We’re finally here. Set up one ad with multiple headlines and a couple of descriptions, and Google will start testing combinations dynamically to serve the combination deemed most likely to achieve the advertiser’s stated goal. Oh, and get more real estate than a standard text ad for giving the new machine learning option a go.
Google’s new responsive search ads are now in beta in AdWords, though not available to all advertisers yet.
They are part of the continuum to let machine learning models do the work of ad creative optimization. Some of the initiatives that have come before it: dynamic search ads, automated ad suggestions (formerly known as Ads Added by Google) and Google’s efforts over the past year to get advertisers to give up manual A/B testing and add at least three ads per ad group. This is the same concept, just more automated. And, of course, there’s the push to automated ad rotation optimization.
The argument for having multiple ad options is that your ad groups will have opportunities to compete in more auctions when there are more options for the keywords to trigger your ads. It also requires relinquishing more control to the machines, which can give those devoted to strictly controlling their ad tests agita. But responsive text ads are just one more indication that the days of manual A/B testing are coming to an end, and fast.
Need an incentive to try them? Google’s giving responsive search ads more character real estate than expanded text ads.
- Show up to three headlines instead of two.
- Show up to two 90-character descriptions instead of one 80-character description.
Writing ad combinations and ‘pinning’
Advertisers can set up as many as 15 headlines and four descriptions in a responsive search ad. The other fields are the same as expanded text ads.
The extra lines and automated order mean you’ll need to think through all the various combination scenarios. Google suggests writing your first three headlines as if they’ll appear together (in whatever order) in the ad.
Try to make the headlines distinct from each other, spotlighting different features, benefits, offers, calls to action and so on.
The best practice for responsive text ads — writing headlines that are relevant to the keywords in the ad group and including at least one of the keywords in your ad group in the headlines — remains valid.
There is an option to “pin” headlines and descriptions to specific positions. This will be particularly helpful for advertisers in sensitive categories that require disclaimers, for example. Keep in mind that if you pin just one headline or description to a position, that will be the only thing allowed to show in that spot. It’s possible to pin a few headlines or descriptions to a position to provide more flexibility in the dynamic matching.
To see reporting metrics on responsive search ads, create a filter for them in the ad type column then download the report or open it in Report Editor.
While they’re in beta, advertisers will only be able to add responsive search ads to ad groups with existing ads. You can find more details on the AdWords support page.
Google AdWords is a highly effective marketing channel for brands to engage with customers.
The auction-based pay-per-click (PPC) model has revolutionized the advertising industry, but beneath the seductive simplicity of this input-output relationship lies a highly sophisticated technology.
Within this article, we round up five advanced features that can help you gain that vital competitive advantage.
Google AdWords has undergone a host of changes over the past 12 months, some cosmetic and some functional. Google’s prime revenue-driver has a new, intuitive look and feel that makes it even easier for marketers to assess performance and spot new opportunities.Under the hood, AdWords is home to some increasingly sophisticated machine learning technology. Everything from bid adjustments to audience behavior and even search intent is now anlyzed by machine learning algorithms to improve ad targeting and performance.
All of this is changing how we run search campaigns, largely for the better.
Meanwhile, there are broad trends that continue to converge with search. Voice-activated digital assistants, visual search, and the ongoing growth of ecommerce all center around Google’s search engine.
At the intersection of Google and these emerging trends, paid search will evolve and new ways to reach audiences will arise.
Though this future-gazing reveals just how exciting the industry is, marketers also need to keep one eye firmly on the present.
As it stands, AdWords provides a vast array of features, all of which can impact campaign performance. Though automation is taking over more aspects of the day-to-day running of an account, there is arguably more need than ever before for seasoned paid search experts how know how to get the most out of the platform.
Below are five advanced AdWords features that can boost any PPC campaign.
For all of AdWords’ virtues, it has not been able to rival Facebook in terms of sheer quantity of demographic targeting options.
As part of Google’s ongoing shift from a keyword focus to a customer-centric approach, demographic targeting has improved very significantly.
This feature now allows advertisers to target customers by income and parental status, along with gender and age. Targeting by income is only available for video advertising and is restricted to the U.S., Japan, Australia, and New Zealand for the moment.
Nonetheless, this is a noteworthy update and provides an advanced feature that many brand will welcome.
The available options now include:
Demographic targeting for Search, Display or Video campaigns:
- Age: “18-24,” “25-34,” “35-44,” “45-54,” “55-64,” “65 or more,” and “Unknown”
- Gender: “Female,” “Male,” and “Unknown”
Demographic targeting for Display or Video campaigns can include:
- Parental status: “Parent,” “Not a parent,” and “Unknown”
Demographic targeting for Video campaigns can include:
- Household income (currently available in the U.S., Japan, Australia, and New Zealand only): “Top 10%,” “11-20%,” “21-30%,” “31-40%,” “41-50%,” “Lower 50%,” and “Unknown”
Combined with the improved user interface, this can lead to some illuminating reports that highlight more detail about audiences than we have ever seen in this platform.
It’s not perfect yet and has some drawbacks in practice, as creating audiences can be quite labor-intensive when combining different filters. Nonetheless, demographic targeting is improving and will be an area of focus for Google this year.
Our previous article on demographic targeting goes into more detail on how to set this feature up.
A very natural byproduct of the increase in mobile searches has been an explosion in the number of calls attributed to paid search.
