I’ve managed a lot of Google Shopping Ads Campaigns that have brought in Millions of pounds for the businesses I work for, so when people ask me “are Google Shopping ads Worth it” I have to tell them, yes they are, if you know what you’re doing. If you run an ecommerce website and you’re looking to run Google shopping ads, this is a no-brainer. I’ve yet to see a campaign that didn’t return at least 6x return on ad spend.
What are Google Shopping Ads?
They are ads that show at the top of the search results when people are looking to buy products online. They contain an image of the product, the name of the company selling the product and possibly some promotional message. They rae used by Ecommerce businesses to sell their products.
Google Shopping Ads Setup
First up, if you run an ecommerce site and you are running search ads but not shopping ads, you’re leaving money on the table. Done properly, the majority of your online sales will originate from shopping ads.
But there are a number of technical hurdles to deal with before your ads can run.
You need to set up a Google Merchant Centre account. Click here to sign up.
Use the same Google account information that you use for your Google Ads account. In fact, life will be much simpler if you use the same Google account for all of the services that they provide for each website that you own.
So you want to be able to use the same login info for:
- Google Tag Manager
- Google Analytics
- Google Search Console
- Google Merchant Centre
- Google Adwords
- Google Sheets
- Google Data Studio
If you don’t use the same login credentials, you’ll find that you have to add users for other email addresses that you use for any one of theses services to each account where the login info is not common. It gets messy. So do it right from the start.
What does Google Merchant Centre Do?
You’ll input some basic information about your website and your business, but the main function is to process a product feed that you will upload. The product information in that feed is usually set to update daily to take account of your stock changes.
How that feed is generated depends on your shopping cart and each cart will generally have a specific method to generate the feed. More often than not, this is done via an extension or a plugin. Your website developers are best placed to advise you on how to generate your product feed.
The feed contains all the data that Google needs in order to show your products in the search results. The quality of your feed will have a direct impact on the results that you get from running shopping campaigns, so it’s important that you spend the time required to make sure that your feed is optimal.
Once your feed is uploaded to your Merchant Centre, Google will validate it. That means that when you log in to your Merchant Cantre, you’ll see any issues that Google has with processing your feed. Solve those issues before you go any further, either with the help of your developer, or if they lack the technical knowlege to do so, with a specialist.
Set up a Google Shopping Campaign in Your Adwords Account
This bit is quite simple. Just set up a new campaign and follow the instructions. You can’t go wrong. I usually just create one ad group and sub-divide by brand. I know there are more complicated ways of splitting this up that are recommended elsewhere, but I like to keep it simple.
Setting up Tracking and Conversions
I’ve seen dozens of ecommerce shopping campaigns in Google Ads where there was either no ecommerce tracking in place, or where there was but no conversion values were being recorded.
And that just doesn’t work.
Often the workaround that’s employed is to work with the cost per conversion numbers. I’ve done that in the past and I have to say, it’s hard to make a profit on that data.
You need both conversions and conversion values recorded accurately to make this work.
That needs a bit of coding work from your developers to push the data to your Google Ads campaign. Any competent Ecommerce developer will have implemented this many times and doing that for you should be no more than a few hours work. If they can’t handle it, find a freelancer who can.
Sidenote: In any retail market online you will face up to 60 competitors who are all trying to get a piece of the market. Some of them will be very large companies with very large budgets, but don’t let that deter you. It is possible to beat them, Amazon included.
There does exist however both a virtuous and declining circle and you have to be on the right side of that to make things work.
In short, the more granular you are in your optimisation, the less money you’ll waste and you’ll obtain a higher return on ad spend.
That forces your less well optinised competitors to bid higher to compete with you, so reducing their margins.
The businesses who win at this game define a targeted return that’s a compromise between volume and profit. Then they optimise every available variable in their campaign to hit that target.
The difference between a poorly optimised camapign hitting 5x return and a very well optimised campaign hitting 9x return; is the difference between not knowing if you’re still in business at the end of the month, and predictably forcasting cash flow and profits.
Highly optimised campaigns win every time.
Now for the fun bit.
How to Optimise a Google Shopping Ads Campaign
1) Set a Brand Level Bid – just set everything to 15p. You won’t go far wrong and you’ll start getting some data
2) Sub-divide each brand down to item ID level. Which means if you sell 100 different items of a brand, you might have 400 items in your sub-division i.e. 100 items x 4 sizes per item.
3) Set a target return on a spend – I recommend 7x
4) Check your campaigns daily and adjust bids based on 30 day data to acheive your target.
5) Adjust all other variables to acheive your targeted return
That’s the basics, but the devil, as always is in the detail – how exactly do you fully optimise a Google Shopping Ads Campaign?
