Google’s bidding strategies evolved radically. I remember when there were only a few, and now you can have even AI helping you get the most out of your budget. Most advertisers would rely on manual bidding or use enhanced CPC, which had a little help from Google adjusting their bid.
Later came the CPA bidding or the cost per acquisition. The first CPA bidding was bad, it did not optimize things as intended, even the Google team admitted it to me during one call when I worked with a very large company (it shall stay unnamed).
However, recent developments revived automated bidding strategies. Google started pushing all accounts to adopt automated bidding as, according to Google, it has many signals and capabilities that humans can’t comprehend. It may be true.
But how exactly does bidding work? And what strategies you should use Today? And is automated bidding truly the solution of all your problems.
Let’s deep dive into the world of bidding strategies.
How bidding works?
Underneath it all Google is just an ad auction system. Since the day it was created. But it’s not a simple auction system where the highest bid wins. No. It was created in a way that advertisers all sizes can participate and win.
For that to work, Google had to make a different auction system and add additional factors. Those factors are:
Your bid
Obviously. This is still a quite important factor as it is still an auction. How it works with Google, is with your bid, you tell it what is the highest amount you want to pay for a click. The great part is usually you pay less than you specify. A lot of people think that Google’s goal is to take every penny, so when you set up your max CPC (cos per click), you think this is how you will be charged all the time. It’s not true.
Combined with account performance and other factors the price might go down. However, you could pay max amount if your ads are bad or you are in a very competitive niche.
The quality of your ads
This bring me to our second factor, the quality. Google is obsessed with ad quality. That’s why it has a Quality Score, that looks at ad relevancy, usefulness, the website and who clicks the ad. Google wants the person clicking to have the best experience. That’s why they reward you if your ads are great, and punish you if they are not. This comes in a form of higher or lower CPC.
Ad assetsThey used to be called extensions. And the rule here is simple. Add as many assets as you can to your ads. Not all will be applicaple, but the ones that are, should be created. Google will treat it as a higher quality ad. Hence lower CPCs.
Your Ad Rank
As I said, Google wants the person search to have the best experience. That’s why ad rank looks at search term, location, device, time of search, nature of a search term, other ads and results on the page, and other signals that I’m sure are a secret. This whole combination signals allows Google to determine the ad rank, each time users makes a search and decide where to show your ad.
You see it’s not just a simple auction. All these factors are the reason small companies can compete with bigger ones for the same keywords. Sure, a company with a million dollars a month will probably win against someone who spends only a $1000. It is still an auction and your bid and budget have a huge impact. But I always says “if you can’t outbid them, outsmart them”.
Now let’s dissect each bidding strategy.
Search campaigns
Manual CPC
This is the basics. Old school. You go from keyword to keyword and set your bids for each of them. Don’t worry, there a ways to set bids for all keywords, at once, if you want them to have the same bid. It will save you time.
But in general, it means that Google will not charge more than your specified CPC (cost per click).
When I would recommend:
– new account and no conversions
– low daily budget (varies per country, but lower than $50)
– when you want to learn
– when you want to outsmart your competition
When you need to try different strategies:
– large accounts (might take a lot of time to manage)
– a lot of conversions, then you might use CPA to your advantage
Enhanced CPC
You can also use Enhanced CPC. Available as an extension of manual CPC. By ticking the check box you will tell Google to adjust your bid to get more conversions. This is a great way to transition to more automated bidding strategies.
You will need to set up conversions in the account, otherwise, this feature will not work.
Once you opt-in to eCPC, your bids can be increased or decreased by about 20% (by Google), depending on how likely Google thinks this click might result in a conversion.
It seems simple but, trust me, under the hood you have a powerful Google algorithm working its ass off.
I would also recommend waiting until you accumulate some conversions so that the algorithm can pick on them as soon as you switch. For the best performance make sure your campaign has at least 100 conversions per month.
You’re still responsible for all the bids for keywords, meaning you set them. The increase or decrease happens on Google’s side, in the algorithm. The actual CPC that you see in your account does not change.
If you have an account with conversions, I see no reason why not have eCPC enabled.
When I would recommend:
– when you have at least 100 conversions per campaign, per month.
– when you want to have control over bids, but you want some help from Google.