In fact, BIA/Kelsey projects that there will be 162 billion calls to businesses from smartphones by 2019.
Search forms a fundamental part of this brand-consumer relationship, so businesses are understandably keen to ensure they are set up to capitalize on such heightened demand.
Click-to-call can be an overlooked opportunity, as it does require a little bit of setup. If advertisers want to add call extensions, report specifically on this activity, and even schedule when these extensions appear, it is necessary to do this manually within AdWords.
Helpfully, it is now possible to enable call extensions across an account, simplfying what was once a cumbersome undertaking.
This is becoming an automated process in some aspects, whereby Google will identify landing pages that contain a phone number and generate call extensions using this information. However, some manual input will be required to get the most out of this feature.
Our step-by-step guide contains a range of handy tips for marketers who woud like to enable click-to-call campaigns.
Optimized ad rotation
Google made some very notable changes to its ad rotation settings in the second half of 2017.
In essence, ad rotation constantly tests different ad variations to find the optimal version for your audience and campaign KPIs.
Google’s machine learning technology is a natural fit for such a task, so it is no surprise that Google wants to take much of the ad rotation process out of the hands of advertisers and turn it into a slick, automated feature.
Perhaps this focus on the machine learning side of things has led advertisers to beleive that the process now requires no input from them.
A recent study by Marin Software across their very sizeable client base found that many ad groups contain fewer than three creatives:
This is very significant, as Google recommends providing at least three ads in every ad group. Their official stance is, “The more of your ads our system can choose from, the better the expected ad performance.”
Creating a range of ads provides the resources Google needs to run statistically significant tests. No matter how sophisticated the machine learning algorithms are, with only one or two ads in each group there is very little they can do to improve performance.
There is a broader lesson to be taken here, beyond just getting the most out of this AdWords feature.
Even the most advanced technology requires the right quantity and quality of inputs. Although more and more elements of AdWords management can be automated, this doesn’t mean we can leave the machines to their own devices.
There are plentiful best practices that we still need to follow. Optimizing your ad rotation by including at least three ads in each group certainly counts as one of these.
Custom intent audiences
Google is clearly making a play for more of the traditionally ‘top of funnel’ marketing approaches.
The launch of more granular custom intent audiences with the Google Display Network is part of a wider strategy to take on the likes of Facebook by providing greater control over target audiences.
Google’s guidelines provide clear definition over how this recently launched feature works:
For Display campaigns, you can create a custom intent audience using in-market keywords – simply entering keywords and URLs related to products and services your ideal audience is researching across sites and apps.
In-market keywords (Display campaigns)
- Enter keywords, URLs, apps or YouTube content to reach an online audience that’s actively researching a related product or service.
- It’s best practice to add keywords and URLs (ideally 15 total) that fit a common theme to help AdWords understand your ideal audience.
- Avoid entering URLs that require people to sign in, such as social media or email services.
- Include keywords related to the products and services that this audience is researching; these will be used as the focal point for building the custom intent audience.
Custom intent audiences: Auto-created (Display campaigns)
To make finding the right people easy, Google uses machine learning technology to analyse your existing campaigns and auto-create custom intent audiences. These audiences are based on the most common keywords and URLs found in content that people browse while researching a given product or service.
For example, insights from existing campaigns may show that people who’ve visited a sporting goods website have also actively researched all-weather running shoes. AdWords may then auto-create a new ‘waterproof trail running shoes’ custom intent audience to simplify the process of reaching this niche segment of customers.
Once more, we see the addition of machine learning into a core Google product.
These automated audience lists are generated based on activity across all of your Google marketing channels, including YouTube and Universal App Campaigns, along with Search and the Google Display Network.
Although this does not yet provide the level of targeting that Facebook can offer, custom intent audiences do dramatically improve the product and they move Google closer to a truly customer-centric approach.
Sophisticated advertisers will find thata this advanced feature improves performance for both prospecting and remarketing.
Smart bidding has some crossover with the other AdWords features on our list. In a nutshell, smart bidding uses machine learning to asses the relationships between a range of variables and improve performance through the AdWords auction.
It is capable of optimizing bids to ensure the best possible return on investment against the advertiser’s target KPIs. Smart bidding does this by looking at the context surrounding bids and isolating the factors that have historically led to specific outcomes. Based on this knowledge, it can automatically bid at the right level to hit the advertiser’s campaign targets.
These targets can be set based on a target CPA (cost per acquisition), ROAS (return on ad spend), or CPC (cost per click).
The latest option available to brands is named ‘maximize conversions’ and this will seek to gain the optial number of conversions (whatver those may be for the brand in question) against their set budge.
As we have noted already, these algorithms require substantial amounts of data, so this is a feature best used by this with an accrual of historical AdWords performance data.
Smart bidding is also not quite a ‘set and forget’ bidding strategy. Some marketers will still prefer the control of manual bidding and it would be fair to say that smart bidding levels the playing field somewhat across all advertisers.
Nonetheless, it is a hugely powerful AdWords feature and can create multiple account performance efficiencies.
The new attribution model for AdWords was the talk of the town when it was first announced in 2016. But since then, it hasn’t received as much attention.
Even so, I was surprised recently when a colleague in the pay-per-click industry confessed that he wasn’t aware of it.
