The Google Ads Bidding Strategy Guide You Wish You’d Always Had

Google Ads

There is no such thing as a successful Google Ads campaign without well thought out bidding. The bids you choose directly influence your visibility, spend and ROAS.

Even if you somehow know the exact optimal bid, search auctions are highly dynamic and vary in conversion intent, so the “right” bid is actually a constantly moving target. Here’s where bid management comes in.

There are various ways of doing this — manually, automatically and somewhere in between those two. All options will be explained here.

Table of Contents

Why do different bidding strategies exist?

The main reason for seeing so many bidding strategy options to choose from in Google Ads is the obvious fact that companies vary so much — their business models, seasonality, marketing strategies and budgets require different approaches and implementation.

Different Google Ads bidding strategies have different purposes. They are all designed to help you manage your bids as effectively as possible towards reaching your business goals. So first, you need to define your goals to choose the right bidding strategy and then set it up properly and follow up with continuous optimization as needed.

What do you need to do before choosing a bidding strategy?

Setting Goals

The first thing that needs to be done is defining your business goals and setting targets for your Google Ads campaigns. All this means you have to be tracking your goals (e.g. purchases, sign-ups, leads, etc) in the first place. This can be done with Google Analytics or Google Ads tracking. Read about the differences between them here.

Your goals could be any of the following:

  • Increasing or maintaining revenue within a set target ROAS (return on ad spend)
  • Getting leads with a set target CPA (cost per acquisition)
  • Maximizing leads
  • Maximizing traffic to your website
  • Raising brand awareness
  • Driving more sales from your website
  • etc

Once your goals are defined, you should get to know your customer and define one or a few various funnels that describe the customers’ expected behavior and steps towards completing the desired action(s). For example, a potential behavioral pattern (points of contact with your business) could be:

Generic paid or organic Google search → remarketing display banners → more specific Google search (with higher purchase intent) → dynamic remarketing banners with sale offer → PURCHASE (→ newsletter → remarketing banners → etc)

You can check your customers’ conversion paths from the Google Analytics report found at Conversions → Multi-Channel Funnels → Top Conversion Paths. When you get to the report, make sure to include only the conversions, path length(s), interaction types and lookback window that you wish to analyze.

Google Analytics MCF report

Get access to the demo account here. Screenshot from Google Merchandise Store Top Conversion Paths report:

Top Conversion Paths

If you have an idea of your customers’ purchasing behavior, it’s easier for you to assign which channels (campaigns) should assist different funnel steps. Campaigns that are targeted for different funnel steps should also have different goals.

For example, at first, your goal could be to get potential (relevant!) customers to your website so they could get acquainted with what you’re offering. If they don’t purchase straight away, you’d want to stay top of their mind and get them to return to your website. If they have an actual purchase intent then you could keep reminding them about your products/services with different messaging, and, if necessary, offer them a promotion or some form of stronger incentive until they have completed the purchase.

So, once you have defined how to reach your customers during their purchasing behavior steps, you can start defining goals for each of these campaigns (steps) separately. Let’s use the same example as before:

  1. Generic Google search GOAL: maximizing (relevant!) visitors to your website to get potential customers acquainted with your brand and what you offer
  2. Remarketing display banners GOAL: maximizing returning visitors
  3. More specific Google search GOAL: maximizing relevant (high-intent) visitors and if possible, getting them to complete the desired action
  4. Dynamic remarketing banners GOAL: getting the visitors that haven’t converted yet to convert (for example, show banners with products that the user viewed or left in their cart to remind them to complete their conversion)

Choosing an attribution model in Google Ads

To make sure your bidding decisions are in line with your goals, do acknowledge the importance of choosing a suitable attribution model in Google Ads. Customers’ journeys towards completing a purchase often include more than one point of contact through Google Ads.

Even though some generic keywords or upper-funnel display banners might have very few or no conversions at all, they are still likely a very important part of your funnel as they are the first touchpoint that brings users to your website in the first place.

Naturally, you’d want to attribute part of the final revenue to all relevant points of contact. In order to do so, choose an attribution model that distributes conversions and revenue between all (Google Ads’) points of contact (Google Ads: Tools → Measurement → Conversions → Attribution model):

  • Data-driven
  • Linear
  • Time-decay
  • Position based

Frederick Vallaeys has summarized this idea very nicely in his article about bid management:

“You may also find yourself potentially hurting your account performance if you don’t consider the entire conversion path. Some keywords may not have a lot of last-click conversions, but they may play an important role early on during the purchase cycle. If your bids are purely based on last-click attribution data, you may start to narrow the upper portion of your funnel, leading to problems down the road.”

What the attribution model in Google Ads does is that it chooses how to distribute “credit” for conversions (the number of conversions and their revenue) between keywords. So, for example, if you have chosen the linear attribution model and 4 different keywords brought a user to your website before they finally completed a conversion, each of those 4 keywords will get 0.25 conversions and 25% of the revenue attributed to them. (If you use last-click attribution, the only keyword that gets credit is the one that brought the user to your website the very last time before they completed their conversion.)

The more you attribute conversions to the earlier-stage touchpoints, the more aggressive your approach is. Meaning, it is riskier, but also geared more towards growth. The more you attribute to the final-stage touchpoints, the “safer” your approach is as you are focusing more on maximizing the performance of rather lower-funnel keywords.

