7 PPC Automations That Save You From Burnout

Episode Description

Matthieu Tran-Van is an independent Google Ads specialist who provides search engine advertising services to clients with a lifetime managed ad spend of $350 million. In this session, he shares 7 automations that helped boost his clients’ results and saved him from burnout.

Episode Takeaways

Automation, when used intelligently, can help you scale the efficiency of your advertising campaigns. Matthieu leverages Optmyzr’s extensive suite of automations to work smarter and more strategically.

He explains sophisticated systems that he uses to shift budget from unprofitable products to star performers, make incremental adjustments to automated bidding, and more. Read on to understand how you can focus more time on strategy development while your campaigns run smoothly on autopilot.

1. Shopping campaign automations

Matthieu shares that shopping campaigns are central for e-commerce clients. But a lot of the success of your campaign depends on feed quality. This makes ongoing audits crucial. Automated merchant center feed can help you quickly identify issues across multiple accounts, while restructuring on the basis of performance buckets is a good way to improve efficiency.

Matthieu and his team use Optmyzr’s Shopping Feed Audits to audit the feed and identify issues like pricing errors, missing product identifiers, or unoptimized descriptions.

“Everything starts with a great audit of their feeds because every single time when we deal with shopping, it all starts with the feed. You cannot do any good shopping campaigns or PMax campaigns if you have a crappy feed.

And here what I’ve love in Optmyzr is basically this one-click merchant center feed audit that you can run. It is just super comprehensive” explains Mat.

2. Dynamic Search Ad (DSA) optimizations

Matthieu also explains that granular DSA campaigns focused on individual product detail pages is a great way to boost ad relevance and boosting conversions. He also suggests using product detail pages as dynamic targets to maximize alignment across search terms, ad copy, and landing page. This will increase your CTR, lead to better conversion rates, and lower your CPCs.

“You can even craft specific ads for each product using the attributes that you have in your feed. You can build those very granular DSAs again in less than one hour even for thousands and thousands of product thanks to Optmyzr” shares Mat.

3. Seasonality-based automations

If you work in seasonal industries, like, say, swimwear, then you know that you need to promote your products at the right time. Weather-based triggers can trigger your campaigns for seasonally relevant products.

Another tip Matthieushares is to use city-specific conditions to trigger your PMax campaigns. Additionally, combining geographic targeting with contextual data allows for more precise targeting and can be used for other types of seasonal automations as well.

“I decided to combine the power of PMax across all Google networks with the weather contextual data that actually is already embedded into Optmyzr so you don’t even have to pull from an API you own weather data” said Mat.

4. Automated bidding automations

When you’re using automated bidding strategies like target ROAS, it is important to not jump straight into it. Rather, take the time to train Google’s algorithm using small, incremental adjustments. This makes your bidding more effective.

Automation plays a huge role in enabling consistent, patient optimization— something we might rush as humans.

Matthieu runs this automation seamlessly using Optmyzr’s Rule Engine. He configured a strategy such that whenever the target ROAS is met, the rule nudges the goal higher. On the other hand, if its not met, the bids are lowered slightly. This way, you can avoid extreme swings in performance and maintain stability in your campaigns.

“What I wanted to achieve was to increase the return on a spend to a certain level. But by doing gradually. So basically every single time that the smart bidding has achieved your desired return on ad spend, then you increase the return on that spend by 1%. And when it does not meet the constraint, then you just decrease the CPC by one cent.” shared Mat.

5. Keyword automations

One other area where Matthieuleverages automations is to add profitable search terms as exact-match keywords. This, in turn, drives incremental growth.

You can set click thresholds to ensure that valid keywords are added. For instance, search terms that generate 100 clicks and have a low CPA get added to the ad group. Plus, automation can also be used to manage and exclude negative keywords so as not to affect campaign performance.

“I’ve decided to build a rule that automatically adds in exact match, the search terms that are actually profitable. They should have proven their profit profitability after at a certain threshold in terms of clicks. I said every time that a search term has more than 99 clicks and it’s profitable, then please add it to the ad group in exact match.

Every single time I run this strategy I always have this kind of fantastic growth because you make sure that all the search term that are super profitable are automatically added to your account” shares Mat.

6. PMax automations

Matthieu emphasizes that despite Performance Max campaigns’ “black box” nature, extracting and analyzing search term data is possible and valuable through automation. While Google provides search term data in the Insights tab, manually accessing this information across multiple accounts is impractical for agencies managing significant ad spend.

“Within Optmyzr, you already have a script that you can download and implement within your account or at the MCC level and automatically you can schedule it so it will automatically extract PMax search term data that your campaigns are giving you and add it to a spreadsheet.” said Mat.

7. Responsive Search Ad automations

You can also evaluate RSA performance beyond the insights Google is able to give you with the help of a systematic framework. Matthieu combines multiple data points to identify underperforming ads that require optimization.

“I’ve developed my own checklist to know if I should improve or optimize responsive search ads. I basically look at the volume of impressions and I also look at the click through rate. The ad strength is also one of the components that I’m looking at.” explains Mat.

He also shares that in order to get a more holistic view of ad performance across the funnel, he looks at the specific ration of conversions divided by impressions. This approach helps identify ads that might have acceptable conversion rates but are underperforming in generating initial user interest, or conversely, ads that attract clicks but don’t lead to conversions.

The ultimate benefit of Mat’s RSA automation is redirecting time from identification to creative improvement.

