Dynamic search ad campaigns utilize Google’s organic website crawling technology to automatically target relevant user-intent based search queries based on relevant website URLs. In this blog entry, we cover everything from best practices to tips and tricks for Dynamic Search Ads.
Google ads have been seeing a shift of landscape from behaviour-based targeting to data-driven targeting. Earlier, most of the targeting parameters revolved around forms of targeting like manual CPC and ECPC. However, recently Google Ads has started to integrate data-driven decisions into its bidding strategy.
Hence why, new strategies have been introduced to the platform such as maximize clicks, maximize conversions, TCPA, and more. Once these bidding strategies were tested with enough data across Google ads, the new campaign types were introduced on the platform which was almost semi-automated such as Dynamic Search Ads and Smart Campaigns. Many marketers have leveraged these campaigns and it’s time you leverage them as well.
Smart shopping campaigns are basically a hybrid of standard shopping campaigns and dynamic remarketing campaigns (display) with automatic bidding and larger reach. From the shopping campaign point of view, the bidding is automated and from the display campaign point of view, the reach has become broader. Earlier, the shopping campaign would run only on the search network but now, it reaches across display networks as well.
There are a few basic requirements for effectively running smart shopping campaigns and I will cover those now, the more advanced information will be covered later in this article.
A smart campaign should have transaction-specific conversion tracking. In a nutshell, when a conversion happens on your website, the sales value should also be passed to the conversion tracking pixel. You can read about it here. You need to dynamically pass to the event snippet the value of the conversion and its currency by setting the value and currency parameters highlighted below, at runtime.
Additionally, you need to have Global Site Tag along with a remarketing list of at least 100 active users.
When you set up the dynamic remarketing tag on your ECommerce website, make sure that you are passing as many as possible parameter values to the platform. The more granular information you pass, the better the performance would be. So, why is dynamic remarketing so important?
In the picture above, you can see that the tag hits stopped being delivered towards the end of September. And how did that impact the performance of the smart shopping campaign?
You can also see here that the conversion took a dive towards the end of September because of certain Ecommerce parameters not being fired. If you do not set up custom parameters, machine learning will be used to match products from your feed based on insights provided by your Google Analytics or global site tag, such as your page titles and landing pages.
If you have not realized until now, then you should know that the optimization capacity of a smart shopping campaign depends on the strength of the remarketing list build. The more data Google ads platform has, the better it will optimize the smart shopping campaign. Have a deeper look at the remarketing list from the Google Ads platform below:
Google ads have automatically created a few remarketing lists from the Google ads remarketing tag. These audience lists work as a sample list using which the smart shopping campaign algorithm optimizes the campaign performance. If you even further notice in the list above, the remarketing lists are built upon:
All of these pieces of information are passed to the platform using the dynamic remarketing tag. That’s why I addressed the importance of dynamic remarketing tag implementation at the beginning.
Smart shopping literally combines the ad spend of shopping and display campaigns. So, if you are planning to start a smart shopping campaign, combine the average daily spend of shopping and display campaigns over the past 30 days and allocate that budget to smart shopping ads.
You should also pause the dynamic remarketing as well as standard shopping campaigns as a smart shopping campaign takes priority over these two later campaign types. In the screenshot below, we had to pause the standard shopping campaign as it stopped spending since the smart shopping campaign kicked off and took over the priority over the standard shopping campaign. Also, if both of the campaigns are running in parallel then, there is an interference in the campaign learning for smart shopping campaigns.
I have seen many people starting with a limited budget and a limited number of products when starting with a smart shopping campaign. That’s not the best practice. The idea here is to make your campaign learn as fast as possible and as reliable as possible. Also, to have comparable or better performance to standard shopping or dynamic remarketing campaign, it is recommended to have at least the same selection of products.
If you did not run any such campaigns in the past then, it is still advisable to have a large set of products so that the campaign has more choices and better decisions to make and are not restricted by a few products.
However, product segregation is also important. Once you give a budget, the Smart Shopping campaigns automatically maximize your conversion value within a given budget. It often results in the machine learning algorithm recognizing the top bestsellers while other products get left out.
