Want to see something impressive?
Spend $1 on Google clicks. Earn $25 in revenue.
Yep. We’re half expecting a crack squad from the Australian Mint to kick our door in to accuse us of printing money.
Because that’s pretty much exactly what this Google Shopping campaign is doing for our client.
Their AdWords Account (aka money printing machine) has turned a $233 Google Ad spend into $5,651 in revenue in 16 days in August so far:
And the good news…
We’re going to show you how we did it.
This post lifts the curtains on a highly profitable Google Shopping AdWords campaign.
Learn how we did it and how you can do the same for your own campaigns.
Alright… let’s get into it.
This client is a current client, so we’re going to keep things anonymous.
But we can give their back story.
And it starts with a problem:
No matter what they tried, they simply couldn’t get their Google Shopping ads to show.
They’d tried everything.
Resubmitting the feeds.
Creating campaigns for specific products and brands.
For five months they were at it.
And no matter what they tried, instead of a flood of traffic, they had an underwhelming trickle.
We’re talking 50 visits a week.
For a store with thousands of products, this was unacceptable.
We were tasked with fixing this problem first:
So our PPC Pros got on the case.
Pretty quickly, impressions jumped from around 5,000 per week to 230,000 per week.
Here’s how we did it:
- Re-built the feed from the ground up. This client was using a standard Magento extension for the feed. We ditched that and moved them over to another feed platform.
- Audited and optimised the Google Merchant Centre account. We ran a fine tooth comb over every setting in the account and tweaked accordingly.
The reason we moved them over to a new feed boils down to one thing – control.
With our new feed, we were able to ensure absolutely every element Google Shopping looks at when deciding what ads to show was there.
- Well optimised product titles with keywords (brands, product names, sizes, variants)
- Optimised product descriptions
- Better categorisation of the products in line with Google Product Taxonomy
- Custom labels
Anyway… some of that is probably gibberish to you, right?
The key takeaway is that the foundation for any good Google Shopping campaign is a high-quality feed.
We Want $5 ROAS!
With the feed optimised and the ads showing 4500% more than they had been, it was happy days.
Well, yes… and no.
In order for the client to be profitable, they needed to hit $5 ROAS.
That’s “Return On Advertising Spend”.
$5 in revenue generated for every $1 spent on the ads.
The goal had been set.
In March they weren’t even close:
That was a $3.28 ROAS for March.
Effectively… the client was losing money if it remained like this.
After we got our PPC Pros on the case, over April, May, June and July the ROAS looked like this:
The ROAS each month was an improvement on the last:
- March – $3.28 return for every $1 spent on ads
- April – $4.47 return for every $1 spent on ads
- May – $5.06 return for every $1 spent on ads
- June – $7.23 return for every $1 spent on ads
- July – $9.10 return for every $1 spent on ads
In August, it went nuts:
- The ads had no visibility, so we fixed the feed.
- The ads suddenly had a tonne of visibility, but were not profitable
- Over the course of 4 months we increased ROAS to $9 (almost double the client’s target)
- So far this month, the party’s really started, and we’re at $25 ROAS
[box type=”info”]Want these results yourself? Download our Google Shopping Optimisation checklist. We use this to optimise our client accounts and now you can too.[/box]
$25 ROAS! Wooooooow! How Did You Do It? Here’s How.
Alright, enough living in ROAS past.
How the hell did we do this?
Well, to tell the truth, through a lot of hours of analysis and optimisations.
But here are some of the “big ticket items” we worked on.
After sweating on the data from March, April & May we noticed a trend.
Mobile traffic for this client absolutely sucked.
Here’s how desktop, mobile and tablet traffic performed over those 3 months:
The key points here are:
- Desktop traffic converted at 1.6% compared to 0.58% for mobile. This means if a person clicked on an ad on their desktop they were nearly three times more likely to buy.
- The revenue generated from the ad spend on desktop was acceptable. Mobile and Tablet… not so much.
We went back to the client’s Google Analytics account.
This trend was repeated.
Mobile traffic never converted at better than 1%, no matter what traffic source it came from.
So how did we fix it?
We applied negative bid adjustments of 100% to mobile and tablet devices.
This tells Google to not show the shopping ads to anyone on those devices:
To do this:
- Click into a campaign then Click Settings
- Click Devices
- Under the “Bid Adj” column add in “-100%” for the devices you want to get rid of.
And that’s all there is to it.
Well, actually, there was one more action point here.
Improving the client’s mobile site.
There is a reason that mobile traffic converts at less than 1% no matter what source they come from.
And that reason is the experience they have when using the site.
We’ve made a bunch of (still to be implemented) recommendations to improve this so that one day we can turn this source of traffic back on.
Killing “Click Wasters”
Click Wasters are products that get clicked on a bunch yet don’t turn into sales.
They simply sit there wasting your spend.
And you don’t want that.
So get rid of them.
Here’s what that looks like:
As per the above screenshot, we look for products that have cost us money over a decent period of time (say 2 or 3 months) but have not converted. Or, if they have converted, they have not done so profitably.
Better to just ditch them and focus your spend on the products you know do convert well.
All you have to do is jump in AdWords and drill down by the product groups.
In the CPC column, where you would normally set your bids, instead of entering your bid you can select the “Excluded” option:
[box type=”info”]Want these results yourself? Download our Google Shopping Optimisation checklist. We use this to optimise our client accounts.[/box]
The next area we looked at was the time of day.
And what times of day are important?
The ones that people are buying up big, of course!
So, we needed to work out when people are buying and tell AdWords we want more of that action.
Here’s how to do it.
We started by running four months worth of sales data in AdWords and segmenting it by the hour of the day.
It looks something like this:
That’s 24 hours of the day, broken out into individual hour blocks, in Military Time.
The red box above highlights a run from 2 pm to 10 pm where the client’s site makes more sales than at other times of the day.
We decided to go after this time of day.
This is done through the use of Schedules:
- From Midnight to 2 pm everything runs as per usual
- From 2 pm to 10 pm we tell Google to bump our bids up by 10% so we have more visibility in this high sales time.
- From 10 pm to Midnight it reverts back to business as usual
It’s worked well so far.
So well, in fact, we’re now going to be tweaking the schedule to also give some love to the 10 am to midday period, which is also a strong performer.
Want your own money printing machine?
That just about wraps things up for this case study.
Hopefully, you pulled a few gleaming nuggets of gold from this post.
So get out there and apply them to your own campaigns.
Or, if you’re not an AdWords nerd (that’s cool, we’re a rare breed) then we’d love to take a look at your account and provide a few pointers (free of charge).
Just apply for a free strategy session with one of our PPC Pros below:
[button link=”/free-strategy-session/” type=”big” color=”lightblue”] Get A Free Strategy Session[/button]