“I can focus like 99% of my time just improving the ad for real. I don’t spend that much time finding which ad I’m going to change. Instead they come to me, I improve the ads, and then I can focus 100% of my work on improving my client’s performance” says Mat.

Tools & Resources

• Shopping Feed Audits
• Campaign Automator
• Rule Engine
• Report Designer

Episode Transcript

Frederick Vallaeys: Hello and welcome to another episode of Automation Layering Master Class, where we teach you how to get the most out of Optmyzr using automation techniques. To do this, we always bring in the people who actually use Optmyzr and work with clients to make them more successful.

Today we have Matthieu Tran-Van from France, who is managing a significant amount of spend and is able to do that much better and more efficiently using Optmyzr. But it’s really not just Optmyzr—it’s the fact that he was at Google for over a decade. He figured out exactly what strategies were working for clients. By the way, he’s written a book about that, so check out his book.

Even if you have all the best strategies in the world, sometimes you realize that you need a tool, you need a little bit of technology to help you actually execute on all of that, and that’s where Optmyzr came in. So we’re very fortunate today to have Matthieu joining us. Let’s get started to learn about automation layering from Mat.

Frederick Vallaeys: Mat, thank you for joining us today.

Mat: Hello, Fred. Thank you very much for having me today. Bonjour!

Frederick Vallaeys: Bonjour! Where in France are you calling us from?

Mat: I used to live for almost 10 years in Paris, but I’m back now to where I was born in the south of France, in Toulouse in the southwest of France to be precise. It’s a great pleasure to be here with you today on the podcast.

I’m a big fan of everything that you have done, not only Optmyzr but also your book and your past at Google. I think we share the same background because it’s true that I’ve been spending 10 years of my life working at Google with global clients, especially handling the Performance Marketing topics for global advertisers. I had also the great pleasure to publish one book that is called “The Google Ads Strategist Handbook,” so please check it out. I will send you a copy, Frederick Vallaeys.

Frederick Vallaeys: I really look forward to it, and that’s amazing. Thank you so much for coming out here and sharing how you use Optmyzr, but also what you’ve learned in your 10 years at Google, obviously working with so many clients and being on the inside. Being on the inside much more recently than I have certainly gives you a fresh perspective, and I think people are going to learn a lot from our session today.

Besides that, you also lecture in some top Business Schools in France, you’re on the jury for SMX in Paris, so clearly very involved in the industry. So then tell us a little bit about after you left Google—you started consulting, right? How did that come about?

Mat: That was actually a total adventure. After 10 years at Google, I decided to leave Google, but not to launch my own business right away, or at least not on my own. I joined the board of a tech startup called Data Yad, which still exists. Here we built a SaaS tool that can dynamically change the landing pages for e-commerce advertisers. The goal was to improve the profitability of their shopping investment.

But after one year, I had a personal event—I had my third child—so I decided to totally flip around my life. I left Paris, I left my job at the startup, and I decided to go with my very first passion, which was actually Google Ads and SEA. Now I deal with a lot of clients across all verticals. In total, I manage 25 million euros—here we are not talking in dollars, we are talking in euros—so it’s 25 million euros under management. I have clients across all verticals, being lead gen advertisers or e-commerce advertisers, and I have a great time working with them all.

Frederick Vallaeys: Great! Well, let’s jump right into it. Let’s start with those e-commerce clients that you have and a couple of the things you do for them. I think it all starts obviously with the structured data, the feed, so let’s talk a little bit about what you do with that.

Mat: Exactly. Shopping for e-commerce is just super strategic for most of my clients. It’s 80% of where they are putting their money, so making sure that you nail down your shopping campaigns or your Performance Max campaigns is very important for those kinds of advertisers. Guess what? Here, Optmyzr can just supercharge the kind of work that you are doing for your shopping advertisers, and that’s exactly what I’m going to try to share with you in this first part.

Everything starts with a great audit of their feeds because, come on Frederick Vallaeys, you know it—every single time when we deal with shopping, it all starts with the feed. You cannot do any good shopping campaigns or Performance Max campaigns if you have a crappy feed.

What I’ve loved with Optmyzr is basically this one-click Merchant Center feed audit that you can run. It’s just super comprehensive and very interesting because here you can even customize the type of attribute you want to scan for each advertiser. So you can have, in a snap of a finger, a full comprehensive view of all the attributes that are missing, maybe the titles that are not optimized, the descriptions that you can improve, and so on.

It can be super actionable and granular because within the audit itself, you can, in one click, download the exact IDs that are concerned by the analysis. It has totally changed the way I’m making recommendations when it comes to feed optimization to my clients. It’s just an amazing tool.

Frederick Vallaeys: That makes sense. I think probably when you’re auditing a new client or a new prospect, this is a great way for you to figure out how much improvement you will be able to do or how much the last agency has messed it up.

Then of course, once you start making promises—you say, “Hey, we’re gonna make sure you always have nice titles, the descriptions are in a certain way, there’s no disapproved products”—but now you actually have that commitment. You have a lot of clients, you have a lot of other stuff to do, so how do you stay on top of that? That’s where that audit thing comes in because it runs consistently and it will point out things where you’re maybe not living up to the promise that you made to the client. You can fix it before the client calls you out or goes to another agency and asks them for an audit. So it helps both with client acquisition and client retention.

Mat: True, true. Even from a maintenance perspective, I used to run this particular Merchant Center feed audit at least once a month for all my clients because e-commerce advertisers’ products come and go, and they are changing their items very often. So it’s pretty common that advertisers do not keep always a very good feed when they are changing their products. Sometimes the description is missing, an attribute is missing, or maybe they have changed the naming convention.