In case, you have certain mandatory business requirements that would not allow you to consolidate the products, then create separate campaigns for such product lines. Some examples of such campaigns could be:
You should give some time to the campaign in order for it to learn. Typically, it starts performing within a week but Google recommends it to be a minimum of two weeks. However, you should also keep a tab on external factors as well such as seasonality, weekends, events, offers, holidays, etc.
As the campaign learns, the performance in terms of CPC or number of clicks might fluctuate. For example, initially, there might not be enough clicks on the campaign. That’s because the campaign strives to optimize for conversion value than clicks. So, it tries to eliminate the non-converting clicks and bid higher on converting clicks with the saved money.
Some campaigns might perform and some might not. It is hard to tell when to kill the campaign. In my opinion and experience, it is better to kill the campaign when you have stability in the data. For example, there is not much fluctuation in the following context:
Dynamic search ad campaigns utilize Google’s organic website crawling technology to automatically target relevant user-intent based search queries based on relevant website URLs. In this blog entry, we will be covering everything from best practices to tips and tricks for Dynamic Search Ads. Let’s dive in!
Dynamic search ads utilize organic web crawling to automatically target relevant user-intent based search queries based on your website content. The keywords that you target through a standard campaign might not have any user intention but through DSA, you directly address the user-intention.
Example:
Most of us will target [Grey Wedding Suit for men] as a keyword. However, the DSA campaign might discover search terms such as “grey wedding suit for short man” or “greyish wedding suit for brown man” you really never know. So, short, intent-based search terms can come in any form.
User-intent is further addressed by dynamically generated headlines. I know they sometimes look weird on Google paid search ads since Google’s machine learning and NLP have not yet become smart enough but DSAs try to address the user intent by generating dynamic headlines. Dynamically generated headlines are slightly longer than standard text ads adding more flexibility, relevance, and visibility for online searchers.
When you update the content of the page, based on the fresh crawl, Google will try to match your updated webpage with the new intent-based search terms.
Time & Traffic
Imagine a situation where you want to expose more online users to your brand because you have awesome products but you can’t find relevant keywords through keyword planner because there isn’t enough keyword volume for those keywords that you are looking for. DSAs can help you quickly scale your campaign by driving user-intent based traffic to the most relevant landing page on your site based on the user search terms. It might also result in budget wastage if not monitored carefully.
In this case, you do not need to do new keyword research from scratch, no ad copywriting, no grouping, no bidding, etc. Google does it all for you. You can even further utilize this free gift from Google to optimize your standard search ads campaign performance. Keep on reading to know why.
Often I have seen that some of the poor performing keywords which were paused in the standard campaigns due to high CPA had begun to perform much better in DSA campaigns. On top of that, I have also seen that the longer you run the campaign, the better the performance of the DSA campaign gets.
Total CTR of my DSA campaign is more than 40% which is not the case with even branded standard search campaigns. DSAs are highly intent oriented.
You might not have full control over how your ad copies and for what keywords or on which landing pages.
Since Google tries to fit in the organically crawled website content in ad copies, sometimes, ad copies might appear weird. Though Google is working on it but there is a long way to go when machines will start writing like humans.
Though one of the key pros of DSA campaigns is that they bring in new traffic and new search terms but that comes at a cost of ad spend. If you cannot control the prospective search terms using negative keywords, Google might show up your website for many unimaginable search terms.
In one of my DSA campaigns, Google threw in a total of 198,426 search terms over a month of 6 months. Out of these search terms, a total of 193,418 search terms produced no conversions and accounting for a budget spend of $307,768 while the total budget spent was $452,278.
Though a conversion rate of 5.5 leveled out the campaign spend. Imagine being able to control the budget wastage and guiding them towards the conversion. Most of the non-converting search terms were long-tail keywords.
Since DSA campaigns utilize auto targets and auto targets belong in an ad group, try to be as granular as possible with the AdGroup to have better control. If you have a product line or product category, try to make ad groups based on them. Here is a screenshot of the sample campaign structure for a tea company.
Note: Don't forget to add other categories/ sub-categories/ product-line as negative keywords in all other ad groups. Please also discern whether you think keywords or search terms should intersect.