I can understand how he missed it. After all, it’s one of those AdWords settings that you can easily overlook if you don’t know it’s there.
Therefore, in this article, I’m going to recap what attribution settings are and reveal what you can expect if you start using the new models.
What is Attribution?
Basically, “attribution” is about assigning credit to clicks that lead to conversions.
For example, say “Dan” wants to buy a baseball jersey. So he searches “baseball jersey” and clicks on an ad.
After a few more searches, he decides he wants a Red Sox baseball jersey. So he does some more searching and clicks on another ad from the same company. He then buys a jersey from that company.
How do you want to attribute credit for the sale? Should the first ad Dan clicked get the credit or the second? Or a combination?
That’s what attribution is all about.
Problems with the Old Model
Before Google rolled out its new attribution model, attribution was given to the last ad clicked. But this didn’t always accurately reflect what was happening in the account.
In the baseball jersey example above, the last ad would have gotten all the credit for the sale.
At the time, we could use Google Analytics assisted data to give us a fuller picture of what was happening. But it could be a hassle to try and “meld” Google Analytics data with what we were seeing in AdWords.
Google’s New Attribution Model
This all changed when Google rolled out its new attribution models.
As Google explains:
Most advertisers measure the success of their online advertising on a “last click” basis. This means they give all the credit for a conversion to the last-clicked ad and corresponding keyword. However, this ignores the other clicks customers may have made along the way.
Attribution models give you more control over how much credit each ad and keyword gets for your conversions.
With this change, PPC pros could now choose which attribution model to use.
These attribution models include:
- Last click
- First click
- Time decay
- Position based
And now instead of having to cobble together data from AdWords and Google Analytics, all this functionality now resides in AdWords!
Here are more details from AdWords about the different attribution models:
Changes You May See With the New Attribution Models
When we first tried some of the new attribution models, we noticed some changes in our accounts. You may also experience the following:
1. Your Branded Conversions May Drop and Non-Branded May Rise
Under some of the new models, you may see a drop in branded conversions but an increase in non-branded conversions (i.e., credit given to a branded ad versus credit given to a non-branded ad).
This isn’t a bad thing. We were always convinced that non-branded ads contributed to conversions, but it wasn’t always easy to show that with the data.
Sponsored Webinar – [Case Study] How to Create Content That Earns Links & Increases Your Traffic 316%
Join North Star Inbound’s founder Nicole DeLeon on Wednesday, April 18 at 2:00 PM Eastern. DeLeon will offer advice on how to leverage white hat link building to improve your backlink profile as well as earn traffic.
But now, the contributions of non-branded ads are more obvious. It confirms what we always suspected: It’s not uncommon for people to click on a non-branded ad earlier in the path to conversion and then click on a branded ad later as they get to know your brand.
Under the last click model, branded campaigns would often get all the credit. (“All our conversions are coming from branded campaigns!”) And we might be tempted to pause our non-brand campaigns.
But now the data shows that would be a bad idea.
Having this fuller (and more accurate) picture of what’s going on makes it easier to optimize client accounts.
2. You May Need to Experiment to Find the Right Model
While having multiple attribution models to choose from is great, it isn’t always easy to know which ones to choose.
We were happy to move away from the last click model, for all the reasons explained above.
We decided that first click wouldn’t make any more sense for most of our clients – it would merely shift the overemphasis from the back end of the conversion path to the front end.
Therefore, we’ve been mostly using position-based attribution and linear attribution for our clients. We like these two options because they give credit to clicks that are distributed throughout the conversion path.
Unfortunately, we haven’t yet had the opportunity to use the data-driven attribution model as none of our clients have accumulated large enough amounts of data to make it available to us.
As explained in the Google help file:
Data-driven attribution gives credit for conversions based on how people search for your business and decide to become your customers. It uses data from your account to determine which ads, keywords, and campaigns have the greatest impact on your business goals. You can use data-driven attribution for website and Google Analytics conversions from Search Network campaigns.
If you’d like more guidance on choosing the right attribution model for your campaigns, check out the Search Engine Journal article, “Searching for the Perfect Attribution Model.”
3. Your Data May Show up Differently
Because some of these models are attribute portions of credit to different clicks, we often see partial conversions (e.g. ¼ conversion, ½ conversion) in accounts.
To minimize confusion, we’ve taken to rounding up these fractions when reporting to clients. We don’t want to get into a lengthy (and distracting) discussion about what, exactly is a “¼ conversion”!
How to Change Your Attribution Model
If you’re inspired to play around with the attribution setting on your accounts, you’ll find it under Conversions in AdWords.
From the table, select the conversion action you want to change. You can then select the desired attribution model from the dropdown menu:
With AdWords changing every day, it isn’t surprising that some industry folks (even experienced ones), might have missed the boat on different attribution models.
While these models might seem a bit confusing, they’re well worth exploring. You may end up with a more accurate picture of which ads are driving your conversions.
A topic that is hugely important for any marketer is that of targeting – making sure your budget gets spent on the people most likely to buy from you.
With all the features available to us across digital advertising platforms, we’ve never had it so good.
Yet most marketers I speak to at events are unaware of the options available to them, and are unfortunately still wasting a lot of their click spend on irrelevant people who simply don’t convert.
In this article, I will explore how to carry out advanced audience targeting in Google AdWords, which features you can use to make your marketing budget work a lot harder, and how to use them.