Google Ads Bidding Strategies – Definitions, Pros, Cons, When & How to Use

Manual Bidding

Manual CPC

manual cpc

Definition

Manual CPC is “a bidding method that lets you set your own maximum cost-per-click (CPC) for your ads. This differs from automated bid strategies, which set bid amounts for you”. (Source: Google Support) Bids can be set on the keyword or ad group level.

Remember

Manual CPC bidding is the option for you if you have a strict limit for the maximum amount you’re willing to pay for a click. Or, if you feel confident enough to maintain 100% control (and responsibility) over the amount you pay for each click. Or you want the feeling of inner peace that Jason Rothman describes he gets from managing bids manually.

Remember, if you’re using manual bidding, you’re setting maximum bids at the keyword level. It’s important to remember that different searches caught by the same keywords may have different intent but manual CPC bidding sets equal maximum bids for all of them.

To use manual bidding efficiently, you need great analysis skills to make sure you’re basing your decisions on statistically relevant data and not missing anything important. There are many details to take into account when trying to find the most optimal bid for every keyword and it requires constant re-reviewing and work. Which isn’t necessarily a bad thing.

To make informed decisions about which bids to set for your keywords you should take into account keyword (or ad group) level conversion rates and target CPAs. A target CPA being the cost per acquisition that you are willing to pay to stay profitable.

Manual CPC is often used together with Enhanced CPC which is discussed further below.

It can also be paired with bid modifiers to take into account the information regarding different signals (time of day, day of the week, gender, age, device, etc) that we have access to and adjust bids accordingly.

Automated rules can also assist you with managing bids based on previous data and its averages. (Bid modifiers and automated rules are explained below.)

Andres Ruul, Google Ads Team Lead at Holini

“Manual CPC bidding is useful if you’re not able to import actual sales numbers from your CRM to Google Ads. E-commerce left aside, most businesses’ conversions are leads (sign-ups, free trial requests, etc). And lead quality can vary enormously. Not all leads convert to be actual paying customers (which is the ultimate end-goal) but Google sees all leads equally if a lead has been defined as a conversion action in Google Ads.

In reality, a lot of leads don’t convert to be useful for the business (paying customers), but Google actually has a lot of conversion data on poor-quality leads and the Machine Learning might be working in the wrong direction. For example, a loan-provider might get loan applications of which 70% get a negative answer. But Google’s Smart Bidding would keep bringing in low-quality leads (as all leads are equal in the eyes of Google because tracking stops once the application has been sent). But manually we have the option to use common sense and make optimization decisions that would increase our share of high-quality leads.

If possible, try to import actual sales numbers from your CRM to Google Ads and you don’t have to worry about the final lead quality when optimizing bids.

Manual bidding is also very important for campaigns with very little and sparse data in which case Smart Bidding simply doesn’t have enough data to make smart decisions.”

Pros of Manual CPC
  • More control over your bids
  • More transparent than Smart Bidding
Cons of Manual CPC
  • Setting maximum bids at auction-time is not possible meaning that you cannot tailor bids at search term level.
  • Requires rather advanced analysis skills to manage efficiently
  • It’s the most time-consuming bidding strategy regarding optimization and account management
  • You can only manage different bid adjustment types (e.g. device, gender, age) separately, you don’t have data on their aggregated effects
When to Use Manual CPC Bidding
  • You feel confident in yourself to create a system for setting, managing and analyzing bids manually.
  • You have a strict set maximum amount that you are willing to pay for each click.
  • You don’t have enough data in your account for Smart Bidding to work as expected.
  • You have created a new campaign and want to gather data before making any further bidding strategy decisions.
  • Manual CPC bidding is often a great option for lead generation businesses that haven’t set up comprehensive conversion tracking which incorporates lead quality.

Automated Bid Strategies

Automated bid strategies are designed to help you reach a specific predefined goal. They set maximum bids at least several times a day for your keywords to help achieve this goal. Remember, Automated Bidding is not the same as Smart Bidding. Smart Bidding is a subcategory of Automated Bidding and is discussed after these Automated Bidding Strategies.

Target Search Page Location

target search page location

Definition

“Target search page location is a type of portfolio bid strategy that automates bidding across multiple campaigns to show your ad on the top of the page or on the first page of Google search results.” (Source: Google Support)

There are two targets to choose from:

  • Top of the first search results page
  • Anywhere on the first search results page
Remember
  • Target Search Page Location bidding is not compatible with dynamic search ads.
  • Your bid is updated several times a day. (The initial update after you have set up this bid strategy may take a few minutes.)
  • Your desired placement isn’t guaranteed. Google will try to place your ads in your desired position (top of the first search results page / anywhere on the first search results page) but your final placement is determined by your quality score and competition in the ad auction.
  • There can be quite heavy fluctuations in competition causing bids to increase substantially. For example, if a new advertiser starts advertising on Google Ads, or if someone runs an aggressive short-term campaign, etc. Definitely still check your average CPCs from time to time to make sure they haven’t gotten unreasonably high.
  • You can choose to not let Google raise bids for low quality keywords (QS < 4) to avoid unreasonably high bids.
  • You can choose whether or not to let Google raise your bids if your campaign becomes limited by budget.
Pros of Target Search Page Location Bidding
  • Great for advertisers with visibility goals
  • Allows to limit raising bids in situations where a “human brain” most likely wouldn’t raise bids (QS < 4, the campaign is limited by budget)
Cons of Target Search Page Location Bidding
  • Not compatible with dynamic search ads
  • Your desired placement isn’t guaranteed
  • If not set up properly might increase bids unreasonably
  • You can only manage different bid adjustment types (e.g. device, gender, age) separately, you don’t have data on their aggregated effects
When to Use Target Search Page Location Bidding

The target search page location bidding strategy is right for you if your main goal is to be visible among top results (or on the first page). This could be the relevant choice for your own brand’s keywords for example.

