It’s finally here (thank you for your patience over the holidays!) the monster recap post of our chats with Google Ads’ liaison Ginny Marvin from December 2023! You all asked and Ginny answered:
Are there any more insights as to extra brand safety controls for Pmax on the horizon?
As you probably saw, there’s a temporary option to request to opt out of Search Partner Network at the account level through March 1st. I don’t have any updates to share at this time, but this is temporary as the teams evaluate additional controls to put in place.
As an FYI, looking at data from October, over 90% of SPN impressions were from Top 100 sites and apps, including YouTube. Programmable Search Engine (ProSE) sites accounted for less than 1% of SPN impressions.
- Here’s a rundown of brand safety controls for PMax that I shared in August
- An article I wrote this month for SEJ on controls in PMax
What are their objectives with the amount of campaign types that there are on Google Ads? Are they going to slowly get rid of different campaign types until the only option available left will be AI-driven campaigns like Pmax?
I don’t have anything to share on that but you’ll continue to see investments and updates in different campaign types, including Search, in the coming year.
Does Google have plans to integrate any of its recent AI into the Ads platform for analysis? Does Google currently have any best practices for using these or similar tools for large Google Ads datasets?
I don’t have anything new to add here. One current area to highlight, though, is predictive analytics. Demand forecasts on the Insights page in Google Ads, for example, can show products and services relevant to your business where demand is forecast to start increasing significantly within the next 180 days. And in GA4, there are predictive metrics and audiences which can identify likely purchasers or those at risk of churning and share with linked Google Ads accounts for advertising.
With AI impact on ad creatives, targeting and other controls, does Google actually even care about businesses in sensitive/regulated verticals that have very specific needs and requirements?
Yes, Google absolutely cares and considers the needs of businesses in these types of verticals. That’s why we ensure there are controls like pinning, for example.
Is a PMax planning tool in our future? I get asked how much headroom there is and that’s how we can plan budgets. I can plan TV & YouTube and then give PMax leftovers, but I don’t think that’s what the product team is into.
Performance Planner now supports Performance Max campaigns. You can find the specific eligibility requirements here.
Impression Share for PMax – Will we ever see this rolled out? Click share is nice, but IS would help determine if there is more room to push into search/shopping inventory within this campaign type.
What is Google’s rationale for not showing network-level breakdowns in PMax? Even with no advertiser controls, this is a big gap in reporting and understanding performance.
PMax reporting is tailored to focus on how the campaign is performing against the goals set by the advertiser. We do not provide raw inventory level data because we’ve observed that it may not always accurately represent performance. That said, we are committed to providing more insights & actionable recommendations over time and I will continue to pass this feedback along to the team.
Why is there a resistance from google to provide easy negative keyword application to Performance Max when its been clearly shown that there is potential for massive wasted spend for small business.
PMax is focused on performance goals – e.g. finding users who are likely to convert / deliver high conversion value based on your selected bidding strategy.
I know there’s some desire to have traditional campaign-level negative keyword or search theme exclusions and will share this with the team, but I’ve also recently written about many of the levers available in Performance Max to help steer the AI to maximize performance.
A few examples:
- Account level negatives, intended for brand suitability/safety concerns, rolled out early this year.
- Brand Exclusions were introduced this summer to exclude branded traffic (including misspellings and foreign scripts of the brand names) from PMax campaigns, as a result of advertiser feedback.
- Additionally, page feeds are now supported in PMax campaigns to provide more control over eligible landing pages.
What tactics do you recommend for new small businesses (SMBs) beginning to advertise on Google? Curious because it’s getting harder and harder to hit conversion thresholds for smart bidding and budget wise they typically can’t afford the volume play to ramp up.
It bears repeating that advertisers of all sizes should be focused on ensuring they have a measurement plan in place that doesn’t rely on third-party cookies and individual identifiers. From there:
- Consider using higher-funnel conversion actions to increase biddable conversion volumes. Also consider simplifying campaign structures, and shared budgets & portfolio strategies to automatically allocate bids and budgets across campaigns.
- There is no conversion volume requirement for most Smart Bidding strategies (for tROAS, you should have at least 15 conversions in the last 30 days that are passing conversion values through Google Ads).
- If you have set budget limits, use Max Conversions or Max Conversion Value.
- If you have specific targets, CPA or ROAS targets can be used like manual bids (i.e. adjusting your targets will change the bids Smart Bidding uses in the auction) while still providing real-time bidding flexibility across different queries. You should have uncapped budgets when using tCPA or tROAS.
Google seems to be pushing Maximize Conversion Value and tROAS. I’d love to utilize this, but in Lead Gen it’s not always practical to assign a value to the conversion action – especially when something like a form fill or phone call can lead to services worth anywhere from $20-several thousand dollars. I think it would be much more practical to set values at the ad group or campaign level. Does Google have any plans to change how conversion values can be assigned? If not, can you recommend a workaround?