When you talk about AdWords and/or PPC, most people first think of search ads, so we’ll also start there. The audience options on the Google Search Network are a good place to begin thinking about how you will adapt spend towards the people most likely to purchase or inquire.
These days it isn’t as simple as choosing a suitable bid for any given keyword. For half a decade we have been able to use bid modifiers to optimize the best performing parts of an account and automatically spend more money there. Advertisers can automatically weight budget and bids depending on a few criteria:
- Device type
- Time (which can be the time of day, or the day of the week)
You can ask Google to apply a higher bid of up to 300% more than you would normally spend if certain criteria are met, or reduce bids by up to minus 100% (i.e. turn it off).
I am yet to meet a business that can’t improve their PPC spend by considering these factors.
Ask yourself questions like:
- Does the location of where a search is made influence the likelihood of conversion?
- Is someone using a mobile device worth more or less than someone on a desktop?
- Are you happy spending as much on traffic during the evenings as you are during the day?
One way or another, each of these should influence how you spend your money.
As an example, if you are a retailer with high-street stores and an ecommerce site, then bid adjustments need serious consideration.
If someone is searching on their phone, within walking distance of one of your stores during opening hours, then they are probably worth bidding more for. Send them into the store where they will probably have a higher average spend than they would online.
If you take away one of those factors (it’s now an out of hours search, or they are on a desktop at home) then lower the bid and send them to the ecommerce site.
It is important to understand the difference in value for each type of search and reflect that in your bidding strategy.
A more recent addition to this is demographic bidding. You can now also choose to modify bids depending on:
Is your target market younger males? Then bid more for them, and lower bids if searches are performed by females or older people. This allows you to bid on more broad keywords that you may have avoided before, as you can be more certain of the person clicking.
For example, a clothing company targeting men can bid on broad keywords like “skinny jeans”, which would traditionally have attracted more clicks from females, but that audience can now be excluded.
Before looking at the really interesting audience targeting options here, I want to talk about the basics.
Even though a frankly outrageous sum of money gets spent on Google search ads every single day, people don’t actually spend that much time searching. We spend the majority of our time online doing other things like reading news articles and special interest blogs, watching videos, chatting in forums etc.
PPC advertisers can also use Google AdWords to appear here, on the Google Display Network (GDN) with videos, banners and text ads.
Targeting options are plentiful. To start, you can use Placements and Keywords. Placements are the websites you hand-pick for showing your ads, whereas keyword targeting is based on the text showing on web pages, which you might want to appear against.
Taking it a stage further, you can also bid demographically (not only sex and age, but on the GDN you can also choose parental status). This isn’t a situation where you have to choose one or the other; instead, you can layer these options to really refine your audience.
For example, if you are launching a new range of baby clothes, then you may choose to advertise to females who are in their 20s and showing as parents, choosing only to place adverts on baby-related websites and on pages where your competitors are named or where certain target words are present.
The first of the more advanced targeting options that you can choose to implement, as well as or instead of the above, is affinity audiences.
Google refer to these as “TV-like audiences” and they are based around topics of interest. Data is collected as users engage with pages, applications, channels, videos and content across YouTube and the GDN. This information is then collated and used to build a profile of who they are, and what advertisements can be tailored to their personality and preferences.
You should view an affinity audience as a group of individuals who have a general, long-standing interest in a specific subject.
As advertisers, we can take advantage of Google analyzing someone’s overall interests, passions and lifestyle to get a better sense of their identity. If you think about your browsing behavior, there will be certain themes and patterns that are easy to spot and brands can choose to advertise to you and everyone else with the same interests.
If the ready-made audiences like “beauty mavens” or “running enthusiasts” aren’t quite specific enough for you, then you have the ability to build your own.
Custom Affinity Audiences can be created, where you give AdWords a list of keywords that detail the area of interest and also websites that would be frequented by your audience (authority news sites and strong rivals are good starting point) and then it creates a theme for you.
Wouldn’t it be amazing if you knew someone was just about to buy a product in your market? Well now you can. AdWords can qualify someone as being “in-market” for a specific product or service, i.e. they are further down the conversion funnel and are actively considered purchasing a product or engaging with a service that is similar to what you offer.
In-market audiences take into account clicks on related ads and subsequent conversions, along with the content of the sites and pages that a person visits, as well as the recency and frequency of the visits.
You should view in-market audiences as individuals who are temporarily interested in a specific segment. For example, if I am not a car enthusiast and I don’t read a lot of automotive publications then I won’t be in the automotive affinity audience, but for a short period of time, every once in a while, I would fit into an automotive in-market audience when making a new car purchase, showing this intent by researching and comparing cars online.
I often get asked how accurate this is and, from the campaigns we’ve run, we have seen good results. This is down to a lot of effort from Google where AdWords examines repeated patterns of behavior, sets personalized algorithms and updates in real-time, so people not interested or who have already spent their money soon drop out.
Whilst you are able to target pre-defined in-market audiences, you may want to consider segmenting existing affinity audiences in order to re-engage with users already aware of your brand. This increases effectiveness and gives you full control of moving that user through the conversion funnel and toward your products or service.
At this point in the process, your marketing messages should showcase exactly what your brand has to offer and how it differs, alongside any promotions. Make use of all sales tools at your disposal here as these people are looking to spend now and may not be back for a while, if at all.