I have already highlighted in the section above on budget wastage on a large number of search terms and long-tail keywords. Imagine controlling negative keywords to make converting keywords total percentage reaches 10% and that gives us the ROAS of 22. So keep on building the negative keywords list in order to minimize unwanted search terms.
In DSAs, two lines of description are the only space that you get to show off your brand. Use those two lines to high top-level brand values, offers or promos. Always keep user experience and intent on your mind.
Though Google will show the remarketing ads to the audience which have already seen you however if we retarget the remarketing audience through DSA, your remarketing audience can land on your website for some other search terms as DSAs bring a whole lot of new audience. You can either use the “Targeting” option or bid higher through “observation”.
The longer you run your DSA campaign, the better it gets. Here is an example of campaign learning. The campaign started slow but learned a lot towards the end of the year as you can see in the screenshot below.
Make sure that there is no internal competition on the products targeted by the DSA campaign through other standard campaigns. Make sure to exclude the potential keywords which might create internal competition. On the other hand, if you are pausing or removing a keyword from any other search campaign, make sure to add those keywords as negative keywords in your DSA campaign as well. Google also considers overall account performance including keywords as well.
Here is a list of pages that should generally be excluded from the dynamic ad targets
The better you do at SEO, the better you will also do with DSA – that’s what I have observed with my ECom clients. Organically organizing your product pages for SEO could be tricky. However, to know how the content targeting the user-based intent should look like, visit Amazon individual product pages.
SEO, apart from paid media, should also be a high priority. Other than bringing in new organic traffic, SEO also helps learn and optimize your DSA campaign quickly.
There are a lot of ways to optimize a DSA campaign. Here are a few ways that on a generic level work for the DSA campaign optimization.
Here is a list of the negative keywords that I have put in the master negative keyword list. Updating the negative keywords helps you cut the cost of the ads being shown on unwanted search terms. Although, I have also observed that the need to update the negative keywords subsides with time as Google ads learn and more data accumulates in the account. You can do the same for the landing pages as well.
If you have multiple ad groups in your campaign, different ad groups will have different levels of performance. And if you see that google ads are allocating more than the required budget to the low performing ad groups, in that case, I weed out the low-performing ad groups and put them in a separate campaign. The easiest way to do so is to duplicate the campaign and do the needful. I usually prefer to give 80% to high performing ad groups campaign and 20% budget to the low performing ad groups of the total budget from the initial DSA campaign.
Facebook Ads optimization is completely different from optimization on other platforms. The nuances of Facebook ads optimization are more dependent on those parameters which don't apply to other platforms like Google ads. By following these optimization tips, you should be able to reduce your cost per conversion and reduce your Facebook ads budget.
Unlike Search engines such as Google or Bing, Facebook ads also consider audience engagement in ranking your ad. Facebook uses three different metrics in evaluating your ranking, with engagement rate ranking being one of those factors unlike other advertising platforms. Engagement rate ranking considers different forms of engagement on your ad such as: likes, reactions, shares, comments, comment threads, and shares. This means that the more engagement you have the better your ads will perform.
Before running any advertisements, use the ad creative as a post via your Facebook page and then boost it. When you boost a post, Facebook optimizes your post for reach and engagement by default. This allows your boosted post to receive lots of extra social engagement and if you have a built in following it will help you leverage that as well.
When Facebook starts delivering your ad set, whether at the start of a campaign or after you edit it, it does not have all the data necessary to deliver it as stable as possible. To gather that data, Facebook shows ads to different types of people to learn who is most likely to get you optimization events. This process is called the “learning phase.”
Once Facebook has all the data, your ad set can experience fewer performance fluctuations. At this point, you can make an informed decision about your ad set. If you’re satisfied with your results, you can let it keep running or increase its budget. If you’re unsatisfied, you can edit the ad set to try to improve its performance or pause it.
During the learning phase, you can expect more performance fluctuations than usual. Do not make any significant edits to your ad set during the learning phase. Doing so can cause the ad set to reenter the learning phase before generating any meaningful information.
Your ad set needs about 50 optimization events after starting to run or since the last significant edit to complete the learning phase. If your conversion window is a 7-day click, the 50 or so conversions you need all have to happen within 7 days of the click that led to them. This means that ad sets with longer conversion windows can need more time to complete the learning phase since they’re likely promoting products or services that have longer consideration times.