This type of bidding can also be suitable for long-tail keywords that already catch searches with very high conversion intent, in which case conversion intent doesn’t need to be taken into account as much when setting the auction-time bids (so you might not need to use a bidding strategy that takes conversion intent into account – eCPC, target CPA, target ROAS). Meaning, you can be pretty sure that by just maximizing traffic from these specific high-intent searches, you’re going to get traffic with a high conversion rate.

Target Outranking Share

target outranking share

Definition

Target outranking share is a portfolio bid strategy that helps your ads outrank ads from another domain. It’s important to understand that this strategy doesn’t necessarily improve your overall ad rank; it only works to help you improve your rank in comparison to another domain’s ads. (Source: Google Support)

Remember
  • Your bids will be updated once a day.
  • As said in Google’s support materials, this bidding strategy doesn’t help you enter auctions that you normally wouldn’t be participating in. Meaning that, for example, if your competitor has keywords in their account that you don’t, you still won’t be entered in auctions for searches that are “caught” by those keywords.
  • You can only target one domain to outrank for each bid strategy.
  • It’s not always profitable to be on top positions with all your keywords so it might make sense to divide your keywords into campaigns according to which ones you want to outrank competitors with and which ones you don’t. And then only use the Target Outranking Share bidding strategy for the keywords (campaign) that you “need” to outrank specific competitors with.
  • Final placement isn’t guaranteed – it’s determined by the outcome of the ad auction, which is influenced by advertiser competition and Quality Score. (Source: Google Support)
  • You can choose if you’d want to raise bids for keywords with a low quality score (QS < 4).
  • You can enter the percentage of auctions in which you’d want to outrank the chosen domain. The higher the percentage → the bigger the needed budget.
  • You can set a maximum bid limit.
  • The Target Outranking Share bidding strategy won’t raise bids if your campaign is limited by budget. In this case, your set targeting goals may not be achieved.
  • You should allow at least 7 days for the strategy to learn and start producing results before changing any settings. (Source: Google Support)
  • If two domains are both using a Target outranking share bid strategy, the strategies will increase each domain’s bid until one of them reaches their maximum bid limit. The participant with the best bid and quality score will earn the higher ad rank(Source: Google Support)
  • The bids set by this strategy will affect your keywords’ bids in all auctions that their ads participate in, not only the ones that you and the targeted domain (that you want to outrank) participate in.
Pros of Target Outranking Share Bidding
  • Great option for making sure you are outranking your competitors with keywords that have proven to be very profitable for you (if they have proven to be profitable when their corresponding ads are shown in top positions or above competitor’s ads).
  • Often a good choice for your own brand keywords if a competitor is also bidding on your brand keywords.
  • You have the option to limit Google raising bids in situations where it wouldn’t make sense.
Cons of Target Outranking Share Bidding
  • You don’t have total control over your bids and might end up spending a lot as your bids are directly affected by the bids of the advertiser you’re trying to outrank. For example, if they decide to test an aggressive strategy and increase their bids, your bids will also be increased.
  • You can only target one domain to outrank for each bid strategy.
  • Requires supervision to avoid bids getting unreasonably high.
  • You can only manage different bid adjustment types (e.g. device, gender, age) separately, you don’t have data on their aggregated effects
When to Use Target Outranking Share Bidding

This strategy is great for advertisers who know of a specific competitor that is taking away a portion of their relevant traffic/sales. You can target this competitor’s domain and try to always show your ads above theirs.

In this case, try to do some keyword research and make sure all potentially relevant keywords are represented in your account.

Maximize Clicks

Maximize clicks

Definition

“Maximize clicks is an automated bid strategy that sets your bids to help get as many clicks as possible within your budget.” (Source: Google Support)

Remember

Before choosing this strategy, ask yourself if focusing on maximizing clicks to your website will help you achieve your larger business goal(s) or if you need to take some other factors, besides just clicks, into account as well. Businesses differ — some have the single goal of maximizing traffic to their website, some have other goals in addition to traffic as well.

In most cases, just getting more clicks doesn’t automatically mean you’d get more conversions. Often, the factors determining your traffic’s quality should also be considered.

For example, it would be rather easy to get a lot of clicks to a website when using some very generic broad match keywords. But only a very small amount (if any) of this traffic might convert. In Google’s eyes, they are doing everything right when you have chosen the Maximize Clicks bidding strategy — they are maximizing clicks to your website, nothing else is considered. But you might not be so happy. In this case, it would make sense to analyze the traffic’s quality from various keywords (at different CPC levels) and make adjustments to your bidding.

Lots of traffic doesn’t equal lots of conversions.
More like:
Lots of quality traffic equals lots of conversions.

Maximize clicks bidding can be combined with enhanced CPC which takes into account potential conversion intent when setting bids. This is covered in the next chapter.

