth your business or use a store sale to measure value of offline sales dynamically and optimize towards success across channels.
To recap on the value, understanding the value of offline action for your business is crucial. As we mentioned previously, you can use this formula on screen to estimate offline revenue. Here's a summary of store visit conversions estimating offline revenue, as well as store sale conversions that allows you to measure offline revenue dynamically or if you're eligible for the solution. Right, now that we have our value next be to update the conversion settings in Google Ads with that store visit value that we've calculated.
So in the ads UI, click on the tools and settings icon in the upper right-hand corner. Next, navigate to conversions, which is under measurement and then click store visits. Click edit settings and then add that value under the value section and click save. Once you've selected your store visit value, you should monitor your store visits regularly, using the per store report.
Stay on top of your active campaign performance. Knowing how each of your locations are performing is more important than ever in this dynamic market. Our reporting allows you to segment your visit data per store so you can make informed decisions based on real data. And then when it comes to fine tuning, if the products or services have a shorter purchase cycle, consider shortening the conversion window to seven days. Shorter windows allow our bidding systems to be more reactive and changes in recent behaviour in order to validate results and act upon that more quickly.
So as a recap on omni bidding, because we covered a lot, for many retailers, the focus is on meeting consumers where they feel most comfortable to shop. This may result in conversion volumes fluctuating across online and offline. Omnichannel bidding adapts to these changes in consumer behaviour to ensure that customers see your message regardless of where they purchase or where they visit. So use smart bidding optimizing towards other store visits or store sales to stay flexible and optimize across online and offline goals. Update your targets to reflect your omnichannel goals and balance between efficiencies and conversion volume. And then lastly, use our performance planning tools such as performance planner and bid simulator to optimize your budgets and targets in this dynamic market.
So now that we've covered the value of omni bidding, now let's focus on driving more people to store if we're looking to grow even further. In this case, you can harness the best of Google advertising with Performance Max or store goals, a new campaign type that uses automation to drive more traffic to your storefront. When people are ready and willing and even comfortable to go into stores, they want to shop locally and near them, performance Max helps you to reach them and engage with those shoppers.
This campaign type optimizes to reach consumers who are most likely to visit your business. It's triggered by their user intent distance to your business, local search queries and devices. So, there's a lightweight implementation requirement using standard assets to allow for a high degree of creative customization to communicate offerings, business hours, safety measures, and so much more.
Performance Max for store goals make it possible for you to reach customers throughout their consumer journey. Each format is dynamically designed to drive the best possible performance for each platform. Your customer may look at content on web or in apps later on in the journey.
If the site is part of the Google Display network, which is the largest in the world that has access to roughly 90% of the global internet population, then this creative will show. And it'll look like what you see on the screen right now. Perhaps the customer then opens Google Maps on their way to your location. They will now be able to see your "Open" feature nearby based on their previous searches.
And even from a search perspective, when someone's searching for a coffee shop near them this would look at a different inventory that is usually used in additional search ads and send it around local search results. You don't need to pick which keywords to show on. We do that for you based on what we know drives high intent and will get people to your location.
As we said, all of these creatives links to your business profile. The business profile can also be used to promote specific products through the creative as well as you can see on the screen of the example showed. So for example, your customer, may be watching videos on YouTube.
We'll show the ad if we believe that they are likely to visit a store like yours. It's your video creative that you've added to the campaign with a link to the stores that are nearby to them. This campaign type can be used across a variety of business types drive several use cases, such as retail, restaurants, auto and telco. For example, if you're a retailer and you want to promote footfall to all of your locations, you're in key moments, or promote specific subset of locations such as reopenings, top stores, outlets, or even just to showcase product specific promotions, this solution is for you.
In terms of customization, you can continue to reflect your holistic marketing message with highly customizable creatives. This will enable you to highlight the changes you have made to your services such as curbside pickups, shop by appointment, safety measures you're taking within the store. Even just promote changes in business hours.
All of these features will help adapt in this new world, allowing customers to get the security they seek, to be confident in going to your physical store locations. And we are constantly driving innovation in the local campaign space, including a product showcasing feature in new services. This product ads feature can help advertisers to provide customers with information they need to visit their stores, including product availability, pricing and promotions to entice potential customers to visit the stores in person. This is available on the new services such as Display, Business Profile and YouTube. Additionally, product ads provided great and safe alternative to paper circulars, pamphlets and flyers. A trend that's accelerated during the pandemic, and even continues to this day.
It can be used as complement to offline media and we see many advertisers using them to promote weekly products both offline and online. Similar to circulars, advertisers have the control and flexibility to select which products will appear within this format. So you can create local product feed by uploading data from your top products, for example. So let's say your 5 to 10 items with description, photo, and the price.
And Google will automatically create ads showcasing these products for each targeted location on display and business profile pages. And with that then I'll be handing over to Denis who'll take us through product data excellence. So stay tuned and thank you.
Hi everyone. Today I would like to talk about Product Data Excellence and why it's a key asset to drive Return on Investment. For those of you who already know lots of this information consider this a thorough checklist to refresh your knowledge about the best practices and to make sure you keep the big picture in mind all the time when working with retailers. For others, these content might be completely new. I'll try to do my best to give you an overview of what should be a high priority and how to think about Product Data Excellence using a simple framework. And I know how difficult it is to keep in mind all the details working with multiple accounts, clients and countries.
Review this presentation from time to time to make sure you don't miss anything. My name is Denis. I'm a Regional Product Lead for Commerce Products in EMEA.
And today I'll walk you through four sections. First we'll briefly touch base on why product data is important. Then we'll review how to maximize visibility of your products and amount of clicks and conversions. And finally, how to achieve your business goals.