Here, the audits can help you spot those improvement areas in less than a minute.

Frederick Vallaeys: And then one thing that Optmyzr does as well is we help with the creation of supplemental feeds. Because one of the worst things is you identify, “Okay, they don’t have a great setup for this, like this title is broken.” Now you go and fix it, but then the next time their automated tool regenerates the feed, it’s probably going to be broken again. So it’s like this constant battle—it keeps breaking. But if you use a supplemental feed, you’re basically saying, “Listen, let them do whatever they do on their end, and then once the product feed comes to us, for that product ID, we know it’s broken so we need to tweak it in this manner.” It really makes it easier to work with your client without having to interrupt all of the other areas of their company that might be using that feed.

Mat: Definitely. Thanks to this first job that you can do on the audit and then implementing it easily, as you said, with the supplemental feed feature of Optmyzr—think about your custom labels, all the attributes that are missing like colors, materials, and things like this—you are ready to really build and maintain very powerful shopping campaigns for your clients.

Mat: What I’ve been also seeing is, once the work is done on the feed part, most of the clients are looking for something else than just your normal shopping campaign or Performance Max campaign that targets all the products.

For instance, I have in mind this advertiser of mine who has a huge catalog—around 18,000 different IDs—and they are changing every day, so it’s pretty massive. The challenge here is, once we fixed the feed and then ran some analysis that you can pull directly from the Google Ads UI, we saw that more than 60% of the client’s budget was spent on products with a return on ad spend below one. So basically, 60% of your money is going into products that are making you lose even more money. That was a real issue.

What I decided to do—and again, thanks to Optmyzr, that’s something that you can implement in less than one hour—was to restructure the entire shopping campaigns based on performance buckets. I defined a matrix where you have, on one hand, the volume as measured by the cost that you put behind each product ID, and the return on that spend.

So we came up with four different categories:

• The star products
• The ones that you can optimize
• The ones that make you bleed money because they’re not profitable
• The zombie products—the ones that you don’t know because they are never shown (and that happens a lot, especially for e-commerce advertisers that have very large catalogs)

By implementing this structure through Optmyzr, we’ve managed to go from only 27% of the budget spent on the very best products (the ones with the highest ROAS) to 53%. That was a massive win for this client because, in just a few weeks, focusing the investment on products that drive a lot of margins has been a game-changer for the economics of this specific advertiser.

Frederick Vallaeys: That’s brilliant. Obviously, the performance metrics keep changing, so something that used to be a zombie had no volume, then it starts getting some volume and maybe goes into the optimized bucket, and maybe it doesn’t work well—something happens in the market so it becomes a bleeder. Then you go and fix it, and then it becomes a star.

That’s the beauty here, right? You’re not just moving things once, but it’s constantly adapting to how things are performing. That might be driven by external market conditions that are not in your control, but at the very least, you’re then able to move things into the right budget bucket. So if something’s going to be bleeding, you’re not going to bleed out the whole budget for that campaign, and you’re going to shift your budget to whatever the stars happen to be that month. I love that implementation.

Mat: Exactly. Then for the same client who said, “After the win we had here on this specific account, I have a surprise for you—I don’t have only one account, I actually market my products across seven different countries, and now I want you to implement a new shopping strategy across all the countries.”

Here’s the kind of nightmare you can have as a PPC practitioner—seven countries, seven accounts, seven different Merchant Center feeds that you all need to audit. Now you know it’s easy to audit a feed thanks to Optmyzr, but you always ask yourself, “How am I going to manage that many shopping campaigns?”

Mat: I decided to go with a slightly different approach because the volume was not the same for all the markets. I decided to fully stop Performance Max everywhere across all the markets and go back to standard shopping campaigns.

One of the insights was that for brand searches, especially people searching for the brand, the CPCs with Performance Max were really high across all the markets. So we decided to go for each market with a structure that consists of three different standard shopping campaigns:

• One for the brand terms only—a campaign that triggers only when users search for the brand terms
• A second one that only has the best sellers across the 18,000 products—only the best sellers for that specific country
• A third one for the long tail of the catalog

Optmyzr helps greatly with the implementation of 20+ campaigns because, with the shopping management feature of the tool, you can build them in five minutes each. You just spend your time thinking on the strategy, and then with the tool, the execution becomes flawless and super productive.

But you still have one problem: how do you steer those campaigns? You can decide to manage individual bidding strategies for each country, for each bucket, or even each campaign. But I decided not to do this because the way my customer was looking at those additional countries was purely to drive incremental conversions and incremental revenues. Basically, we don’t care if one conversion comes from Spain or from Germany—we just want to have more conversions for a given return on ad spend.

What I decided to do is to build portfolio bid strategies at the MCC level. So every bucket of campaigns (for instance, all the brand shopping campaigns) is managed by only one strategy.

Some key benefits:

1. You get a lot of volume very rapidly. In two weeks, your smart bidding is already on top of things, it has a huge volume of data, and it can deliver the return on spend that you expect.
2. In terms of arbitrage and CPCs between markets, it’s super easy because now you let the smart bidding at the top of your MCC manage this complexity for you. If there are cheap conversions in Austria instead of Spain, the smart bidding can make those arbitrages in real time.

In less than two weeks, we saw not only conversions going up for this client, but also the return on ad spend is super well controlled. When you turn the return on ad spend bid up by 5%, 24 hours later the smart bidding is following you.