Pros of Maximize Clicks Bidding
  • Enables you to get the maximum amount of traffic to your website for your set budget.
  • You can set a bid limit (but make sure you’re not limiting performance for no good reason)
Cons of Maximize Clicks Bidding
  • Only focuses on traffic (clicks), doesn’t take into account the quality of this traffic.
  • You can only manage different bid adjustment types (e.g. device, gender, age) separately, you don’t have data on their aggregated effects
When to Use Maximize Clicks Bidding

Maximize clicks bidding is great for advertisers whose main goal is to get as much traffic to their site as possible. Or for campaigns with the no. 1 goal of driving traffic.

Maximize clicks could also work great for long-tail keywords with high conversion intent (as explained under the target search page location section). If the search already has high conversion intent then it’s likely that you’d want to maximize traffic from these kinds of searches as much as possible.

Maximize clicks is also often used for RLSA and own brand keywords’ campaigns. Again, the reasoning behind this being that searches “caught” by keywords in these types of campaigns already have a higher-than-average conversion intent. Here’s a great article about RLSA campaigns from WordStream.

What to Consider Before Moving to Smart Bidding Strategies

The very-very first thing you need to have sorted is conversion tracking. The main component that feeds Smart Bidding’s machine learning and all its bidding decisions is conversion data. And if it’s incorrect or incomplete then it just simply can’t work as expected.

Equally as important — you need to make sure you’re counting relevant actions as conversions.

If you have set up multiple goals or conversions in Google Analytics or Google Ads, think through which ones you want to “include in the Conversions column” in Google Ads. The ones that you choose to include will be the ones that Google will try to bring you with your set target CPA or target ROAS for example.

For e-commerce sites, usually, the only conversion that is counted as the “main conversion” or “macro conversion” and is included in the Conversions column, is a purchase. All other actions that you have also defined as (micro-)conversions and you’d still like to track and base some decisions on (newsletter subscriptions, sign-ups, visiting over 10 pages, etc), should be excluded from the Conversions column and they can be viewed under the All Conversions column.

Conversions

To check all this, go to:

Google Ads > Tools > Measurement > Conversions

Measurement - Conversions

Check the “Include in conversions” column.

Include in Conversions

The conversion(s) that you have chosen to “Include in Conversions” will be the conversion that Smart Bidding will optimize towards.

Brad Geddes wrote in his article about Target CPA bidding:

“If you are going to add additional conversions for CPA bidding, make sure you really want the optimizer working off of those goals.”

You can view the performance of other conversions by adding the Conversion Action segment to your reports:

Conversion Action

Conversion Volume

There is obviously a learning period involved with each Smart Bidding strategy and the more relevant data your account has, the quicker the machines learn. So if you have very small volumes of traffic and conversions, the time Smart Bidding takes to learn can be too long to wait and manual CPC may be a better choice for you.

The number of required conversions for Smart Bidding to work well depends on:

  • the market segment you’re working with,
  • if you’re working with B2C or B2B,
  • the complexity of your customers’ journeys towards conversions

Detailed information about conversion volumes for each bidding strategy is listed under their specific chapters below.

Attribution Model

Before diving into the world of Smart Bidding, acknowledge your choice of attribution model again. As explained earlier, you want to attribute credit (revenue and conversions) to all relevant touch-points that led the potential customer to your website and ultimately — to a conversion.

Conversion Tracking (Google Analytics vs. Google Ads)

In most cases, it’s beneficial to use Google Ads tracking as this takes into account the total effect of Google Ads and Smart Bidding has more conversion data to work with.

Explanation:

If you’re importing transactions from Google Analytics to Google Ads and the person’s path to purchase (his/her points of contact with your website) looked like this:

Conversion Funnel

In this case, the conversion is entirely attributed to “facebook.com / referral” in the non-MCF reports in Google Analytics as it was the last non-direct click before the conversion happened. Even though Google Ads brought the person to your website two times, it doesn’t get any credit. Find out more about the differences between Google Ads conversion tracking and Google Analytics tracking here.

NB! The attribution model you choose inside Google Ads only distributes the conversion and revenue between Google Ads keywords if the conversion was attributed to Google Ads in the first place. Meaning that Google Ads was the last non-direct click before the conversion.

However, if you use Google Ads conversion tracking then Google Ads gets credit for the conversion no matter which touchpoint it was before the purchase (first or second or fifth or last, etc). This way Google’s Machine Learning (which is used for Smart Bidding) has more data to base its decisions on and it can evaluate the total effect of Google Ads more adequately.

You also need to decide which conversion actions you want Smart Bidding to optimize towards. If you have set up multiple goals or conversions in Google Analytics or Google Ads, think through which ones you want to “include in the Conversions column”. The ones that you choose to include will be the ones that Google will try to bring you with your set target CPA or target ROAS for example.

About Google Ads & Machine Learning

The main concept behind Google’s Smart Bidding is that it has access to more information (than us humans do) and Google’s almighty all-knowing brain can make more informed decisions and predictions about the most optimal bid for each situation. So it only makes sense to “feed it” as much information about your site’s conversions and people’s preceding behavior as possible.

Smart Bidding changes your keywords’ maximum CPC bids for every single search (auction) whereas manual bidding sets one maximum CPC for all searches that trigger your keyword. This is referred to as auction-time bidding — a suitable bid is determined at the time of each auction to best reach the goals that you have set, taking into account the user signals and their different combinations’ likelihood of converting.