When you need to try different strategies:
– same as with manual CPC, large accounts might be time consuming to manage.
– a lot of conversions, then you might use CPA to your advantage. CPA is a different breed of automation. Sort of a next step from eCPC.
Maximize clicks
This strategy does what it says. For your given budget it will try to get as many clicks as possible.
Here you are no longer in control of your individual bids. You can’t adjust bids on the keyword level, Google does that for you. And this is a small downside. You can, however, tell Google what is your CPC maximum limit. If you don’t want Google going bonkers with the cost per click.
I don’t like this strategy and I will explain why. Clicks are not created equal. Using this strategy you tell Google that you don’t care about the click quality as long as you get a lot of them.
Sure, Google will still look at your keywords, but if you have more than one keyword, Google might decide to bring clicks from certain keywords, just because they get more clicks. But that does not mean, they will convert on your website.
Let’s try with an example. Say you have a store and sell kids clothes. Your audience is basically moms, that buy clothes for their kids. I know dads buy clothing too, but for the purpose of this example, let’s assume only moms do. But even here it’s not all moms. You don’t want to pay for all moms coming to your site, because not all moms will like your selection, or pricing or anything else. You want moms who would buy, not the moms who just come to the store.
Same with clicks, you don’t want any clicks, even if they are from your keywords. You want clicks from your keywords, that buy. But this is not the goal for Google under this strategy.
On the other hand, sometimes, all you need is to get traffic. Have in mind that Google will still show ads based on your keywords. It’s not like the clicks will be random. But it will try to optimize the ad position, time of day, day of the week and etc. to get you the most clicks.
This does work, but I recommend it when you are sure that your website converts (engages people) well. Meaning that those clicks will not be wasted. And measure results by looking at conversions or revenue, not traffic.
When I would recommend:
– never
When you might try it:
– when you need to get targeted traffic and it’s very easy to engage with your site. Like, entertaining website or news.
Maximize conversions
It’s the same as with clicks, but this time Google will work hard to get you the most conversions within your budget. You can also set the maximum price for the conversion.
If you have already conversions flowing in, based on your previous strategy (probably CPC), then you can test at least one campaign by switching to this bidding strategy.
Works better for conversions with fixed value, for example leads, registrations, email signups. For ecommerce, not that often, because each conversion has a different sale value. Unless you sell one product with one price, then it’s OK.
The reason why is the same as with maximise clicks. Not all conversions are equal. So Google might optimize towards conversions that are, cheaper products, but you make less money from those products, hence your revenue suffers.
If you do try it, make sure to keep an eye on the total conversions and the cost per conversion. If you have different offerings, make sure it’s not leaning more to least profitable options.
When I would recommend:
– when you have a lot of conversions in the account
– ideally your conversions do not fluctuate often
– your offering has one price, rather than a lot of different ones.
When you need to try different strategies:
– I would go for CPA bidding in this case (coming up later in the post)
Target impression share
This strategy is great when you want to get more impressions. You can select 3 targets for this strategy: anywhere on the search page, top of the results page, and absolute top of the results page.
Anywhere on the page means that it will get you an impression in any of the positions. It could be the top or bottom. Top of results means 1 of 4 positions above. And absolute Top means it will strive for number one at the top.
Then you can also select Percent (%) impression share to target. There is a limit of how many impressions each keyword has. Think of an impression share like a pie, it’s limited. So what percentage piece of that pie you want.
Google automatically adjusts your bids depending on the chosen target. And having your budget in mind will try to get the most impressions. Most often it will mean more clicks, so watch out for your total spend and of course impact on your conversions.
If you’re afraid to spend too much limit the budget for that campaign or limit the max CPC.
This strategy is quite good if you want to outrank some of your competitors and conversions are a secondary goal.
When I would recommend:
– never, unless your boss want to see your ad above your competitors. Yes, I had similar requests. Not conversions or sales. Just to beat the competition. Crazy.
When you might try it:
– maybe with branding goals in mind. But even then there are better options.
Target CPA
Now we’re moving into automated bidding strategies. First up is the target CPA, which means cost per acquisition, or how much are you paying for your conversion. Conversion can be anything from a sale to an email subscriber.