Currently, we recommend structuring by value targets at the campaign-level. So in this example, I’d advise separating these by campaign rather than by ad group. Target ROAS is also available for portfolio strategies if you have similar targets across multiple campaigns.
You might also find this article helpful.
Any insight as to why Google pushes so much harder for tROAS over tCPA? The latter is a vital cost-controlling lever, and there are fears it will fade away over time, given the current trajectory of Google’s prioritization.
Target ROAS can enable you to prioritize and optimize when you have goals such as sales revenue, LTV, profit margins and are able to feed that data back into Google Ads. You can also continue to use Target CPA.
Enhance ROAS bidding – Are there any plans to allow COGS (cost of goods sold) to feed back into the calculation from a given CMS?
Using conversions with cart data allows you to bring data about the products sold for each transaction. Conversions with cart data is currently supported for PMax, Shopping, and Search campaigns linked to a Google Merchant Center account. More details here.
Additionally, here’s more information on how GA4 conversions are imported into Ads, and why you may see data discrepancies between the two.
Have they considered bringing back expanded text ads? In most cases, they performed the same or better than RSAs.
There are no plans to move away from RSA. One thing to note when looking at ETA vs. RSA performance is to resist comparing metrics (ex. CTR) between those ads in the same ad group. RSA is able to automatically show different combinations depending on the search query. Particularly when used with Smart Bidding, RSAs have the flexibility to serve combinations tailored to specific user context.
You might find this guide to RSAs useful. It covers the steps that occur before RSA combinations advance in the auction, Ad Strength, metrics to evaluate success and more.
Why can we NOT see data on what Automated Assets, photo variations and new asset/extension formats are doing to our ad CTR’s? Google seems to testing different variations in basically EVERYTHING these days and im curious as to why we are NOT shown this data and how it effects our ads. To me this has nothing to do with privacy but more towards the push to full automation and less transparency. It also has implications for smaller advertisers. So I am wondering why this data is not given to advertisers?
Correct, CTR reporting isn’t related to privacy needs. It used to be that the ads we created were served to every user, regardless of context. Now, Google’s ad systems can take a host of signals and performance history into account when determining which combination of assets to serve to a user. With RSAs, metrics like clickthrough rate and conversion rate are not always helpful indicators of your performance.
This is also why we don’t recommend using the combinations report to create “static” versions of ads – they can’t be guaranteed to perform as well because RSAs look at a number of real time signals to maximize performance. Additionally, you can also now see sitelinks, callouts, and images in combinations reports.
You can also use Ad variations to test RSA messaging.
And the guide to RSAs (also linked above) is a helpful resource.
Does Google Ads have some way to test Assets (didn’t find much via search)? Has it been useful? Do you know if they are building it for Demand Gen or PMax? Like A/B test different types of assets: videos, images, etc?
You can use Ad variations to test RSA messaging.
With Demand Gen experiments, you can test the impact of your video and image assets.
For PMax, the new asset group reporting shows conv, conv value, cost, etc at the asset group level. To increase conversions for an asset group, we recommend focusing on adding more assets. To increase conversions for your campaign overall, consider whether it makes sense to add more asset groups. Stay tuned for updates on additional asset testing capabilities in PMax.
How do we stop search queries from matching to our ads in different languages? Even when we are ONLY targeting english languages queries and have ONLY english language selected in settings in our search campaigns? Its becoming super annoying, would love to know if there is a way to stop them.
Ads are eligible for queries where the keywords match and Google believes that the user understands at least one targeted language.
So on the Search network, signals like the query language, user settings, and others are used to determine which languages the user knows. As an example, if a user has their mobile browser set to Spanish and their activity on Google strongly suggests they also speak English, they’d be eligible to see ads that target either English or Spanish – when the keywords match.
Can she explain how this search term [mr carve] is a close variant of the phrase “fiber laser machines” we’re targeting for a client. We’ve seen this happen more and more recently.
Thanks for sharing this example. I’ve passed it along to the team. While this type of matching technically falls within the definition of phrase match, we are taking advertiser feedback into account for future development of our product to reduce such undesirable matches.
When adding new keywords to an existing campaign that uses a Target CPA/ROAS bid strategy, I often see the new keywords get very little traffic. Is there a recommended way to add new keywords in this scenario? Would adding them in their own campaign initially help the account learn about those keywords before moving them into the existing campaign?
Some reasons I can think of why new keywords may not get a lot of traffic are 1. They’re closely related to existing keywords and essentially already covered. 2. The assets/ads haven’t been tailored to the new keyword so there’s a lack of relevance. 3. There is low search volume. This Help Center article has tips for diagnosing keyword performance.
