It seems a bit counter-intuitive, but there really are things every advertiser knows that Google doesn’t – like how people act after they’ve converted from one of your Smart search or shopping campaigns. These are things like:
- Customers who made a purchase but returned all or part of their order
- People who submitted a form and took 12 actions outside of Google’s sight before converting
- Loyal buyers who split repeat purchases between digital and in-store
There’s no question that Smart Bidding and offline conversions have completely changed advertising, but are you one of the few who feed information back to Google to get the most out of those systems?
Ad platforms like Google are automating rapidly, but they need all the information you can provide to speed up machine learning and make better decisions on your behalf. If you want to your ads in front of people who will give you profits instead of just sales, you have to rethink the way you perceive PPC optimization.
We sat down with two people at Google who understand this process best:
- Emi Wayner, Platform Partner Lead, Channel Sales
- Alex Ioch, Regional Product Lead, Automation
Alex is part of the team that builds the products at Google that enable this new wave of search advertising, and Emi helps businesses adapt to the new realities.
We talked about how to use offline conversions to get better leads, why non-retail models need to import post-conversion data, and how to run Smart Bidding to its full potential.
Customer data is the key to sustainable growth.
A study by Boston Consulting Group indicates that online advertisers see average gains of 20% incremental revenue and 30% cost efficiency when they integrate strategies with customer data across the purchasing journey.
Whether you choose to import or track offline conversions, there aren’t many arguments against the importance of first-party data when optimizing modern PPC campaigns.
Just remember that phased growth is the only way you can achieve results with any of these recommendations.
Automations like Smart Bidding require a considerable volume of manual conversions before they can begin making the right decisions for you. Lay the groundwork with manual optimizations, and allow at least two weeks for learning. That goes for search as well as shopping.
In other words, don’t expect huge gains from your campaign metrics within days of feeding in customer or post-conversion data.
Don’t measure everything. Here’s what to focus on.
While Google Ads can measure online activity, sometimes a user who clicks on your ad and enters your sales system converts offline where Google can’t see it, like your CRM or in person.
So there are two groups of components you want to keep track of.
Online: leads & sales
When you track only leads, you get a fair idea of the initial demand for your product or service. Extending tracking to the final sale gives you a better picture of your overall purchase journey:
- How much interest does your brand generate online?
- How long can you sustain that interest?
- How well can you persuade people to give your business money?
Sometimes, even the final sale happens offline. And in these cases, it’s critical to feed that information back into Google Ads with offline conversion imports.
But even if you were to track both leads and sales online, you’d still only see a part of the picture. If a university gets 2,000 requests for information that ends up with 20 enrollments, all they know is they have a 1% conversion rate.
But what about all the other stuff that happens in between? How much did that contribute, if at all?
Offline: The middle funnel
There’s usually a big disconnect between what advertisers tell Google to track and what actually matters to their business. An online form submission is easy to fire a pixel on; you can’t do that for things that happen offline or on a channel that isn’t trackable.
But when you do measure what happens between lead and sale, you get a better idea of what actions are most profitable for your business. Let’s take the same university example Google showed on PPC Town Hall:
- 2,000 requests for info yield 200 applications started (10% conversion rate)
- 200 applications started end with 80 applications completed (40%)
- 80 applications completed lead to 20 students enrolled (25%)
All of a sudden, you can see more of what impacts the final conversion. When you feed that information back to Google is when the magic really happens.
It’s okay to estimate expected value.
When you import offline conversions into Google Ads, it helps to be able to assign values to them based on what level of value they contributed to your business.
The best way to do that is to work backwards:
- You know that a student who enrolls contributes an average of $10,000 to your university
- You account for some fluctuation and tell Google that a completed application is worth $2,000 (25%)
- A started application is then worth $800 (40%)
- And a request for info is worth $80 (10%)
These numbers make it easier for Google’s algorithms to piece together the signals that make a user worth $80/$800/$2,000/$10,000 to your university.
Important: “Don’t get hung up on the numbers,” Alex says. Conversion values are a tool to help Google score and prioritize different users, and these numbers do not directly impact your ad spend or ROAS.
Smart Bidding works best when it has offline data.
When you use Smart Bidding without offline data, Google will look at tens of millions of signals that are familiar to most online advertisers:
- Demographics
- Conversion rate
- Audience segment
- Browser
- Language
- Keywords
- Bids
- Time
- Location
- And much more
But when you import offline data, you can feed so much more information to Google that it uses to optimize future targeting. In the case of a university, this might include parameters like:
- SAT scores
- Preferred courses
- Income
- Credit history and score
- Referrals and awards
- Existing credits
All of a sudden, the jigsaw Google puts together of your ideal customer gets a bunch of new pieces added to it. They can tell with greater accuracy which bids to which clusters of users will give you the best long-term returns.
Value-Based Bidding strategies let you optimize Smart campaigns.
Assigning conversion values lets you differentiate potential customers and bid toward more valuable outcomes.
For education, that could look like assigning different values to students who request a brochure and ones who sign up for a free online course. In retail, it might mean segmenting customers who make frequent returns from those who end up keeping their purchases.
A value-based bidding strategy tells the Smart Bidding algorithms that one group of users/customers is worth more to the profitability of your business than another.
And so Smart Bidding then knows to optimize for the group that is worth 2x/3x/10x more to your bottom line, giving you greater value from your ad spend that goes beyond just ROAS.
Additionally, fixed-value bidding skews your spend and limits your returns. User segments with a higher average bid than your fixed value become missed opportunities, while you over-invest in segments with lower average bids.
Monitor. Intervene. Educate.
There’s no shortcut or a way around the work. Finding sweet spots for ROAS and CPA targets take time, so don’t resist or be intimidated by the learning process. Test your ideas, monitor the outcomes, and optimize as you hit subsequent rounds of statistical significance.
At no point should you keep Google’s machine learning algorithms in the dark, even if data visibility is something all advertisers would like more of from the platform. It only impacts your business negatively; Google’s automation will grow with or without your accounts.
Once you know what works, automate the solution and keep control over how and when you feed information back to Google. A third-party tool can make monitoring and intervention easier and limit damage from malfunctions in the ad platforms.
Optmyzr offers a 2-week free trial that’s helped teams like Sean’s preserve their time and mental fortitude for what matters, allowing them to double their managed ad spend.