Frederick Vallaeys: That’s the beauty of the tool, right? If you said, “We have different goals for different campaigns or different goals in different markets,” then you could actually set it to not be on a portfolio strategy, or you could have multiple portfolios. It would be just as easy with Optmyzr to maintain it because we handle that complexity of all the different campaigns. I’ll talk about bid management efficiency a little bit later as well, but there are techniques that you can use to automate that at scale very nicely.

I love that, and obviously it’s great when the client comes to you and says, “Hey Mat, we got a little bit more work for you,” because things are working.

Frederick Vallaeys: The next thing that you ended up doing had something to do with the brand and non-brand structure. Let’s take a look at that.

Mat: One of the challenges when you want to split brand and non-brand in shopping campaigns is how you maintain your negative keywords. Even if you put different strategies (for brand shopping campaigns, you’ll put a higher return on ad spend, maybe add some CPC caps in your portfolio bid strategy to make sure you don’t pay more than a certain amount per click), you will see that every single day, thousands of non-brand search terms will trigger your brand campaign.

You don’t want this because what you want is to perfectly keep this campaign on the branded search term scope so you can fine-tune the efficiency at the maximum and get the exact ROAS that you want with the exact CPC that is worth each click.

Here again, I use Optmyzr to do the work for me because I’m a lazy guy. To be precise, I use the rule engine feature of the tool to set an automation that says, “For this specific campaign, please look every single day at all the search terms, and when you don’t see the brand, please exclude it at the ad group level in exact match.”

Every single day, thousands of keywords get excluded from your brand shopping campaign. What does that mean? After one week or two weeks, or even sometimes after a month depending on the power of the brand, you have a perfect brand shopping campaign that only triggers for branded shopping search terms.

Now you can decide to maybe show different products for this specific audience. You know exactly at which return on spend you can push this campaign to, and you know exactly at which CPC you can bid most efficiently by making sure that you don’t lose that much impression share but at the same time don’t pay more than you should for a branded term.

That was super helpful to combine the tools that Optmyzr offers on the shopping management campaign front with the rule engine that can help you stay on top of the hygiene of what you built.

Frederick Vallaeys: That’s another good reason to sometimes manage a standard shopping campaign as opposed to Performance Max, because standard shopping campaigns give you the ability to put in negative keywords and maintain that brand consistency.

Now, you could kind of do this with Google—you could probably call your rep if you have one and tell them to exclude these things—but if you came to them daily with hundreds of negative keywords, they’d probably not be super excited. This is a great implementation, and it speaks to your strategy of thinking about why we use Performance Max vs. standard shopping—there are pros and cons to each.

Mat: The great thing is that even for the strategy we discussed earlier with the star products, the optimized products, the bleeder products, the zombies—with Optmyzr, you can actually build exactly the same strategy but fully on Performance Max. It’s also an option.

In my case, I decided to go with standard shopping, but I’ve made exactly the same strategy for another advertiser where everything was built on Performance Max. So here you have all the benefits of Performance Max—you can add all the assets, images, videos, and so on.

That’s pretty neat that Optmyzr can handle both types of campaigns. It’s not only something you can do with standard shopping; you can also do it with Performance Max. It gives you room as the PPC strategist to really think about what is best for this client, and you don’t have to care that much about the execution because the tool can handle it.

Frederick Vallaeys: Thank you to my team because they work very hard to maintain that parity with Google. Basically, the philosophy we have is we are not going to tell you how you should manage your campaigns. We’re not going to come in and say, “Hey, Performance Max is not a thing, you shouldn’t do it,” because clearly it does work in certain instances, and it’s a big focus area for Google. We’re going to help you figure it out, and we’re going to help you scale regardless of what campaign type you choose to use.

Frederick Vallaeys: Let’s talk about shifting gears here a little bit—DSA campaigns and how you optimize those.

Mat: DSA campaigns, I think, might be the best product ever that Google has launched when it comes to advertising products. I love DSA campaigns.

There is this one strategy that I love doing for my e-commerce advertisers: building very granular DSA campaigns. When I say granular DSA campaigns, I approach it a little bit like a shopping standard campaign. What does that mean? It means that what I love to do is to build one ad group equals one product equals one dynamic target, and the dynamic target, of course, is the PDP (product detail page) of the product itself.

I love to do this because first, you can increase drastically the coverage of your dynamic search campaigns—you increase the coverage in terms of pages that are scanned and that you can redirect traffic to, way better than if you just keep the dynamic target as “all website.” It’s way more granular, and you will get more pages scanned.

But also in terms of inventory, suddenly, when you directly plug your DSA on the PDP pages, you see new search terms appearing in your DSA campaigns, and that’s great. The goal, of course, is not just to be more visible but to drive incremental conversion and incremental value.

For this specific advertiser who had a catalog of around 5,000 products, with this specific strategy, we increased the traffic by 20%, which translated into an increase of 52% in sales and 27% in conversion rates.

The secret sauce here is that, thanks to Optmyzr, you can even craft specific ads for each product using the attributes you have in your feed. Basically, when someone searches for a specific Nike pair of shoes in blue from a specific model, in your ad directly, you can use the exact product name with the price, maybe the sales price if it’s on promotion, etc.

You have a perfect alignment between the search term, the exact ad, and the landing page. So of course, when people are clicking on those ads, they are super ready to convert.