Very similar to in-market audiences is an additional type of targeting that came out last year – the ability to advertise to people just before or just after a few big events that take place in their lives. Namely:
- Graduating from college
- Getting married
- Moving house
Even though there are just three options to choose from, there are huge industries that surround each one. Take someone moving home, for example: there are property lawyers, estate agents, removal companies, furniture retailers, kitchen fitters, and many more that should all be considering this audience and factoring it into their advertising strategy.
So, when you’ve done all of this work and now have all of these well-targeted people visiting your site, the sobering reality is that most of them still won’t do what you want them to.
The stats show that you should expect over 90% of people to leave your site without converting, 70% of people to abandon your shopping cart without purchasing, and 2-3 visits before someone crosses the line. You need to have a remarketing strategy in place to start turning these stats around and getting visitors back to convert.
Remarketing works after a cookie is placed into the browser of website users and it gives you the ability to show different ads to people depending on the actions they have (or often more importantly, have not) performed when on your site. It’s powerful – Google data tells us that people on one of your remarketing list are twice as likely to convert as a regular visitor.
To make the most of this, think about what you’d say differently to someone who purchased your most expensive product, versus someone who purchased your cheapest. For the former you may use ads to invite them into a VIP club or ask them to refer a friend, whereas the latter you may just try to upsell.
What about someone who abandons the shopping cart? Could you entice them back with a discount? In this situation you could even layer up your targeting with in-market audiences – when someone who previously visited but didn’t buy from you now shows signs that they are close to purchasing, it is time to bring out your strongest offer.
Remarketing Lists for Search Ads (RLSA)
When remarketing was first launched in AdWords, it was available across the Google Display Network, but the search side of things was left out. However, a few years ago we saw the launch of Remarketing Lists for Search Ads (RLSAs), which allows advertisers to also remarket on the search network.
This opens up huge new options for advertisers, as you can display different ad copy and choose different bids and landing pages for your Google search ads if the people searching are on your remarketing lists.
A good example of how game-changing this can be is in the retail sector. If I have a shop that sells gifts online and want more purchases at Christmas, then traditionally I would have had to be very careful with the keywords I bid on, making sure I only cover searches specific to my products.
Broad keywords like “presents” and “gifts for men” ordinarily would not be profitable and would just burn through my budget. RLSAs let me target those keywords, only showing to people who had purchased gifts from me last Christmas.
As I know they are far more likely to convert, I can afford to out-bid the other advertisers and can write ad copy that rewards repeat custom. You can think in a completely different way and have the freedom to almost abandon traditional PPC rules when someone is on your remarketing lists.
If you have a database of customers and prospects that you want to advertise to, perhaps to get them to repeat purchase, or to get them to actually purchase for the first time, then you can use Customer Match to speak to these audiences.
Your email lists can be uploaded into AdWords and you can then approach these people in a similar way to how you’d remarket them. When they are logged in to Google you can show them unique adverts and use different bids across Search, Shopping, YouTube, and Gmail advertising.
The big win here is that it works cross-device, as people tend to be signed in to their Google accounts on phones, tablets and desktops. This is where remarketing often fails, due to the cookie being device- (and even browser-) specific.
This can be powerful for cross-selling (e.g. if someone bought a flight with you, use this to sell them car hire) and it is great for informing existing customers of new releases and any special limited-edition runs.
It is also quite a safe environment to test out new sales offers, as you already have a relationship of some kind with these people so can judge how well things are received here before rolling them out to unknown audiences.
Remember that for remarketing to work really well, you need to be as granular as possible, meaning that by their very nature, the lists often aren’t huge. But if your remarking lists are performing strongly and you’d love the chance to have more traffic just like them, what can you do? This is where something called Similar Audiences comes into play.
If you have used lookalike targeting on Facebook or the prospecting tools within many of the programmatic platforms then this will be familiar. It’s where you ask Google to look at the audiences within your remarketing lists and go find more people just like them.
Available on both the search and display networks, this is a simple way to find new, prequalified users and often returns additional audiences around 5 times the size of your remarketing ones. It automatically updates, so when someone clicks on an ad and joins your site, they become a remarketing list member and are removed from the similar audience list.
Google data shows that brands typically experience a 41% uplift in conversions here. In our work for clients in retail and finance in particular, we’ve seen this go even higher.
Start your audience strategy
If you are keen to try out these features but don’t know where to start, take a look in your Analytics data.
Begin to identify the types of people coming to your site and those that convert. Where are they located? What devices are they using? What times are conversions highest during the day or week?
Bring those themes into your search and display targeting and use the tools within AdWords to find out where these people hang out online, what topics they are passionate about. Additionally, start thinking about how you’d show them different offers once they’ve already been to your site. You’ll be glad you did.
I have a confession to make.
The odds of my instantly deleting one of the many marketing emails I receive each day are about as good as Tom Brady and the Patriots making the playoffs — meaning it’s pretty likely to happen.
Unfortunately for all you email marketers out there, I’m not alone. According to email marketing service MailChimp, the average email open rate across industries is below 25 percent,with a click rate of 2 to 3 percent. That means that, on average, you’d need to send 100 emails to get two or three people to take any action. All that time and energy spent crafting the perfect email marketing campaign will be wasted if you don’t create a complementary strategy to get more sales from your hard-earned email list.