Signals that Google uses for adjusting bids (the list is constantly evolving):

  • Device
  • Physical location
  • Location intent
  • Weekday & time of day
  • Remarketing list
  • Ad characteristics
  • Interface language
  • Browser
  • Operating system
  • Demographics
  • Actual search query
  • Search Network partner

Find out more about user signals that are used by Google’s machine learning here.

Here’s a great (& entertaining 😛 ) video that very generally explains how machine learning works:
https://www.youtube.com/watch?v=R9OHn5ZF4Uo

After watching the video you can decide for yourself if you’d trust machine learning to be smarter than you when it comes to setting the most optimal keyword bids.

Other experts’ videos discussing machine learning (in a bit more serious manner):

1) Man vs Machine: How to Future-proof Your PPC Job – Brad Geddes:

2) Academy on Air – Introduction to Machine Learning – Google

Smart Bidding Strategies

Google’s Smart Bidding strategies use machine learning to make bidding decisions that should bring you the most optimal results within your predefined goals. It uses:

  1. Auction-time bidding
  2. Adaptive learning at the query level
  3. Richer user-signals and cross-signal analysis

What makes Google Ads machine learning unique is the fact that it uses query-level data across similar auctions. This way, the algorithms that determine your bids have way more precise data to base their decisions on. This approach also helps reduce performance fluctuations when keyword-level conversion data is scarce.

Another small benefit is that if you restructure your account (e.g. move keywords to a different ad group/campaign) then the algorithms don’t need to start the learning process all over again. As learning happens at the query level, the location of a keyword in your account doesn’t affect the data Google has about it. If a search query has already matched a keyword in your account, Google already has data about it and can apply its learnings from historic performance to the decisions that are made for the newly located keyword. (Source: Setting Smarter Search Bids. Inside Automated Bidding with AdWords. Google 2018)

Enhanced CPC

enhanced cpc

Definition

Enhanced CPC is a great stepping-stone for starting with Smart Bidding. “Enhanced cost-per-click (ECPC) helps you get more conversions from manual bidding. ECPC works by automatically adjusting your manual bids for clicks that seem more or less likely to lead to a sale or conversion on your website. ECPC is constrained by your max CPC bids when optimizing for conversions” (Source: Google Support)

“ECPC looks for ad auctions that are more likely to lead to conversions and then raises your max CPC bid (after applying any bid adjustments you’ve set) to compete harder for those clicks. If a click seems less likely to convert, Google Ads will lower your bid. ECPC will try to keep your average CPC below the max CPC you set (including bid adjustments), but may exceed your max CPC for short periods of time.” (Source: Google Support)

Remember

Enhanced CPC is constrained by the max CPC bid limit you’ve set. So while it does raise bids for clicks more likely to convert, it has a limit it can’t exceed. Also, don’t forget that Enhanced CPC takes bid adjustments that you’ve set into account before determining final bids.

When deciding bids, ECPC automatically considers different conversion rates of the potential traffic (mobile devices’ bids are set separately, though). So no additional bid adjustments (besides mobile) are necessary for ECPC to maximize conversions. But if you have proven data that some types of traffic work especially well, you can choose to set bid adjustments and they will be applied on top of ECPC’s automatic adjustments.

Remember that Enhanced CPC bidding bids towards the likelihood of a conversion (other Smart Bidding strategies optimize towards the final conversion).

When you first enable Enhanced CPC bidding, the campaign will have an approximately 50/50 budget split between using manual bids and a fully automated bid (that tries to deliver more conversions at manual bid’s CPA). Once the fully automated bids’ performance is proven, the campaign will take more towards the fully automated bidding.

Required Conversion Volumes for Enhanced CPC Bidding

Using Enhanced CPC bidding does not require having any prior conversions in your account.

Pros of Enhanced CPC Bidding
  • Enabling the enhanced CPC option is a great way to test what Google’s machine learning can do for you while still maintaining control over the average maximum CPC (and bid adjustments).
Cons of Enhanced CPC Bidding
  • If you have set the max CPC limit very high, you should beware of potentially higher costs than usual as Google can be quite bold when setting bids during the learning period (when estimating the likelihood of a conversion from the given click). As already mentioned several times before, the less data you have the longer Google’s machine learning learns. In the meantime, you are quite literally paying more because of its mistakes.
  • ECPC bidding is not available for App Campaigns (previously Universal App Campaigns).
When to Use Enhanced CPC Bidding

As mentioned in Google’s support materials, Enhanced CPC is a great stepping-stone to move from manual bidding to Smart Bidding. It combines them so that conversion probability is taken into account when setting bids but you maintain control over the maximum bid you’re willing to pay.

Enhanced CPC is a great option for when you don’t have enough conversion volume or data in your account to fully leverage other Smart Bidding strategies. You can join forces with Google’s machine learning and work together towards increasing conversions. You need to acknowledge though, that Enhanced CPC bidding also takes time to learn so it might not bring you the best results right away (as with all other Smart Bidding strategies).

Maximize Conversions

maximize conversions

Definition

“Using historical information about your campaign and evaluating the contextual signals present at auction-time, Maximize Conversions bidding automatically finds an optimal CPC bid for your ad each time it’s eligible to appear. Google Ads sets these bids to help get the most conversions for your campaign while spending your budget.” (Source: Google Support)

Remember

One of the key things to take note of here is that maximize conversions bidding will attempt to fully spend your daily budget. If your campaign previously had a budget that wasn’t being fully spent, you might be in for an increase in costs. This is also a reason why campaigns with maximize conversions bidding can’t be using a shared budget — they must have an individual budget.