This strategy is very easy to implement thanks to Optmyzr, specifically with the Campaign Automator feature. This feature helps you build DSA campaigns or other types of campaigns based on any kind of feeds. You can take contextual data feeds, but here you can also directly leverage the Merchant Center feed of the account within Optmyzr.

By using the Merchant Center feed, you can use all the attributes of your feed in your ads super easily. You can say, “I want my dynamic target to be the attribute link, I want my description line to include the product name and the price,” etc. You can build those very granular DSAs in less than one hour, even for thousands of products, thanks to Optmyzr.

Frederick Vallaeys: That’s fantastic. The beauty here is, I think the underlying strategy is not rocket science—after 20+ years of Google, we understand that relevance matters. But now you’ve actually freed up your time, and you can say, “Should we include the product price? Should we include some brand attributes?” and you can very easily modify the template and next week run a different template to see if that helped boost your conversion rates.

One of the super interesting things that you just pointed out is, you would think if you give Google your generic website and say, “DSA this, figure out any page within this,” you’d get as many impressions as listing every page independently. But it sounds like by listing them independently and focusing Google on those PDP pages, it actually drove quite a bit more volume. I think that’s a brilliant insight and something everybody should learn from.

Mat: The relevancy is even better here because thanks to the ad that is super tailored for this PDP, when someone is searching for this product, you are one of the most relevant options with perfect alignment between your ad and your landing page.

In the past, I used to do this strategy, but it took me around one day to implement the entire strategy for my clients. Now, thanks to the Campaign Automator, it takes me like one hour. It gives you some time to go home and play with the three kids in the evening.

Frederick Vallaeys: The next one we should talk about is seasonality, so “maillot de bain” (swimsuits) in French. What did you find out?

Mat: Here it’s still this idea of leveraging different data feeds. We’ve talked about how we can leverage the Merchant Center feed with DSA campaigns, and here it’s more or less the same idea at the beginning, but in this case with Performance Max and with contextual data.

The advertiser is in the swimwear industry. As you can guess, swimwear is highly seasonal—people only buy swimwear between April to mid-July, and then it’s the end of the season. You cannot afford to lose some sales in this very high peak period.

The strategy I decided to go with for this advertiser was, on top of the always-on search campaigns and shopping campaigns, I wanted to put a boost with Performance Max.

First, I looked at the geo data in the account to map the cities that were presenting both the biggest volume of conversion and also the best return on ad spend. This analysis helped me scope the top 10 cities where it was really interesting for them to do a massive push.

The problem is that people only buy swimwear when the weather is pretty good. In rainy April, no one is thinking about buying a swimsuit, but when it’s sunny June or May, everybody is thinking about when they will be on the beach this summer.

That’s where the magic happens. I decided to combine the power of Performance Max across all Google networks with the weather contextual data that is already embedded into Optmyzr. You don’t even have to pull from an API your own weather data—Optmyzr is already doing the job for you on this one.

The strategy was: every single time that in one of our top 10 cities, the temperature is above 27 degrees, then automatically activate the Performance Max campaign or add the geo area in the Performance Max campaign to advertise on a specific audience signal. This audience signal was basically the persona of the brand, where we leveraged the customer list, the brand traffic, etc. We knew that this specific audience signal was really sharp for this brand.

Here you can truly blast from an advertising perspective across search, shopping, display, YouTube, and you can also think about different creative strategies. For instance, we decided to go more with new collection videos on YouTube and lifestyle ads on display because we knew that people were looking at the advertising on a very sunny day when it was super warm.

You can even play a little bit and do some testing with the creatives to see what performs best. But the idea was to leverage this contextual data to activate, on top of all always-on campaigns, a Performance Max campaign that can truly meet the moment when the user is in the exact precise moment of thinking about the swimsuit because it’s hot, warm, and sunny.

Frederick Vallaeys: I love it because when we started talking about scripts more than a decade ago, bidding by weather was sort of the classic example, and people continue to talk about that. But here’s a brilliant example—this is not rocket science, but people are going to buy bathing suits depending on the weather.

We make that super easy to test out because it’s a thesis, right? You have to test out if it actually makes a difference. As Matthieu is saying, some of these things to implement manually—you’d be spending your whole day, integrating with new APIs, it’d be super complicated. But now you can put your marketing hat on, talk to your client, figure out what seasonality and temperature patterns they’ve seen, and if that seems to be a factor, it’s very easy to set it up.

Then you set it up, and you get what Matthieu is talking about: thinking about audiences, shopping categories, and all of the other things that are really Google Ads-specific. All of this weather-based bidding and enabling of campaigns is handled for you automatically. A huge time-saver!

Mat: It’s a great addition in Optmyzr, to be honest. The beauty of it is that you can also add other contextual feeds. I’m thinking about traffic updates or financial markets updates. I have some clients in finance who might want to increase ad pressure on some campaigns when certain stocks are going down or up. Like when the NASDAQ is going up, I want to trigger or push or increase the bid on some specific campaigns or ad groups.

You can actually do it with the Campaign Automator—you can bring your own contextual feed of data and replicate the same kind of approach.

Frederick Vallaeys: Campaign Automator, Rule Engine—you can use all of those. That’s where I would recommend if you ever have a strategic idea and you are not quite sure how to implement it, there are so many ways you can go about it. Talk to your account manager at Optmyzr—everybody gets an account manager, so they can certainly point you in the right direction and help you figure out how to connect that.

In the case of weather, we have that data integrated, but in the case of what you’re saying, maybe traffic data, we don’t have that yet. We’ll just talk to you, figure out where to bring it in from, and how to connect it to all of those location-specific elements.