If the keywords are on a new theme that you don’t already have covered, then you’ll likely want to create a new ad group (if the conv/conv value target of the product or service you’re advertising with these keywords is in line with the existing campaign), or create a new campaign if the conv/conv value target is different from your existing campaigns.
Google Ads used to be a goldmine for consumer insights, these new campaigns seem to lean more heavily into being ambiguous. Have the teams heard this feedback, and will we see a return to search queries instead of search “insights”?
Here’s an overview of all the insights that may be available in your account, including search terms and search trends insights.
Regarding search terms reporting, the privacy landscape has changed significantly – from both consumer expectation and regulatory perspectives. And with it, search terms reporting has also had to evolve – along with fundamental aspects of digital advertising more broadly.
The privacy thresholds in all search query reporting aren’t changing, but the search terms report remains available and search terms insights is designed to help provide more about the types of queries your ads show for. Updates to search terms insights this year include: more categories, API integration, custom date ranges & the ability to download.
There seem to be a lot more disapprovals of ads happening in the last 6 months. Has there been a change internally at Google Ads that would explain this? And how can advertisers get incorrect disapprovals fixed in a timely manner?
Ads policies are necessary to protect advertisers, publishers and consumers and to comply with local laws and regulations, but I know navigating them can be frustrating and there are occasions of false flags. I don’t know if there have actually been more disapprovals recently, however.
I typically recommend taking a close look at the Help Center pages on the respective policies and carefully reviewing your ads, site, etc. before appealing and then providing as much detail as possible when doing so.
Striking the right balance between blocking bad actors (of which there are many!) and reducing friction for good actors to advertise isn’t easy and something I know we need to keep working on.
I have no words to describe how bad Google Ads support is. Are there any plans to improve it … or is support being completely phased out altogether? To what extent has Bard/Google AI taken over support?
Support isn’t being phased out but changes are being made. There’ve long been challenges on this front, as everyone here is likely aware. I’ve talked about this before, but it was one of the areas I wanted to understand better when I joined. With the scope of inquiries, it’s not an easy solve. That said, I know there are real frustrations about the current state, including chat. I do think Support is an area where LLMs/Google AI will be able to make big strides in improving experiences. That’s not happening yet but work is underway, but stay tuned. [To add on a bit more here, the Help Center, specifically, is an area where we see AI playing a role.]
Are account and or campaign optimization scores a factor in keyword click costs?
No. Optimization score is a tool and isn’t a factor in the auction. Same goes for Ad Strength, and Quality Score. None of these tools are used in the auction or directly affect CPCs.
Optimization Score is an estimate of how well your account is set up to perform. Use it as a guide and evaluate if the associated recommendations make sense to apply to your account.
If Ginny was to start a new account from scratch (ecomm or lead gen) – how would she start?
Well, there’s a whole lot of “it depends” in this answer – goals, budget, market, resources, etc.
The first thing I’d do to get started, regardless though, is make sure my measurement was set up properly and fully. This is going to be absolutely critical in the coming year with the removal of third-party cookies for ads measurement. This site has a good overview of solutions available and steps to take.
Many people are seeing instances of overspending on a campaign by as much as double the daily budget. Aside from us using scripts/APIs or manually pausing an ad group/campaign, is there a setting within Google Ads that will keep Google from overspending the daily campaign budget?
This has been the case since 2017 (as a throwback, here’s how I covered the change at SEL back then). Campaigns can spend up to 2x the daily budget on a given day, but your monthly spend won’t exceed your spending limit (30.4 days X your avg. daily budget) in order to account for predicted fluctuations in search traffic – more here.
Why is the [Google Ads] certification going back to having a charge?
The free certification exams are still available and there are no changes to those on Skillshop.
The new professional certificates in Display, Search & video are currently in pilot. They focus on implementation & optimization skills, are optional, and don’t count toward the Partner program requirements.
They came out of requests from agencies who wanted higher exam integrity, security and more content focused on managing campaigns.
Why are PMax conversion values as reported in Google Ads so far off from what is reported as attributed to the campaign in GA4? (Google Ads is significantly higher than the GA4 number – have heard/seen this for multiple clients and from other people as well).
There are several possible drivers of discrepancies between Google Analytics and Google Ads. First, Performance Max campaigns are still a Google Ads campaign, and all considerations about Google Analytics and Google Ads apply to Performance Max campaigns too.
Another possibility for discrepancies is related to Performance Max reporting important metrics like engaged view conversions (EVCs) while Google Analytics by default does not (learn more about how to set up and manage conversion events in GA4).
When should we expect chrome to stop browser led cookie tracking?
Chrome announced last [month] that it will begin testing Tracking Protection, which restricts website access to third-party cookies in order to limit cross-site tracking, for 1% of Chrome users globally – starting Jan 4. More here.
And you can see the full timeline for testing and third-party cookie phase-out here.
We’ll have a lot of resources to help advertisers navigate the changes ahead. This privacy planner site is a good place to start.