Required Conversion Volumes for Maximize Conversions Bidding

The following information was provided by a Google representative via email upon request:

No starting requirements (can be used from the start of the campaign), yet, here are some best practices:

Clicks: We [Google] recommend campaigns have at least 20 clicks/day.

Conversions: Any account with conversion tracking can utilize Maximize Conversions from the start of any search campaign, and it is the default bid strategy in the Google Ads front end for all new search campaigns for conversion tracked advertisers. Therefore, there is no requirement for conversions before adding Maximize Conversions. Make sure the right conversion actions are included in the “Conversions” column by using the “Include in Conversions” setting.

Budget: The campaign should already be spending most of the existing budget, as Maximize Conversions will spend the entire budget once enabled. If the campaign is spending far below the budget before opting in, spend will increase significantly and CPA may also increase even though conversion volume will also increase.

Impression Share: The campaign should have additional available inventory (imp. share <90%) before switching to Maximize Conversions so that the algorithm has headroom to increase volume. If the campaign is already maxing out on impression share, there is less value to adding Maximize Conversions to drive conversion volume. You can also look to add additional broad match keywords before opting into the strategy to increase available inventory if needed.

Pros of Maximize Conversions Bidding
  • Requires the least historical conversion data (compared to target CPA & target ROAS bidding)
  • Bids are optimized at the search query level, not keyword level
  • Search context is considered to predict the likelihood of a conversion from the click (bid set accordingly)
  • Bids are adjusted for every auction
  • User’s historical behavior is taken into account when determining bids.
  • Different signals’ performance is aggregated and bids are set accordingly (with manual bidding you can only look at all signals e.g. device, gender, age separately)
  • Has access to auction-time signals that are only available to Google (browser, language settings, operation system, app, actual query, ad creative)
Cons of Maximize Conversions Bidding
  • Google will attempt to fully spend your budget while maximizing conversions so you might see a significant increase in costs (if you budget hasn’t been fully spent before).
When to Use Maximize Conversions Bidding

The Maximize Conversions bidding strategy could be the one for you if you haven’t defined a set target CPA or target ROAS and your goal is to drive more conversions within your current budget. It is worth trying if your campaign is limited by budget and you have <90% impression share.

If the amount of conversions fluctuates (also fluctuating your CPA and ROAS) it might sometimes be best to just try to get the maximum out of your campaign → use Maximize Conversions bidding.

But if you do have specific ROI goals — a target CPA or ROAS, it is recommended to use Target CPA or Target ROAS bidding.

Here’s a great article about the maximize conversions bidding strategy by Dan Roberts from PPC hubbub.

Target CPA

target CPA bidding

Definition

“Target CPA is a Google Ads Smart Bidding strategy that sets bids to help get as many conversions as possible at the target cost-per-acquisition (CPA) you set. It uses advanced machine learning to automatically optimize bids and offers auction-time bidding capabilities that tailor bids for each and every auction. Target CPA is available as either a standard strategy in a single campaign or as a portfolio strategy across multiple campaigns.” (Source: Google Support)

When deciding on a suitable bid, Target CPA bidding factors in:

  • Contextual signals at auction time
  • Search query level performance across your account
  • Predicted conversion rate
  • Your inputted Target CPA
Remember

So basically, bidding-wise, all you need to do is define your target CPA — cost per acquisition (or multiple different CPAs that you can set at ad group level) and let Google optimize your bids (at auction-time) to achieve it.

You do need to be realistic though. It’s recommended to start with a bit of a higher CPA than your ideal to allow more “spending room” for the learning process. This way Google can bid more aggressively at the start of your campaign to get more traffic and eventually if it has enough data, you should get more conversions at a lower CPA as well.

If you set your target CPA too low, you’ll be missing out on a lot of potentially converting clicks because Google will try to be extra cautious with spending your budget and setting bids (to avoid exceeding the set target CPA). Here’s a summary of testing different target CPAs and how they affected actual CPAs and conversion volumes by Kristina McLane from Hanapin.

Again, it’s recommended to first test it with an experiment so you can compare performances and decide for yourself if target CPA is the strategy for you.

For Target CPA bidding to work best, you should set a daily budget that is >10x the Target CPA bid. This way the campaign has sufficient room to explore all hours of the day. To control volumes, use your CPA bid. A budget >5x the Target CPA is still good, the absolute minimum daily budget should be 2x your Target CPA.

If you have significant variations between different ad groups’ actual CPAs, you should consider using Target CPA on the ad group level.

Required Conversion Volumes for Target CPA Bidding

The following information was provided by a Google representative via email upon request:

“A campaign must have had 15 converted clicks in the last 30 days to start a new Target CPA/CO campaign. This does not mean that conversion history needs to be at least 30 days long, but it is the minimum amount of data that the system needs to predict bids.

The campaign also needs to have had 10 converted clicks in the last 30 days as we would have measured it five days ago. This is done to prevent the situation where you get more than 15 conversions in just an hour or two and then opt-in to Conversion Optimizer [Target CPA], as that would not give the tool enough data to work with. For example, if you just set up the campaign and had all your conversions yesterday, you would not qualify. Think of it as going backwards in time five days and then asking yourself: Have I had 10 conversions in the last 30 days?

