Bidding strategies and why Target ROAS is the best

Target ROAS is the best bidding strategy as it leverages Google’s ability to assign a probability of a user converting: bidding more for high converters and less for low converters (plus it also takes into account the value of that conversion).

Review of bidding strategies

Why Manual CPC doesn’t work so well

Let’s take a fictional high-performing keyword that has been on a Target ROAS bidding strategy of 500%.

Here’s a summary of how that keyword has performed over the last 6 months:

The keyword has delivered £83,032 of conversion value (revenue) with an average revenue per click of £24.83.

The ‘Search exact match IS’ for this keyword is 70% (i.e. 30% of the time an ad didn’t show for this keyword either because Google didn’t bid or didn’t bid high enough).

You think you can drive more volume through this keyword by changing the bidding strategy to Manual CPC with a max. CPC of £5.

(For a click costing £5 you hope to get £25 in revenue ⇒ 500% ROAS. This is a max. CPC, so you hope you’ll scoop up a lot of lower cost clicks as well – still at a revenue per click of £25 – so perhaps you can drive more revenue and a higher ROAS than 500%?).

However, after running on Manual CPC for 6 months you find the performance has dropped:

Fewer clicks
Much lower revenue per click
Lower ROAS
Much lower Conversion Value
Much lower CPC

What happened?

If we segment the clicks into ‘CPC bands’ then the performance of the Target ROAS campaign might look something like this:

The first thing to note is the huge spread in CPCs paid – reflecting the quality of the user (and that Google can bid very aggressively if it thinks someone has a high chance of converting).

Secondly, there is broadly the same ROAS for all CPCs between £2 and £9. As the CPC goes up so does the revenue per click ⇒ steady ROAS (the ratio of revenue to cost is steady).

When we swapped to the Manual CPC campaign with a max. CPC of £5 we cut off all the orders that have a CPC above £5.

In addition, we are now bidding more aggressively/over-bidding in all bands under £5 (including bidding on some users that we weren’t bidding on at all previously).

Users we might have previously bid for in the £2 – £4 bands we are now paying over £4 for, so the performance in the £4 – £5 band has dropped too:

To improve the manual CPC performance you have to make bid adjustments – telling Google to bid more or less for device, location, time of day, demographic data etc. about a user.

The number of adjustments explodes when multiplied by keywords – and Google only lets you make adjustments on some of the criteria that it can use (for example it can see prior web visits) when it sets a bid using a smart bidding strategy.

In summary, doing Manual CPC well requires a lot of time and effort in making adjustments to bids, and all this effort only partially replicates what Target ROAS is doing automatically.

Want to see if ROAS is steady across CPCs for your data?

Here’s how to look at your ROAS by CPC band for an Ad Group in one of your campaigns:

Go to Reports and create a new report with the following dimensions and metrics:

Dimensions of:

Ad group;
Day;
Hour of the day;
Device

and Metrics of:

Clicks;
Cost;
Conv. value;
Impr.;
Impr. (Top) %;
Impr. (Abs Top) %;
Search exact match IS

And then filter for:

an ad group with a decent volume of clicks but ideally has a single exact match keyword in it (or as close to that as you can get).
Network (with search partners) = Google Search.

If you download the data and filter for hours where there was only one click (we need single clicks in an hour so we can directly map the cost of a click to a conversion) and assign each click to a CPC band.

Then use a pivot table to create a table as per the examples above.

You should see that, as per above, the ROAS stays fairly steady as the cost per click grows – if you are using Target ROAS as the bidding strategy.

[mapflo is a data analysis tool and is particularly good for sorting data – give it a try].

If you have been using a bidding strategy other than Target ROAS (or Target CPA) you will still see that the more expensive clicks will typically have a higher revenue per click (the click cost of a particular user will be a function of how competitors are bidding) but you probably won’t see ROAS stay steady as CPC increases. It’s more likely that you’ll see the cheaper clicks driving better ROAS (but you’ve probably overpaid for these even).

We’ve added in ‘Search exact match IS’ so you can see for exact matches what visibility you are getting for that keyword – if it is a keyword with an inherent ROAS much higher than the Target ROAS for the campaign then you should see a high ‘Exact match IS’ percentage and the ratio of Abs Top % / Top % to be high (i.e. your bid nearly always comes top in the auction).

What about ‘Enhanced CPC’ bidding strategy?

This is a halfway house where you set a max. CPC for a keyword but then allow Google to apply some intelligence to adjust that bid for users with a higher probability of converting.

If you are trying to do Manual CPC properly – i.e. set very granular bid adjustments for location, demographics, device, audience etc. then it seems strange to then let Google apply its own adjustments (if you use bid adjustments with enhanced CPC then Google will apply its adjustments on top of those) – enhanced CPC should do better than manual CPC bidding but why not let Google decide the value of a user without tying one arm behind its back?

Target Impression Share – Beware!

If we have bought into the logic of paying more for higher converting clicks then the same logic makes it clear that Target Impression Share delivers the exact opposite: paying more for lower converting clicks.