Frederick Vallaeys: Let’s talk about the next one, which is smart bidding. Obviously, one of the biggest automations that most advertisers do is automating their bidding. But then there’s this misconception, I think, that “Hey, you’ve turned on automated bidding from Google, my job is done, I’m not going to think about bidding anymore.”

Mat: (Evil laugh) Or is that like, “Oh my God, you’re so dumb for just stopping there!”

Frederick Vallaeys: That’s the opportunity. That was my second book, “Unlevel the Playing Field”—if you thought you were done bidding, here’s your edge. Mat, share what you do.

Mat: I think bidding actually became even more complex since smart bidding exists. Come on, Frederick Vallaeys, it was even easier sometimes just to put a given CPC according to your revenue per click—that was it. Now it’s a complex machinery.

One of the most powerful things you can do when it comes to your bidding is to train them and make them smoothly adapt themselves to your very own constraints. If your client wants a return on ad spend of 500%, you can try launching your campaign and just put 500%, but guess what? I think it’s not going to work, at least not at the beginning.

The idea is, when you want to drive more performance with smart bidding, you need to train and help the smart bidding little by little to go where you want it to go. It’s a job that you need to do on a regular basis, and everything that needs to be done on a regular basis, Optmyzr can really help you with.

For almost all my clients, but for this specific example, I created a rule on portfolio bid strategies (I’m a big fan of portfolio bid strategies because you can cap CPCs). What I wanted to achieve was to increase the return on ad spend to a certain level, but by doing it with little touches.

Basically, every single time that the smart bidding has achieved your desired return on ad spend, then you increase the return on ad spend by 1%. And when it does not meet the constraint, then you just decrease the CPC by one cent. So it’s very granular fine-tuning.

By doing it on a regular basis on autopilot—because Optmyzr and Rule Engine do it for you—you can achieve pretty good results. That’s what you see on the visual where not only does the smart bidding take the lift and follow your constraint very well (because you don’t just go from zero to 500%, you increase things by 1% each time, every week or every day), but it’s also going to exceed the expectation.

For this specific advertiser, my goal was to have a 500% ROAS. I started around 350% and went little by little, increasing the return on ad spend thanks to the Rule Engine. Guess what? The system has actually gone beyond our expectation and delivered more than 600% ROAS for this specific client.

What is interesting and what you don’t see in the visual is that the volume or the conversion value has not fallen down. The conversion value kept going up and up. Just by teaching little by little how you want the smart bidding to behave, not only can you achieve your profitability constraint, but you can also deliver the volume of conversion that your advertiser is expecting.

Frederick Vallaeys: I think that’s the beauty—what you just said is volume would be a big concern, but you just bake that into the Rule Engine strategy. Where you’re saying, “Look at the ROAS hitting a certain target,” you can have multiple conditions on that. So you could also say, “If my ROAS is being met and my volume is still above a certain threshold that I want, or it’s not decreasing by too much, then keep pushing that ROAS goal a couple percent more aggressively.”

I think the other beautiful thing that you’re doing here is, if you didn’t have automation, you’d probably make bigger pushes—you’d make 10% or 15% pushes, and that might cause more erratic changes. It might modulate a little bit more wildly, and you might more quickly think, “Hey, it’s not actually working, it’s not doing what it should do.” But by automating it and saying, “Listen, every week we’re just going to make a little push and we’re going to see if it’s still working,” you don’t have to worry about it. It’s a more precise, more granular method of getting where you need to be.

Mat: Yes, and the system has the patience that maybe you don’t have as a PPC practitioner. Here you put it on autopilot, and guess what? It works! It’s just a matter of being patient and at the beginning setting the right constraint and the right variation that you want.

It definitely works—it’s such a powerful strategy, and it also frees me up. I’m saving at least one day of work because when you have to do this exact strategy across 25-26 different accounts, it’s painful. Now with Optmyzr, you can automate everything and truly scale very granular optimization.

What is really amazing with Optmyzr is that it’s not just one rule that says, “I’m going to increase ROAS when it misses ROAS.” You can precisely define what the thresholds are on your end, you can precisely define the variation. It’s a tool that has been thought by PPC professionals for PPC professionals, so you can really put your very own custom strategy and automate and scale it. For this client, I put 3%, but for another client, I will put 5%, etc. It depends really on each client, but the mechanic of the strategy remains the same.

Frederick Vallaeys: Exactly. Some clients are more aggressive—they want to get there faster, so you push some of the limits a bit faster with them. Customization and personalization is the key here.

Frederick Vallaeys: Let’s talk about keywords. Is that still a thing these days?

Mat: Come on, keywords are still a huge thing, and I love keywords! I do a lot of keyword management, and again, thanks to Optmyzr, I’ve been able to reach a new level.

One of the very basic strategies that we all know as PPC practitioners is keyword expansion—adding new keywords very often from your search term reports that have been profitable or that have a reasonable or better CPA. That’s something you do on a regular basis as a PPC practitioner. You need to open your search term reports, download the search terms, rank them, and decide if it makes sense to add them.

With Optmyzr, because I’m a lazy guy, I’ve decided to build a rule that automatically adds, in exact match, the search terms that are actually profitable (and we know this for a fact) and that have proven their profitability after a certain threshold in terms of clicks.

What you see on the visual: I said every time that a search term has more than 99 clicks (so it gets to 100) and it’s profitable, then please add it to the ad group in exact match. You can also do the same for exclusion—when you know that the return on ad spend is not profitable after one click, please add it as a negative.