The good news is that you can use Google AdWords as your complementary strategy by simply leveraging the existing data you have on your email subscribers. Let’s dive into the best ways to make that happen.
Learn the ins and outs of Customer Match in AdWords
Customer Match in AdWords might be the greatest secret weapon for email marketers that Google has to offer. It allows you to target or exclude your existing customers on Google Search, Display and YouTube by simply uploading your customer email list to AdWords. Think of it as another way to nurture your sales leads besides sending them more emails.
The best thing about Customer Match is that it’s not that difficult to get up and running. Here’s what you need to do to get started:
- Click on the “Wrench” icon in the top right corner of your AdWords Dashboard.
- Click on “Audience Manager” under the Shared Library section.
- Click on “Audience Lists” from the Page Menu on the left.
- Click on the blue “+” button to create a new audience list.
- Select “Customer List.”
- Choose the option to upload a plain text data file or a hashed data file.
- Choose your new file.
- Check the box that says “This data was collected and is being shared with Google in compliance with Google’s policies.”
- Set a membership duration (this should be determined by the types of customers that make up the list).
- Click “Upload and Create List.”
Please note that these instructions are for the “new” version of the AdWords dashboard. If you’re interested in Customer Match but are still using the “old” version of the AdWords dashboard, see here for more instructions.
Segment your email list
Now that you have a better understanding of Customer Match, let’s take a look at how you might want to slice and dice your email list to more effectively target your sales leads on AdWords.
Take a look at the following email audience segments we use at AdHawk (my company) for a moment:
- New and engaged email subscribers who have not become customers.
- Email subscribers who have not opened an email recently.
- Email subscribers who are existing customers and would be a good fit for an upgraded product or service.
Each of these email audience segments has an entirely different relationship with our business and needs to be messaged to differently. If you have a similar breakdown of your marketing emails, you can repurpose your email list segmentation for your AdWords campaigns via Customer Match. This will allow you to tailor the messaging of your ads for each segment, and as a result, help to nudge your sales leads farther down your funnel.
Create a different AdWords strategy for each segment of your email list
Once you have your email audience segments in place, it’s time to develop a unique AdWords strategy for each segment.
I’m going to use the three email audience segments noted above as examples. Your approach might be different, and that’s okay. Just make sure you’re not using general ads for every email audience segment you have on your list.
Converting new and engaged email subscribers
When a new lead signs up to learn more about AdHawk, our team goes into “educate” mode. The goal is to get them to see the value of our product and services as quickly as possible so we can move them down the funnel.
Our “Welcome” email flow takes the first steps in educating our leads, and it performs pretty well compared to the industry average. But our secret weapon emerges when we take a list of our “new” sales leads and turn it into a Customer Match campaign in AdWords.
Here’s what a typical flow for this segment looks at AdHawk:
- Step 1: Potential customer signs up to learn more about AdHawk.
- Step 2: After signing up, the potential customer receives the first email in the “Welcome” email flow, with a call to action to book a time with our sales team.
- Step 3: A Customer Match segment is created for all “new” prospective customers that didn’t take action on the first email in the “Welcome” email flow.
By using a Customer Match segment for all new and engaged AdHawk sales leads, we’re able to bid up on more generic keywords that would be too risky to bid up on for a general search campaign. We’re also able to create Gmail Ads with a similar look and feel to our “Welcome” emails series that prompt a strong customer recall.
Converting unengaged email subscribers
Converting unengaged email subscribers can be a huge pain in the butt. They’ve stopped engaging with your emails, so the worst thing you could do is continue to bash them over the head with more emails.
Here’s the flow we use to re-engage leads that have left us hanging:
- Step 1: Potential customer signs up to learn more about AdHawk but does not engage with our emails for 30 days.
- Step 2: A Customer Match segment is created for all “unengaged” prospective customers.
- Step 3: A Remarketing campaign is created to target prospective customers that have not converted after 30 days.
- Step 4: We tailor the Customer Match and Remarketing ads to promote a special offer.
This group is the least likely to convert, so any new business scraped up is a huge win! It’s important to educate these stale leads on what we do and remind them why they signed up in the first place.
Upselling existing customers to a new product or service
Most marketers are so intent on attracting new business that they often forget that there is a wealth of opportunity under their noses. Don’t sleep on marketing to those that have bought something from you in the past! We use our existing customer segment to promote new features or products we feel they will be a good fit for.
Here’s the flow we use to target existing customers:
- Step 1: A Customer Match segment is created for our “Existing Customers.”
- Step 2: We further segment this list by renewal date to ensure that customers see our ads when their contract is up.
- Step 3: Tailor the ads to promote additional services we offer that our customers are not leveraging.
We’ve structured our flow this way because our product runs on a subscription basis. If you’re selling physical goods that can be repurchased often, break down your segment by the products your customers have shown the most interest in. That way, you can tailor your ads to the specific products you believe would resonate most with them.
Are you leveraging AdWords as part of your email marketing strategy? If you are, I’d love to learn more about what strategies you have used that have been successful.
When it comes to PPC (pay-per-click) advertising, marketers are always looking for ways to improve performance. The goal is to optimize the account continuously, finding ways to save money and spend more efficiently. Anyone with PPC experience knows that there are countless ways to do this, including adjusting keyword bids, setting daily ad schedules, adding negative keywords, and so on.