If you are familiar with Google ads, it's more likely that you see A/B testing as ad variation and using different ad copies. With Facebook, it's a completely different ball game. With Facebook you can get multiple ideas to test. So, I recommend you divy up your testing into to different sections: Ideas to Theme & Theme to Creatives.
With Ideas to Theme, try to homogenize your ideas into one theme. Do the first A/B test within your theme and decide which theme is performing better. Next, for Theme to Creatives, test creatives within the same theme.
Before we jump right away into exclusion marketing to lower Facebook CPC, you need to fully wrap your head around how the Facebook algorithm works.
By selecting a campaign type, you are selecting an objective for your campaign. Facebook will optimize your campaign for that objective. Their targeting algorithms will automatically try to find people within your target audience that are most likely to convert based on the campaign objective. One of the biggest factors in determining whether someone is likely to click on your ad or not is whether they have already clicked on it previously.
People that have already clicked on your ad are obviously interested in what you’re advertising, so they are highly likely to click on it again. Facebook’s targeting algorithms know this and they will often show your ad to people that have clicked on it in the last few days. On the whole, this is a good thing, because not everyone will convert into a sale the first time they click on your ad and come through to your website.
After that first click, a lot of people will take some time to think about what you’re offering or just become distracted and simply forget about it. You don’t want to miss out on easy leads and sales, so advertising to these people is important. But of course, a lot of the people that have clicked on your Facebook ad will have converted into a sale and continuing to advertise to those people is a waste of money.
You need to distinguish between those people that have clicked and converted into a lead or sale and those people that have clicked and not converted. Once you do that, you can then exclude people that have already converted from your Facebook Ad campaigns. This is called Exclusion Marketing.
In my personal experience of doing Facebook ads optimization, this is a huge factor in reducing cost. That’s why I have dedicated a whole blog to Facebook Exclusion marketing and I would suggest you read that as well.
Facebook ads are more about attention and creatives. If your ads are not attention-grabbing, they might not work. But, the story does not end there, after someone decides to read or scroll through the carousel ads or watch the video, you have only the first few seconds to grab their attention. Here are some of the tips on optimizing Facebook Ads attention:
Under Facebook exclusion marketing, we have already discussed how Facebook prefers to show the same ad to the same engaging audience again and again. After a certain number of ad impressions also called ad frequency, the non-converting users start to get ad fatigue, when that occurs, you need to address ad fatigue. Here are some tips on addressing ad fatigue to improve your facebook ads campaign.
Take a long look at the screenshot below and try to see if you notice anything unusual...
The main reason, the CPA is consistently increasing is because of the following reasons:
Audience Exhaustion
The overall Reach of the AdSet is around 210K but the overall reach combined over the entire month is 385K because Facebook can't find any new audience based on the targeting defined at the AdSet level. Which is causing Facebook to recycle the same audience again and again. Again, we cover how Facebook decides to show the ads to a set of audience in Facebook Exclusion Marketing
Ad Fatigue
Due to audience exhaustion, Ad Fatigue is happening when the same audience is being reused over and over again.
Sometimes overly narrow targeting can limit your reach and result in a saturated audience. Here is a screenshot from such an AdSet:
If you look at the above data, this AdSet has spent around $1300 and reached only 739 with a lifetime frequency of around 37.
You can find the audience saturation levels of any Adset by clicking on 'Inspect' below the AdSet name.
The lowest cost bid strategy provides no direct control over costs. Costs may rise over time as cheaper opportunities are exhausted or the budget is increased. If you’re trying to keep your average cost at or below a specific amount, consider switching to the cost cap bid strategy.
How does Cost Cap Bidding work on Facebook?
The cost cap bid strategy controls how high costs can rise. It aims to get the cheapest opportunities possible below your cost control to maximize volume. Facebook may not be able to spend your entire budget if there aren’t enough opportunities at or below the cost control you entered. If you hit your cost control, increasing your budget may not get more results unless you also increase your cost control.
How does the lowest Cost Auction Work on Facebook?
The lowest cost bid strategy aims to get you the cheapest opportunities possible while spending your entire budget. Costs may rise over time as fewer opportunities are available. Costs may rise as budget is increased. The lowest cost bid strategy prioritizes spending the entire budget, and costs can rise with no limit until your budget is fully spent.