I use this strategy—you can use the same automation for your DSA campaigns. When a search term is not profitable, please exclude it automatically from my DSA campaign. You just build a shared negative keyword list at the account level and put all the non-profitable keywords in this list, and the job is done on autopilot.

I pushed this very far for one client, and the strategy ended up adding almost 500 new keywords based on performance. Of course, you spend more money because you increase and expand your keyword coverage, and the inventory expands, but because you only pick the best search terms—the ones that are super highly profitable (and you know they are profitable because they have had more than 100 clicks, so it’s proven performance)—even if your spend goes up, the CPA went down.

The conversions started to really scale—we multiplied by three the volume of conversions. Not only was the return on investment better, but also the volume of profits for this client, as you can see in the chart on the top right corner, is a hockey stick curve.

With that kind of chart, I think anyone who’s not hiring you is crazy! Every single time I run this strategy, I always have this kind of hockey stick growth. Why? Because you automatically are sure that all the search terms that are super profitable are automatically added to your account. So of course, you start to stack all the incremental conversions that you can get from those search terms.

At some point, three months, six months, one year after, the client’s business has totally changed. I even have some clients that told me, “Please, can you stop now because my team is full, we cannot ship more orders.” So we want to go smoother.

Frederick Vallaeys: That’s amazing. I love those stories. I remember in my early days at Google, we had some clients—I was working with all the Dutch speakers—and I’d get those same calls. It was a company that was building personal computers, and they called up and said, “I don’t have enough people, I don’t have enough components, I can’t keep up with the sales you’re generating. Take a break.” That’s sort of the best story you can hope for, right?

Mat: Definitely. It’s kind of an achievement, I think, for every PPC practitioner when your advertiser and your clients tell you this. It’s like you have won the award, you have won the competition, you cannot do better.

Frederick Vallaeys: Now, what about keywords or search terms in the context of Performance Max, in a place where we have a little bit less control? Is that still something you can work on?

Mat: Yes, of course. Thanks to a recent update on the Google side that now gives you the ability to download all the search terms that you can find under the “Insights” tab. To be honest, it’s pretty cool because you have all those category searches, and then you have all the details for the search terms, and you have the performance that goes with it in terms of clicks, conversions, conversion value, and so on.

But let’s face it, especially when you are a PPC practitioner that manages millions of dollars, it’s pretty painful to go into every single account, go into the Insights tab, and download the data every single time. So guess what? Optmyzr can do it for you through a script.

Within Optmyzr, you already have a script that you can download and implement as any script. You can implement it within your account at the account level or at the MCC level, and you can schedule it. It will automatically pull into a Google spreadsheet all the search terms that Performance Max has delivered on.

That’s a game-changer because you can do a lot of things. I love to do two things with this data:

1. Exclude more keywords from my shared negative list if, as Frederick Vallaeys said earlier, you have asked your Google rep to implement some negative keyword lists on your Performance Max campaign. You can keep enriching this negative keyword list when you see terms that aren’t super relevant or have very low performance.
2. Feed and increase the keyword coverage of your normal search campaigns. When Performance Max spots a search term that performs very well in terms of volume or return on ad spend, why don’t you add it to your normal search campaigns? Now I can leverage not only all my search terms coming from my search campaigns, but I can also see all the terms coming from Performance Max that often offer new territories. It’s true that Performance Max goes very far sometimes and reaches some audiences that you may not have thought about at the beginning, but it works! This kind of intent that was not on your radar, and suddenly people are converting out of this intent—now I can build a totally new ad group on this new intent, or maybe a campaign.

You have some synergies that you can play with, but all of those things are possible because on Monday morning, you have all your reports and Google spreadsheets filled with all the Performance Max search terms in one place. That’s magic, to be honest.

Frederick Vallaeys: I think that’s one of the things that’s going to give you an edge—not just sitting and waiting for the automation to handle everything for you, but seeing the automation as something that tests new areas. Once it finds something interesting, go back and actually think about a strategy: Should that have had different messaging? Should that have had different budget associated with it? Was there a seasonality to that new thing it found? There are so many questions you can ask to take that new finding to the next level.

Frederick Vallaeys: Speaking of new messaging, messaging these days is largely done through RSAs (responsive search ads). What do you do with responsive search ads to facilitate that?

Mat: Responsive search ads are always a challenge. It has always been a challenge for different topics. I think about testing, for instance. We never know: Should I pin my headlines or not? Does the ad strength score of Google worth something? Yes? No? We don’t know.

Basically, what I’ve been doing is I’ve developed my own checklist to know if I should improve or optimize a responsive search ad. I look at the volume of impressions, the click-through rate, and the ad strength is one of the components I’m looking at. But you also need to change your ads when you have more difficult problems like a disapproved ad or an ad that is limited because you have kept one exclamation mark in one headline that limits everything, or when the conversion rate is lower than for the rest of your campaign or ad group.

To be even more precise, usually I don’t look at the conversion rate of my ads; I look at the specific ratio of conversions divided by impressions. What I really want to know is: from the impression, does this ad actually help you to convert more?

The way I was doing this is that I built specific filters within the Google Ads UI to filter out all the RSAs that I needed to improve and rework on. But now, thanks to Optmyzr, I’ve built a rule in the Rule Engine that combines all my criteria and can push me one report with all the ads that I need to improve.