Why is it that one of the most obvious ways to spend more efficiently is often overlooked? While everyone is spending time making little tweaks to try and push the needle forward, they often lose sight of the bigger picture, which is the account’s overall budget allocation throughout the year. Using both historical data and competitor’s data and being prepared to adjust on the fly, you can ensure that you are not limited when business is booming and that you’re tightening up and maintaining discipline during slower seasons.
SEASONAL BUDGET ADJUSTMENTS
The most common tactic employed when trying to optimize spend throughout the year is seasonal budget adjustments. This is especially true in the vacation rental industry. Many travel destinations experience busy seasons where it is important fill all vacation rental properties, meaning it is important to allocate additional ad spend to these months.
Property managers near ski resorts need to ensure that their winter ad spend budgets can keep up with increased winter demand. If budgets are limited during these important months, it can mean leaving easy money on the table. On the opposite end of the spectrum, once the ski season ends and the snow begins to melt, these destinations tend to experience a lull in traffic. This can be a good time to reduce bids and budgets to save up for the next big push. This is also a good opportunity to shift focus toward homeowner acquisition as opposed to reservations.
On a similar note, it is also important to understand how specific holidays affect performance. Holidays should be considered their own entities, and even during a slow month or season it may be important to jack up budgets for a brief period to ensure maximum efficiency.
Sticking with the ski town example, April and May (a.k.a. “mud season”) tend to be very slow booking months as snow is melting, but Memorial Day can lead to a nice influx of visitors, especially if the specific town is hosting an event. Thus, it is important that spend begins to increase leading up to Memorial Day to ensure that your budget is keeping up with traffic.
It is also important to keep an eye on competitors and their actions, as it may be an important factor in how you spend your budget. If a competitor begins to increase their budget and bids, it is going to lead to an increase in CPC for your account. If this is during a busy period, it may mean that you need to increase the budget even further to keep up. Even though it’ll be more expensive to generate conversions, many times it is too profitable of a time period to let the competition win.
BE PREPARED FOR CHANGES
Lastly, it is important to be able to adjust strategy on the fly due to unforeseen circumstances. While a lot of the previous examples involved using historical data to formulate a plan of attack, things can change out of the blue because of unforeseen weather events, natural disasters, or other large-scale events. I imagine most businesses in Houston would have benefited from completely pausing ad spend during and in the immediate aftermath of Hurricane Harvey, as the focus of the entire city shifted squarely toward recovery. Similar things can be said for locations dealing with forest fires, mass protests, and so on.
All in all, the moral of the story is that an easy—but often overlooked—strategy for improving account efficiency is better allocating your overall budget throughout the year. Be prepared to take full advantage of busy seasons and important holidays and to spend more efficiently during slower time periods. And most important, be ready to adapt to any unforeseen events. By being prepared you can continue to decrease the amount of wasted spend in your account and improve overall efficiency.
If you really want to improve your Google AdWords’ results, perform these three maintenance tasks and watch your CTR skyrocket.
There are three crucial Google AdWords checks you should do regularly that ensure your pay-per-click (PPC) campaigns remain healthy.
1. Ongoing management: checks.
This part of managing accounts is relatively easy to carry out and is essential if the rest of your AdWords work is to be effective. There are three things you want to measure in this step:
The first step to any success in any business system is deciding what measurement is going to tell you that your business is, in fact, running like it should. Setting up conversion tracking is simple, but it’s also easy for things to break down. If you’re using code on key pages to measure conversions, check regularly to be sure the code is still present and installed correctly.
Also, check the conversion process itself to ensure no glitches have cropped up. Working tracking code on your “thank-you” page is pointless if the lead-capture form is broken. Make sure these pieces of your funnel are installed and functioning.
The settings inside your AdWords account aren’t likely to change without your noticing. But, don’t take this for granted, especially if you’re not the only person administrating your campaign. Keep a written record of your settings and occasionally check them to be sure nothing’s been moved, adjusted, paused or unintentionally reset.
Start by looking for obvious glitches such as broken formatting or dead links. Then take a high-level view and ask whether you’re matching the right landing page to the right ad copy. Does what you promise in the ad get delivered in the landing page? Is the connection between the two obvious to the visitor? Is there a better landing page you could be using?
2. Ongoing management: optimizing.
“Outliers” is a term made popular by Malcolm Gladwell’s book Outliers: The Story of Success. It refers to fringe elements that, in some way, behave differently than everything else. The “outliers” in AdWords are the campaigns, ad groups, keywords, ads or placements that perform significantly better — or significantly worse — than everything else.
You could spend your time optimizing every last element of your PPC campaigns, but that’s not smart. Go for your outliers first. The goal is simple: Increase the good outliers, and decrease or fix the bad ones. If you have a particularly high-performing keyword, for instance, you might want to raise the bid and get more impressions, clicks and conversions. If you have a poorly performing keyword, try lowering the bid or even removing it from the campaign altogether.
Before you reach for the “nuke” button, however, ask whether you can improve the bad keyword by writing a better ad or building a better landing page for it. Sometimes your worst player can turn into your star player.
3. Ongoing management: expansion.
Once your AdWords account is well-optimized, think expansion: more impressions, more clicks and more conversions. With PPC, you can never have too much of a good thing.
There are a number of different ways you can expand an account.