Let's start with a big why questions. Why is product data critical? Product data powers different level of things. Small and big things that users care about and of course, it powers the performance of your campaigns. Our AI is only as powerful as data you provide it with. By small things I mean everything that helps to differentiate a product. Product ratings, promotions that highlight special offers shipping data, sale price badge and returns data.
As you can see, all these small things influence the click through and conversion rates. Big things include new visual forms that we develop. The shopping experience on Search is evolving and becoming more visual.
Visual content inspires users to discover new styles brands, categories and products. Google is a great place to get this experience. The richer the product data the more places your products can appear on.
So, investing into product data you improve your current performance and ensure your future success. And of course, your performance depends on the quality of the data you feed to the system. As the research conducted by Boston Consulting Group it shows that the combination of really good first-party data can drive twofold increase in revenue. And product data plays a role in it as it the first-party data a unique data that only you or your clients have. In my own experience I see how our clients grow adopting Digital Marketing Maturity Framework and powering AI with rich product data. The first part of the framework I will keep in mind maximize visibility and triggering.
I love number three, so I offer you a three step approach checklist. Upload products, get basic quality and check Diagnostics Tab. As you can see, visibility is all about basic quality and just presence on Google. When uploading the product data, make sure that you upload the key attributes. Price GTIN, title image, description and shipping data. These are the ones that matter the most.
Without these attributes your product won't be served on Google or their performance will be very limited. Out of all those attributes which one is the most important? GTIN or Global Trade Item Number. These very attributes done correctly can drastically increase visibility and clicks. Without GTIN, you'll get limited performance and won't see a number of reports.
If you don't have GTIN because maybe your products are unique there is an alternative MPN or Manufacturer Product Number. And we rarely have studies about such basics as GTIN but this one from Indonesia is unique. Blibli is basically a giant store like Amazon. One of the biggest players in the region. For some reason they just hadn't had GTIN data. After fixing the basic attribute our system picked up this additional data which resulted in 23% return increase and a huge increase in clickthrough rates.
So, fixing the basics, you ensure the visibility of your product and triggering. Images, it's another component of basic quality. As I mentioned, the shopping experience in Search is evolving and becoming more visual. GTIN is for machine image is for people, for consumers.
It's highly important to do it right. On the left you can see what's the best practice and on the right, what we do not recommend doing. Let me pick an example.
Take a look at one in the red frame, a cap. We do not recommend using any promotional text or overlays in your images. Such images won't be approved and moreover, we have a specific tool to run promotion so, I encourage you to use that one instead. To sum it up, clean image on the wide background is the best you can do for your business. But what if you would like to show lifestyle images with some real models in real life? There is a certain use case for it.
Use relevant for this attribute called lifestyle image. At the moment, it powers discovery campaigns only. And I know that many specialists working with data miss this attribute. But it's a great step towards rich data and creatives.
And one more that can help you to ensure basic quality is shipping data. It's a must have attribute in most countries. Shipping data can be set up at the account or more granular individual item level. In addition to shipping cost we recommend uploading handling and transit time attributes. It makes you eligible for more shipping speed-based annotations. All studies I've seen show that shipping is one of those top attributes consumers pay attention to.
Of course, the first one was price. But the second or third was shipping. And finally to maximize visibility and triggering check that all products are live using the diagnostics and Opportunity Tabs in Merchant Centre. These two can help you to build a pipeline of things that should be fixed.
Opportunities Tab is a great tool to not only know what to fix but also, to see what's the impact it may have. If you work in agency it's a nice way to quantify the impact of your work. And a little bonus to protect your accounts from an expected event I recommend enabling Automatic Improvements and product protection features. The first one helps to timely extract data from your website and the second one will notify you if something is wrong with the product. With Automatic Improvements we can call your website to match the price availability and condition of a product. The same is available for images too.
We fix image issues to increase their approval rate. Well, that's it about visibility and triggering. Just to quickly recap the first part of this framework upload products ensure basic quality and check the Diagnostics and Opportunity Tabs. Moving on to the second part, clicks and conversions. Now customers can see your products.
Customers love your products and images. They see prices and shipping information but they don't click much. Let's fix it.
Three steps to increase the number of clicks. Increase coverage using many destinations available for a business. Optimize product data.
I'll briefly touch top tips on how to optimize data. And differentiate your offers. In Merchant Centre, multiple destinations are available at your disposal. Free Listings Shopping Ads and Display Remarketing and Local Inventory Ads for those businesses that are presented offline. Activate all relevant destinations and formats. Local Inventory Ads is a special interesting one for those who do sell offline.
It's a great format to provide users with a smooth transition between online and offline. With Local Inventory Ads customers can see if a product is available in store and how far the store is. Check out Programs Tab in Merchant Centre to see the full range of available programs and formats. Another way to increase coverage is to export products. Multi-country targeting helps to easily export to new countries.
And I know how it's not A very easy task for a business to expand and export to new countries to build a new logistics channel. But these little tricks help to at least find the audience find those incremental clicks abroad. And English-speaking countries is a great example here on the slide. Let's jump to optimization. Plenty of resources exist about optimization about how you can optimize product data on Google. And I would like to just pick one and highlight that one in this presentation.
To maximize clicks and conversions I recommend first focus on the title. And follow some best practices. Remember, these data are something that users read and click on.
It should be relevant for users and include relevant keywords. And you hit both goals here. Users see the right information and our machine reads these to define relevancy. For specific product categories we have specific title structures. Example, Maybelline lipstick the ideal title would look like this on the slide And of course, it's not set in stone and the title can change over time to adapt to commonly searched specific keywords.
Keep in mind that each product should have a unique title. Important information that should go first. Try to list all the relevant information and keywords in the title. An easy way to optimize titles and not only them is to use Feed Rules. It's a tool inside M
2023-05-18 01:50