Although Facebook tries to define the right audience by learning from the initial learning phase, it might not always be correct and so you need to keep a check on that. The best way to do this is by breaking down your campaign or AdSet performance by age and gender and that will allow you to see which audience is performing better.
For example, in the above case, Facebook decided to spend more on females from 18-24 but females from 25-34 were actually the best-performing ones. You can use similar breakdowns to optimize your Facebook campaigns based on:
Using automated rules, you can achieve any of these four conditions on ads platforms.
Automated Rules:
Make sure that you implement a proper budget allocation strategy to reduce the cost of your Facebook Ads campaign and increase your overall return. Here are some examples of the right budget allocation
It's not uncommon that a person builds out their entire Facebook Ads campaign expecting it to perform really well but instead actually underperforms. There are many factors that can contribute to this, some of them are in your control and some factors are completely out of your control. Today we're going to pick apart each of these contributing factors and diagnose why your Facebook Ads campaign is underperforming.
Audience saturation is one of the most common reasons why a Facebook Ads campaign is not performing and various parts of your campaign could cause this. For example: improper targeting could easily limit your reach which in turn would result in a campaign with almost no or limited performance.
Sometimes overly narrow targeting can limit your reach and result in a saturated audience. Here is a screenshot from such an AdSet:
If you look at the above data, this AdSet has spent around $1300 and reached only 739 with a lifetime frequency of around 37.
You can find the audience saturation levels of any Adset by clicking on 'Inspect' below the AdSet name.
Another common issue that can cause your Facebook campaign to stop performing or cost per conversion to go up is an increase in competition. When you run an ad set, it gets entered into many auctions with other ad sets that are trying to reach the same target audience. The winner of each auction gets to show an ad to someone in that target audience. Additionally, some times of day/week/month/year are more competitive than others.
In the above screenshot, you can see that the auction competition has gone up (Circled Blue line). You can also notice that the moving average of cost per lead has gone up overall since the auction competition went up. This is because many advertisers are trying to reach the same people and are willing to pay more to do so. When this happens, you may have trouble spending your budget or have to pay a higher cost per result.
During festivals or special seasons, many advertisers broaden their targeting. For example, you might be targeting the people who are interested in just fishing. However, during summer, the same people could also get targeted by resorts, hotels, and fishing companies.
Let's say I am running Facebook ads for online doctors & prescription services in Canada, and I know that some of my competitors are Rexall and Get Maple. To monitor their current activity, I would find their Facebook pages and view their page transparency.
Once you click on the “See More” option on Page Transparency, you will see an option to go to their Ads Library at the bottom, as shown below:
Once you enter the Ads library, you can see the kind of ads your competitors are running. You can also see their ad copy, you can guess their targeting parameters and find out if your competitors are responsible for a decline in your campaign performance. Be mindful that Facebook might not show you entire ads and ad spend is not accurate.
Sometimes we force Facebook to go into the learning phase with limited reach. On top of that, we often force the platform to do repetitive learning on campaigns with a similar objectives. Before we dive further into this though, I am going to explain how Facebook's initial learning works.
When Facebook starts delivering your ad set, it doesn't have all of the data needed to conduct a stable delivery. It can happen either at the start of the campaign or after an edit is made to an AdSet. To provide stability to the AdSet delivery, Facebook shows ads to different types of people to learn who is most likely to get you optimization events. This process is called the "learning phase".
Once Facebook has enough data about the AdSet it will experience fewer performance fluctuations. At this point, you can make an informed decision about what to do with your AdSet moving forward. If you’re satisfied with your results, you can let it keep running or increase its budget. If you’re unsatisfied, you can edit the ad set to try to improve its performance or pause it. However, there is more to it as the initial learning period is also dependent on the AdSet optimization event and conversion window.
Facebook says that “Your ad set needs about 50 optimization events after starting to run or since the last significant edit to complete the learning phase”.
Additionally, if your conversion window is a 7-day click, the 50 or so conversions you need all have to happen within 7 days of the click leading up to them. This means that ad sets with longer conversion windows can require more time to complete the learning phase since they’re likely promoting products or services that have longer consideration times.