Now I don’t spend that much time just in the account going to the right tab, waiting for the UI to load (I think I’ve lost one life just waiting for Google Ads to load), adding my filters, and then changing the account in MCC and redoing the same process. Now all those ads that I need to rework are pushed to me directly in my inbox.

That truly changed my productivity because now I don’t spend that much time searching or looking for the ads that I need to improve. I can focus 99% of my time just improving the ads for real. I don’t spend that much time finding which ad I’m going to change—they come to me, I improve the ads, and then I can directly focus 100% of my work on improving my client’s performance.

Frederick Vallaeys: One thing that I like here that you’re doing is taking advantage of the fact that the Rule Engine can do relative comparisons. So that last line you see there, “The conversions per impression of the ad is less than the conversions per impression average for the ad group”—that’s beautiful.

It’s one of the ways that we find, for example, expensive keywords or expensive things. A brand keyword is always less expensive than the average of the account, but if you look within the brand campaign, you can look for it to be relatively expensive compared to other brand keywords.

That’s one of these things where you unlevel the playing field. Sure, you could use the Google interface to find some simple things, but by combining the level that you look at or maybe looking across different date ranges—like, is this ad progressively getting slowly worse? It’s not noticeable from one week to the next, but if you look over a time range of seven weeks and every week it’s dropping a little bit, maybe now you know you need to do something with it. It’s that flexibility and being able to look at a more strategic level that really helps you drive great results for clients.

Frederick Vallaeys: And then once you have those great results, you’ve got to do reporting, right? That can be time-consuming.

Mat: Reporting is a nightmare! Reporting is a nightmare, plus, to be honest with you, I also challenge a lot the usefulness of reporting. Every single time, I ask my client, “If I send you a report, what kind of decision are you going to make?” And 90% of the time, they don’t know. But still, you need to give some numbers just for the CMO to look good.

In the past, it could have been a task that takes you ages. It’s not just pulling all the data, pulling it into slides, making it look nice, etc. You spend a lot of time actually not driving value for your clients but building slides.

Here also, one thing that I love with Optmyzr is that you can have very high-quality, very flexible reporting. The word is flexible because you can find some reporting solutions that will say, “Okay, here we go, that’s a report,” but you don’t want this because you want to customize it for each client. Maybe one client is going to be more focused on the search terms, another one on the performance of their ads, and you want to be able to change all those elements.

With Optmyzr, you can build automated reporting that will be sent every week, every month, every quarter—whatever cadence you want—which is going to precisely show what you want to show. You have different widgets that you can combine, but you can also customize the widgets, the columns, the tables, etc. Then it automatically generates the reporting for you and sends it to your client. You don’t even have to download it and send it.

Frederick Vallaeys: The best one is a thing that says, “If my conversions are below a certain threshold, then don’t automatically send it to my client—send it to me, let me fix some things up.”

Mat: I love this feature because sometimes you have a tracking problem and you don’t realize it for the week, and then you send a report that says, “This week you have had one conversion,” and the client calls you, “What’s happening?” So that’s a great feature, it’s kind of a safety net.

But even with this, what I truly want to underline is this ability to customize very deeply the reporting that you are sending to your client. You can pick the personalized custom columns that you have built in the account and bring and report them back on your automatic reporting. That’s super powerful. It’s not something that just tells you “conversions.”

Think about the lead gen advertisers where they have CRM conversions—you can see the form, but then after that, the qualified lead, the proposal, the customer. Here you can report on all those columns that you have in the UI; you can add them to your reporting. Not all reporting solutions are actually proposing this kind of feature.

Frederick Vallaeys: One of the latest things—I don’t know if you’re using this, but we have the CRM integration directly into HubSpot. So you can bring that in, that can become part of your bidding strategies, and again, it’s about telling the automation what it is you’re working towards.

We always talk about conversions, right? But what is the conversion? Is it the sale, or is it the lead, or is it some stage in between? That’s where it then becomes a question of: do we have enough data at the level that we want to make meaningful decisions on a frequent enough basis? And if we don’t, let’s find something earlier in the funnel and let’s be directional as opposed to super precise.

That’s why we have people like Matthieu and agencies who need to think about these things, because if it was as simple as writing a blank check to Google and saying, “Hey, get me more sales,” that would be kind of nice, but that’s not the way it works.

Frederick Vallaeys: Mat, this has been absolutely fantastic. Honestly, I think you’ve shared more insights and tricks than anyone we’ve had, so thank you for doing that.

Now, if people want some help with this—because at the end of the day, you do have to know how to use Optmyzr, you do have to have a smart strategist in place—how do people find you?

Mat: They can find me on LinkedIn. If they speak French, I also have my very own website, which is my name (matthieudastranvan.com). You can also find me on Amazon with my book, “The Google Ads Strategist Handbook,” that you can find directly on Amazon.

Frederick Vallaeys: But it sounds like you take clients who speak English—you don’t have to speak French?

Mat: I do have several clients in the US and even in other countries. So yes, if you can bear the French accent, then please reach out. I will be pleased to talk about your account.

Frederick Vallaeys: Big thank you to you, Mat, for sharing everything you did today, and thanks everyone for watching. We do this occasionally—Automation Layering Master Class, the greatest tips and tricks for using Optmyzr and automation in general to run your PPC accounts better.

Matthieu Tran-Van is one of the best, clearly, in the business. Read his book, check out his website, and give him a shout out or connect with him—figure out how to work with him.

Thanks for watching. We’ll see you for the next one.

Mat: Thank you very much, Fred. Thank you for having me.