Also, note that conversions acquired from CPM or vCPM bidding strategies do not count towards eligibility. The Campaign will need to get an additional 15 conversions while on a CPC bidding strategy because the infrastructure is not in place to learn from the bidding of these conversions.”

Here’s an explanation of recommended conversion volumes (on the campaign level) for using Target CPA bidding, how they affect performance and duration of the learning period.

Recommended Number of Conversions
(Setting Smarter Search Bids. Inside Automated Bidding with AdWords. Google 2018)

NB! As of H1 2019, it is stated by Google that enabling Target CPA bidding no longer requires prior conversion data. It still works better if there is prior data available, but it is possible to choose Target CPA bidding from the very start of the campaign. The previously mentioned required conversion volumes are still recommended.

Pros of Target CPA Bidding
  • Bids are optimized at the search query level, not keyword level.
  • Search context is considered to predict the likelihood of a conversion from the click (bid set accordingly).
  • Bids are adjusted for every auction
  • User’s historical behavior is taken into account when determining bids.
  • Different signals’ performance is aggregated and bids are set accordingly (with manual bidding you can only look at all signals e.g. device, gender, age separately)
  • Has access to auction-time signals that are only available to Google (browser, language settings, operation system, app, actual query, ad creative)
Cons of Target CPA Bidding
  • Doesn’t consider conversion value (revenue). (But there’s a different bidding strategy for that — target ROAS.)
  • Performance can fluctuate a lot if your conversions are distributed between a lot of different keywords (search queries). E.g. You have 500 keywords in your account with a total of 50 conversions so that 50 keywords out of your 500 all have 1 conversion. (And next month maybe a different 50 will get conversions.) So the more “stable” your conversions are, the more stable your CPA will usually be and the better the machine learning will work (as it has repetitive performance data to base future decisions on).
When to Use Target CPA Bidding

Target CPA bidding is often used by advertisers who are trying to collect leads and have calculated their ideal CPA that they’re willing to pay for one lead.

Or by advertisers who have products with a similar profit margin in one campaign and are willing to invest a similar amount to drive a purchase from that campaign.

Target ROAS

target ROAS bidding

Definition

“Target ROAS lets you bid based on a target return on ad spend (ROAS). This Google Ads Smart Bidding strategy helps you get more conversion value or revenue at the target return-on-ad-spend (ROAS) you set. Your bids are automatically optimized at auction-time, allowing you to tailor bids for each auction. Target ROAS is available as either a standard strategy for a single campaign or a portfolio strategy across multiple campaigns.” (Source: Google Support)

When deciding on a suitable bid, Target ROAS bidding factors in:

  • Contextual signals at auction time
  • Search query level performance across your account
  • Predicted conversion rate
  • Your inputted Target ROAS
  • Predicted conversion value per click
Remember

Out of the previously mentioned Smart Bidding strategies, target ROAS bidding requires the most historical conversion data to work well.

You need to be realistic when setting the ROAS goal and give Google time to learn and reach a state of constantly bringing in revenue with the same ROAS. The more conversions you have and the more stable they are (keep coming in at similar volumes) the easier it is for Google to “maintain the flow”.

Also, remember that different campaigns, keywords or product groups might have a different ROAS so it makes total sense to have several different target ROAS goals within one account.

For example, your brand keywords will probably have a much higher ROAS than generic keywords (the so-called “first step keywords” that bring potential customers to your website in the first place when they are still at the top of their purchasing funnel). So it makes sense to have different goals for these differently behaving keywords.

Another thing to take note of is that bid adjustments aren’t used with Target ROAS bidding. You don’t need to remove them if you have used them before, they just won’t be used anymore once you switch to Target ROAS bidding. This is due to the fact that auction bids are set according to real-time data. You can, however, set a mobile bid adjustment of -100% if you want to exclude all mobile traffic.

Required Conversion Volumes for Target ROAS Bidding

“To use Target ROAS bidding, your campaign must have at least 20 conversions in the past 45 days. To maximize results and give machine learning algorithms enough data to make informed bidding decisions, we recommend that you have at least 50 conversions in the past 30 days. It also helps if your campaign has received conversion values at a similar rate for at least a few days.” (Google’s support materials)

Here’s an explanation of recommended conversion volumes (on the campaign level) for using Target ROAS bidding, how they affect performance and duration of the learning period.

Recommended Number of Conversions for Target ROAS
(Setting Smarter Search Bids. Inside Automated Bidding with AdWords. Google 2018)

Pros of Target ROAS Bidding
  • Bids are optimized at the search query level, not keyword level
  • Search context is considered to predict the likelihood of a conversion from the click (bid set accordingly)
  • Bids are adjusted for every auction using real-time data
  • User’s historical behavior is taken into account when determining bids.
  • Different signals’ performance is aggregated and bids are set accordingly (with manual bidding you can only look at all signals e.g. device, gender, age separately)
  • Has access to auction-time signals that are only available to Google (browser, language settings, operation system, app, actual query, ad creative)
Cons of Target ROAS Bidding
  • Requires enough historical data and conversion volumes to work efficiently and constantly bring revenue at a stable ROAS level.
When to Use Target ROAS Bidding

Target ROAS bidding is, by its nature, suitable for e-commerce businesses — define your desired ROAS and let Google do the work to achieve it.