The above sets of AdSet are a perfect example of limited initial learning. Here is how:
As of recent, many eCommerce brands cutting down on their advertising efforts due to the Coronavirus. Which is an entirely understandable response especially if your manufacturing or supply lines have been affected. However, if your business hasn't been strongly impacted from an operational standpoint...stopping advertising right now to save money is like stopping a clock to save time. It's completely counterintuitive.
Here is why — though our overarching concern and moral obligation should be to do no harm to our employees, business partners and customers. We shouldn't let that prevent us from seeing the unique opportunity being presented for eCommerce brands to expand.
Due to an enormous increase in online activity and a considerable decrease in online advertising, CPMs are the lowest they have been in the last five years. In essence, you can advertise for cents on the dollar. Additionally, brick and mortar competition is non existent and many online stores are also over-extended or scaling back their efforts out of fear. Hence why, if you own an eCommerce store right now, this is by far the best time to triple down on your advertising efforts and aggressively expand.
Here is where most brands are going wrong — you don't want to triple down on what you have been currently doing. At least from what we have seen with our clients that won't work. Instead, you need to change and adapt your approach and triple down on that. Whether you're coming in from different angles with your copy to address different problems, switching up landing pages, optimizing for different KPIs, whatever it may be, you should adapt. We've adapted, and as you can see in the screenshot below, we are still achieving incredible results for our clients. So, here are four pieces of actionable advice you and your team can implement today to maintain and even substantially increase your return on ad spend in our current climate.
Address your community — There are two kinds of eCommerce brands that exist in this industry, the kind with no purpose other than making as much money as possible and the kind with a deeper purpose that empower their customers. In times of distress and chaos, it becomes incredibly apparent, who are who. That's why you should reach out to your customers addressing our current circumstances and how your brand plans to take action. People buy from who they trust so, establish trust, and you will create a strong base of loyal customers who keep coming back.
This email broadcast Cloudflare CEO Matthew Prince sent out to address Coronavirus is an excellent example of capitalizing on the circumstances, while at the same time still be extremely sensitive to the issue. People will remember this and become lifelong customers.
Change your messaging — Instead of solely focusing on the benefits of your product or how your product can make a positive impact on your customer's life. Emphasize the fact that your product will get delivered directly to their doorstep, how throughout the entire process there won't be any contact, that you are donating a percentage of their purchase to finding a cure, etc. Adapting your ads in this way will increase your retention rate, average order value and cost per conversion.
Switch up landing pages — This will have a profound impact on the conversion rate of traffic that clicks through to your website. For example, as opposed to sending traffic directly to a product page. Send traffic to a separate landing page where you, as the brand's owner, are addressing what's going on in the world right now and how you want to do your part. While addressing them, you could provide a coupon code or ask them to sign up to your email list so you can update them on your contributions to finding a cure and from there send them to the product page.
Optimize for different KPIs — Now, I would personally say this is the last resort, as optimizing for purchases should ALWAYS be a top priority. However, if you are one of the eCommerce brands that can't operate your business as usual, I would recommend adapting by changing your key performance indicators.
An example of this, would be optimizing for email sign-ups or leads so that later down the line you can run some cheap retargeting ads and turn them into paying customers. Although from face value, it may seem as though you are losing money by implementing this strategy, that's not the case. It's just not instantaneous profit and right now, advertising costs cents on the dollar so it's an excellent time to take action.
That's everything for now, however, depending on the duration and severity of COVID-19 we may release a flow schematic detailing how to properly implement these practices as well as others. If you are a client of ours you will have already received an email earlier this week addressing all changes we are implementing in response to this rapidly developing situation. To all our readers, I hope that you found this advice valuable. My heart truly goes out to all those who are suffering from this craziness. I just wanted to give you my thoughts on this pandemic we are facing from a business standpoint and share some tactics we are implementing for our clients. Finding success when it comes to producing results during this time is about adapting your approach and we've found incredible success in doing so for our clients and now you can as well.
Stay Safe Everyone ❤️
It's time to make your ad-budget count, scale your business and blow up your sales. This is not your typical "marketing agency". We provide you with everything you need to be successful.