Here’s my daily process:
Open up the campaign and set the timescale to yesterday.
For any brand where there’s a conversion, open it up and find the items that converted, click the checkbox nest to them.
Set the timescale to last 30 days
With Conversion value/cost as the metric to pay attention to, adjust the bid to acheive your targeted return on ad spend. So, for example, if an item has a conversion value/cost of 6.3 and a bid of 35p with an actual click cost of 35p – I’m going to reduce the bid to 31p to acheive a 7x return on ad spend. And if the conversion value to cost is 16.3 with a bid of 35p and an actual click cost of 35p; I’m going to set the bid to 70p in the knowlege that I can pay twice as much for a click and still hit a 7x return.
That’s it roughly, but there are subtleties:
If the impression share is low, I’ll bid higher and if it’s already high, I’ll be wary of bidding more when I’m already highly visible.
One conversion in the last 30 days could be an anomaly, so I’ll tread carefully adjusting bids either up or down.
You get a feel for what works with specific campaigns over time.
Obviously, the higher the ad spend per item, the more important it is to have the bid correctly set.
Sidenote: All Google Ads optimisation is reactive in that it uses the data about what actually sells when traffic hits your website. The way your website is presented, your pricing, the desirability of your stock and your pricing v your competitors, all have a part to play.
Bonus Google Shopping Ads Optimisation Tip:
Something else that I see frequently is this – thousands of items have bids that are too high as they don’t convert at an acceptable return. Thousands of items have bids that are too low because they convert like crazy but have low visiblity. Thousands of items have bids too low to even show the product when someone searches for it.
And it’s that last point that if mastered can be the icing on the cake for your profits.
Let’s take a typical scenario for a large, mature, shopping campaign:
The click through rate and conversion rate are 1% and 2.3% respectively, which is about standard.
Which means that a shopping ad has to be shown 100 times to get a single click and 4,347 times to get a sale.
So if you’ve got items in your campaign that have only 100 impressions over the last 30 days, then that inventory, is going to take a very long time to shift.
Yet I’d guess that for a lot of campaigns, 1/3 of the stock is in that scenario. And by the way, if you only split down to brand level, you’ll never know.
So when you optimise, pay as much attention to the number of impressions as anything else and make sure that all your stock is visible by increasing bids on stock with low impressions.
Device Bidding Optimisation:
This varies wildly, so you need to get it right.
When you’re in “product groups” click on devices and set the time period to the last 30 days or 14 days if you have a lot of conversions. Adjust the bids for desktop, mobile and tablet to acheive a uniform targeted return. Do this daily.
Day and Time of Day optimisation
Peoples browsing and buying habits vary on different days of the week and at different times of day. For example they will often browse more during the week and buy more at weekends. Fortunately, the data is available to you in Google Ads in the Ad Schedule / Day and Hour Report.
This is invaluable.
You can set a schedule in your campaign settings and use a bid adjustment to increase or decrease bids based on the data in your report. But the standard setup only allows you to set a maximum of 6 time slots daily. So at best you get six x four hour time slots per day.
Ideally, we want to be able to adjust bids hourly ie. have 24x one hour schedule adjustments daily. Fortunately, there are 24/7 bidding scripts available which allow you to do just that.
If you’re doing enough volume of sales to make that worthwhile, you’ll get around a 25% boost in performance by using it.
You can set the locations that your ads show in in your campaign settings. Often, for UK retailers, this is simply left as “United Kingdom”.
But we can go further than that by setting specific countries, counties, cities and towns. That then give us the option to use bid adjusments for each location.
As a minimum, you’d add England, Scotland, Wales and Northern Ireland. But it’s better to then also add each county and then the major cities and towns with each county.
You’ll need a lot of data to get any statistical relevance, but there’s another 10% performance boost waiting for you if you implement this and optimise for it.
Negative Keyword Optimisation
Unlike search campaigns, for shopping campaigns, we don’t add the keywords that specify when we want our ads to show. Google shows the ads when it thinks it’s appropriate. So how do we make sure that our ads don’t show for innefective keywords?
The answer is to add negative keywords i.e specify keywords that block our ads from showing when certain search terms are typed into Google.
We can find these keywords by looking at the data in the search terms report.
Use this sparingly though.
If you check in Google Analytics, you’ll find that one of the most common paths to purchase involves multiple Google Ads clicks. You don’t want to block searches that don’t themselves convert but which do lead to further clicks, that do convert.
There are methods to combat this with different shopping campaigns for top, middle and bottom of funnel and different priorities for each campaign. This strategy uses negative keywords extensively to deal with broad keywords that don’t have buyer intent. In other words to engage people at the start of the buying journey. The bids are set lower as a consequence.
My preference is to look at “attributuion” and optimise bids at item ID level.