It’s reasonable to at least test target ROAS bidding if:

  • Your conversions vary in value and your goal is to maximize conversion value within a set target return on ad spend
  • You have a large volume of conversions (absolute minimum being 15 conversions/month in Search/Display or 25 conversions/45 days on Shopping campaigns)
  • You send revenue data to Google Ads
  • You have a specific ROAS goal

To quickly summarize all bidding strategies mentioned so far:

Google Bidding Strategies

Changing Your Bidding Strategy

If you are thinking of changing your bidding strategy it’s recommended to set up the change as an experiment so you’d have side-by-side comparisons of the performance of two bidding strategies.

Here’s an example of how you’ll see the comparison between your original campaign and experiment. The displayed larger numbers are those of the experiment and below them, you can see the increase or decrease percentage compared to your original campaign settings.

Experiment Results

When you hover over the KPI (e.g. Conversions), you’ll be displayed more detailed information:

Statistical Significance

The key thing to remember — allow the experiment enough time to run. All experiments should run at least as long as there are statistically relevant results and enough conversion volume available to make final conclusive decisions. Google shows you when statistical relevance has been reached by adding the blue asterisk. Also, make sure to take your conversion lags into account before assessing final results.

We have often seen that the performance of a newly added smart bidding strategy is initially worse than your original settings but it can perform much better after the learning process is complete. So give it time, observe and when the time is right, make well-informed data-driven decisions.

Other Methods for Managing Bids

There are other methods for managing bids as well besides all the tactics mentioned in the previous chapters.

One of the key things to remember when managing bids with the help of any of the following methods is that you shouldn’t change the bids too often. All types of bidding need time to work, gather data and only after you have seen the effect of the previous change, should you move on to making the next changes.

For example, if your typical conversion has a lag of 7 days then you should wait at least seven days to assess the effect of your last bidding change. Otherwise, if you change bids daily or hourly, none of the changes you make have enough time to take effect and you might end up lowering bids unreasonably.

Bid Adjustments

Bid adjustments enable you to have different bids based on:

  • Device
  • Location
  • Weekday
  • Time of day
  • Top content (Display Network, YouTube)
  • Targeting method
  • Remarketing lists for search ads (Search Network)
  • Interactions (call adjustments)
  • Demographics

You can analyze historical performance data from the reports found in Google Ads and set the optimal increase or decrease for bids. Multiple bid adjustments are typically multiplied together to get the “total adjustment”. The maximum allowed increase for multiplied bid adjustments is 900% and the minimum multiplied decrease in -90%.

To find out more about bid adjustments check Google’s support materials.

Bidding Rules

Bidding with the help of rules takes advantage of Google Ads’ Automated Rules. Automated Rules enable you to create specific conditions and if they are matched, a set of actions will follow. Brad Geddes said in his article about streamlining bidding with Automated Rules:

“Automated rules are best used in place of repeated basic actions in response to a certain threshold of data.” He also highlights some good points on what to be careful with when setting up your Automated Rules.

An automated rule can be for example:

“If a keyword has brought 5 conversions in the past 7 days AND its average position is 2.5 or lower → increase its bid by 10%”.

The options are endless. It’s crucial to thoroughly understand what you’re doing and still monitor results on a regular basis to avoid bids getting out of control. And as also mentioned in the introduction, don’t run these rules too often or without having a clear and thought-out bid management system, goals and strategy in place.

This is especially important for any rules that lower bids. You could easily end up with too low bids for keywords if, for example, you want to lower not-so-good performing keywords’ bids but you run this rule too often. So the lowering of bids would not have had time to take effect and the bid would be constantly re-lowered.

For example, if you have a rule:

“If a keyword has spent >100€ in the last 7 days AND has brought <2 conversions AND the average position was 1.5 or higher → lower bid by 30%.”

If you run this too often (e.g. every hour) and performance data doesn’t gather quickly enough, which is most likely always the case for periods of 1 hour, the effect of lowering the bid the first time hasn’t had time to really make a difference yet and you’ll just keep lowering it until it has become too low.

Here are detailed instructions for setting them up and some common ways automated rules are used.

Bidding Scripts

Bidding can also be managed with the help of scripts. Scripts are used for many other things besides bid management as well.

Basically, all you need to do is

  1. upload the script to your account (Tools > Bulk Actions > Scripts > +Add Script > paste the script you’ve found),
  2. enter your account’s specifics and details in the script where marked,
  3. authorize it

and you’re good to go.

Sophie Howell from Koozai has compiled a list of 100 scripts to check out. This list would be a reasonably good place to start to get an idea of what’s out there.

Summary

As you just read, there are loads of different options for managing your bids but the key thing about them all is to take your time, understand how they work and why they work the way they do. Also, don’t stay content with the option you may have chosen, say, a year ago. Keep on tweaking it and testing further to find the most optimal option for you. Baring in mind, that option might also be a constantly changing one.

Start off by defining your main goals on account and campaign level to get an idea of which types of bidding would fit them best. Then set specific KPI goals (e.g. target ROAS, target CPA, traffic volumes, etc) for your campaigns or ad groups, monitor performance and make adjustments based on statistically relevant data as it accumulates.

Do let us know in the comments below what your experiences have been with different bidding methods — what has worked best for you (+ add context), what option has disappointed you, etc. This way we can keep adding valuable insights to this post and make it as useful of a resource as possible.

Menu
60 Shares
Share
Share
Pocket
Buffer
Tweet