I always advise using it when you have more conversions, so it’s not very good at the beginning of your campaign. Try to get 100-150 conversions per month, per campaign. Then the algorithm catches on quicker and is more precise.
I know Google and some experts recommend starting with 30 conversions. Don’t listen. The only time it works if the conversions are coming in quick. For example, you started your campaign and within a week you have 30 conversions. Then you can move into CPA bidding and see if you can scale it even more.
Google does recommend the price based on historic metrics, but in general, you should know how much you can pay for each conversion.
Quick tip, if being on manual CPC, you are paying $30 for conversion, but you can easily pay $50 when switching to target CPA, increasing it by $20. In other words $30 + 20%, $36. This way you will give more room for the algorithm to play. Once it catches on, you can slowly reduce the target back to $30.
I always recommend setting a target CPA. Otherwise, Google might take it too far with the conversion price. You might get more conversions, but the price will be too high, meaning that you will be losing money on each conversion.
When I would recommend:
– 100+ conversions per month per campaign
– want to see if AI (machine learning) can help you scale faster.
– Large budget accounts
When you need to try different strategies:
– want more control over keywords and bids, use eCPC
Conversion value + target ROAS
This strategy is good when you have dynamic values for your conversions, like an eCommerce store. Your products are priced differently so obviously you can pay a different price to acquire them.
This strategy uses Return on ad spend (ROAS), meaning that you specify what return you want to get in percentages from your spend.
Say you spend 100$, and you want to get a return of 5x, you specify 500%.
It is a bit confusing as you most likely will have to specify a number in the hundreds and sometimes thousands. Also, you have to know how much money each sale or conversion brings. Ideally, you have your tracking set up so that the value is taken into Google Ads on each sale dynamically. So in the account, you see all the value (or revenue) from the sales, and you see how much you spent.
Again, this is not for beginners. I would recommend it at a later stage and only if you have many differently priced products.
Besides having conversions and tracking set up properly. There is another thing. Google will focus on the total value if you won’t specify the Target ROAS. But don’t forget that Google does not know your margins and other costs. When the value increases, so might the spending. You have to watch if the return on spend is positive. Otherwise, you might spend 5000$ to get 5500$ in value (revenue). The value is increased, but the campaign itself, in terms of profit, is not performing.
Always set Target ROAS or just keep an eye on the performance.
When I would recommend:
– 100+ conversions per month per campaign
– eCommerce store mostly
– Large budget accounts
– proper tracking with dynamic values
When you need to try different strategies:
– CPA bidding can do the tricky quite nicely.
Don’t just switch, test
When I try this or the other, ideally you should test it. That’s why I always recommend starting with manual CPC, then testing a bidding strategy for that campaign.
It’s easy to do as Google has a built-in experiments. They are simple do set up, and will show you if switching to one or the other strategy has any impact.
Let me show you how to set it up. Go to your Campaigns, in the middle menu column find Experiments, click and select Custom experiments. Now click the blue plus button.
Here select custom and search.
Now enter experiment name. Ideally something description so you would know what is being tested. And bellow select the campaign you want to test.
Google created a draft and you need to make changes. Go to campaign settings, find bidding strategies and change to the one you want. Once you made the changes, it’s time to schedule a test.
Click Schedule above. Choose metrics you want to watch. I recommend adding conversions and you can and price per conversion. Scroll down and select the start date. The rest of the setting leave as they are.
Click schedule and wait. You can check in in a few days to see if it’s running.
You will see a small dashboard that shows the results and bellow your original and test variation, so you cna copare metrics. I would advise to always wait until it’s significant. That’s why I don’t recommend runnign on experiment on low traffic or conversions campaign.
Conclusion
I hope you have a better understanding of each bidding strategy. You have to remember that there is no one strategy that is better than others. It all depends on the product, budget, country and many other factors. So please, pretty please, don’t listen to any experts saying one is better than the other.
I most of the time start with eCPC and once I have enough data, I optimized ads, keywords I try to test and find the bidding strategy that works best for this particular business.
I have accounts with mixed strategies, some of them are on manual some of them on automated strategies and both bring great results. No matter the strategy you will still have to do some work. There is not set and forget option here. All campaigns require work.