Let’s say we set a goal for Google to achieve 50% of impressions ‘absolute top of the page’ then the easiest users for Google to get top of the page are the users least likely to convert.

Google will start disproportionately pushing poorer-performing devices, locations, demographics etc. This isn’t going to end well…

Target Impression Share for Brand terms

In fairness, Google only suggests using Target Impression Share for Brand terms, where you may want to have a top impression nearly 100% of the time.

It is likely that your brand terms (such as your company name or names of leading own-brand products or sub-brands) will have the best ROAS of your keywords – but there will still be the variation in likelihood to purchase and Target Impression Share still prioritises the lower quality users.

Some companies fall into the trap of thinking ‘we must be top for our own-brand terms 100% of the time as each visit is so valuable’ – but you might end up paying £5 for each click versus £0.20 with a sensible Target ROAS strategy – with very little increase in traffic (and remember if people want to come to your site then they will click on a lower ranked ad or on your organic link, especially if that is top of the organic listings which it is likely to be for your brand terms).

Your CPC for your brand terms should be relatively low – lower than your non-brand terms.

You should be getting excellent click-through rates even if you are not absolute top and your ad quality should be high as you are the most relevant destination for the user.

If you really want to use Target Impression Share for brand campaigns then we recommend having a max. CPC bid in place and running an experiment with one campaign having a lower maximum bid than the other to find the sweet spot that drives maximum CM3.

However, the rule of ‘clicks not being equal’ still applies to brand terms so we think a better approach is to use Target ROAS with a max. CPC – that aligns with the ability to bid more on the users most likely to convert. You can then use an experiment to find the optimal Target ROAS / max. CPC combination.

Maximise Conversion Value Strategy

Maximise Conversion Value (without Target ROAS) will spend all your budget each day while trying to get as much conversion value as possible for that spend.

This sounds reasonable, and a plus is that it uses Google’s Smart bidding to identify users with a higher probability of converting.

However, think of this strategy as ‘spend all your money’ – as that is what it optimises to. Regardless of the search volume on a given day, it will aim to spend all your money.

On days when there is more opportunity to drive conversions than you have the budget for then your ads will stop running and you’ll miss revenue-earning opportunities (but the revenue you do win will be at an efficient ROAS).

On days where there is little opportunity to drive conversions then Google will spend all your budget anyway and your ROAS will be terrible (plus likely to be some eye-wateringly high CPCs – if there aren’t many clicks to be had then it’ll need to pay a high CPC to achieve its goal of spending all your money).

To do well with this strategy you need to estimate in advance what the likely demand will be on each day and adjust your budget such that the spend and demand match and you hit your Target ROAS without leaving opportunities on the table. Hmm – perhaps you should just use a Target ROAS strategy?

The only time this bidding strategy makes sense is if you have a very constrained budget and even on slow days your spend will be well below the volume of opportunity or if you do not have sufficient conversion volume for Target ROAS to work.

You can also use it when launching a new campaign when you are not sure how the campaign will perform and want to tread carefully – but we think the better option is to launch with Target ROAS and a constrained daily budget so Google can’t spend wildly. You can adjust the ROAS Target to drive sensible volume for Google to learn audiences and conversion rates.

>>> Read the next article: Keywords have a relatively fixed ROAS

>>> Back to main menu

Please be really careful making changes to your Google Ads account

Google doesn’t always respond how you (or we) think it will. The way we think about Google Ads may not be the best set-up for your account.
Only change one thing at a time.
If possible, always use an experiment to test a change – particularly for significant changes such as moving bidding strategy to Maximize conversion value (Target ROAS).
Protect your financial downside by testing with limited spend in the experiment/change. Note that moving to a smart bidding strategy requires a learning phase where Google may not be efficient.
Be careful if adding/removing primary conversion actions – changing what Google is converting to can radically change what and who Google targets and how much it’s willing to spend.
Remember, all changes to your account are at your own risk. Mapflo shall not be liable for any damages; losses; lost revenue or lost profit.

Glossary of Terms

AOV = Average Order Value

CM1 = Contribution Margin 1 = revenue minus COGS (cost of goods sold) in an order.

CM2 = Contribution Margin 2 = margin on an order after all costs directly attributable to that order such as COGS, shipping, payment fees, customer service etc. (except for marketing).

CM3 = Contribution Margin 3 = CM2 less marketing spend. An ‘Estimated CM3’ value uses an assumed CM2 %.

CPA = Cost Per Action. In this report taken to mean cost per conversion or cost per order.

Keywords = words or phrases (assigned to an ad group) that match a user’s search term and trigger Google to bid to show an ad.

Lifetime CM3 = CM3 from all orders (or subscription payments) for a customer.

Profit = CM3 less all fixed overheads (such as salaries and office rent). Hence Optimising CM3 also optimises profit at the same cost base

ROAS = ‘Return On Ad Spend’ = conversion value divided by cost. A ROAS of 400% means you get four pounds of revenue back for every pound of ad spend.

Search term = the word or phrase that a user searches for on Google.

Additional Insights

Use mapflo instead of a spreadsheet for data analysis

Coming in 2023.