- Start with new keywords. The search query report is a goldmine for this information. (To access this, look under the Dimensions tab, and in the “View” drop-down box, select “Search Terms.”) This shows you the actual search terms people typed in that Google chose to show your ad for. Look for common phrases that aren’t yet in any of your ad groups. Add them in.
- Look in the Opportunities tab for Google’s list of additional keyword suggestions. This section is useful for finding new ideas, but beware: Don’t be too quickly sucked into Google’s insistence that the best thing you can do is increase your maximum bids. There’s a time and place to raise bids. Don’t do it in kneejerk response to Google’s pestering.
- Pay regular visits to Google’s Keyword Planner. We also recommend third-party applications like SpyFu, SEMrush and WordStream. Go digging there on a regular basis to find new keyword ideas.
- Try aiming for the top positions. When your ad moves to the top of the page above the organic results, the positive difference in click-through rate (CTR) is massive. Use your “Top vs. Other” report to see this spelled out in hard numbers.
Display campaigns with vibrant ads that show everywhere
- Never assume that the performance of your display campaign ads has hit its ceiling. Keep testing new ads, especially image ads and try to beat your best CTR. You can often get a quick win just by testing a vibrant new image or a new headline.
- If you’re using a managed placement campaign, look around for new sites where you can feature ads. If you’re using contextual targeting, look for new keywords or topics you can introduce that will expand the range of sites where your ads can show.
- Experiment with different targeting methods. If you’re only using managed placements, give contextual targeting a try and vice versa. And if you’re not using remarketing, this should be at the top of your idea list.
Want a head start in making the shift to expanded text ads in AdWords? Columnist Frederick Vallaeys has an AdWords script that pulls information from your SEO titles and meta descriptions.
Just about a week ago, all AdWords advertisers got access to expanded text ads. Have you converted your whole account over to take advantage of the new format yet?
If you haven’t because you’ve been too busy with the day-to-day of managing accounts (or because you took that well-earned summer vacation), read on for an easy to use AdWords Script that can help.
The goal is to leverage all the work already done by your SEO team to give the PPC team a head start with creating longer text ads. After all, the recommended lengths for SEO titles and descriptions are pretty close to the new limits Google allows for expanded text ads.
While it’s very hard to fully automate the creation of ads, the goal of the script is to give advertisers a starting point that is better than their existing ads or worse, a blank bulksheet. Depending on how your SEO tags are set up, it’s possible that the script will produce acceptable results right out of the box.
Leveraging existing script code
The script is a simple adaptation of one of the oldest scripts out there: the broken URL checker. Instead of fetching landing pages to see if they return errors, I modified the code to fetch the metatags of the page, specifically the title, description and first h1 tag of each page that is used as a landing page by an existing ad.
I’ve said it before, and I’ll say it again, but the beauty of AdWords Scripts is that they are easy to modify. If you also had the idea to use metadata to construct expanded text ads, you could have grabbed an existing code sample and with some small tweaks, adapted it to complete this task.
I know the majority of readers don’t write code, though, so you can simply copy and paste a fully working version below.
Turning the title tag into headline ad text
The recommended length for the title tag is 50 to 60 characters, according to Moz. Google allows 30 characters each for headline 1 and headline 2, for a total of 60 characters. Beware, though, that titles that exceed a certain pixel count may get truncated, and, according to Google, titles that are 33 characters or shorter are the least likely to be truncated.
The script fetches the title and splits it into two lines of up to 30 characters each. In case this truncates your original title tag, we’ve included it in the spreadsheet so you can reference it while you rewrite your headlines.
I’ve also added an optional setting to specify how long you want your overall title to be so if you would like to stay within 33 characters because you’re in a vertical where there are strict regulations about ad text, then this is a good option for you.
The setting is called “maxHeadlineLength,” and you can set the value to “33,” some other number, or leave it at “60” to use the full allowed length. Do keep in mind that many headlines will be shorter than 60 characters even when you are allowing the script to use all characters. This is because we have to split the title between two lines, so we look for the first word break before the end of the line.
Turning the description tag into the ad description
If our SEO colleagues stuck to Moz’s recommendation for a description tag length, our landing pages should have a description between 150 and 160 characters in length. That’s twice as long as the 80 characters Google now allows us to include in expanded text ad descriptions.
The script pulls the description tag in its entirety and truncates it at the last full word before 80 characters. So once again, you can refer to the original description if you need to rewrite what the script suggested for the starting point.
There are no more mobile preferred ads
While the announcement and eventual launch of expanded text ads got quite a bit of media attention, the removal of a mobile preferred ad hasn’t garnered quite the same attention. So if you haven’t played with the new format at all, when you do, you will see that there is no difference between expanded text ads for different devices. The only difference is that advertisers get to set an optional mobile final URL.
My script fetches both the old regular final URL and that of the mobile preferred ad. When it generates the bulksheet, it combines the two together on a single line so that you can write a single new ad that will have both the mobile and regular final URL.
The metadata is only fetched from the regular landing page, so if you use different metatags on your mobile site, you’d have to edit the code to start fetching those.
The script’s output
When the script finishes, it will have created a Google Sheet in your Google Drive named “Expanded Text Ads From Meta Data (Your_Account_Name).”
As I mentioned, this automation is intended to give you a head start on creating new ads but it will not be perfect, so you can edit our baseline suggestions in your favorite spreadsheet tool; then, copy and paste it all into the newest version of the AdWords Editor, where you can make final tweaks and upload it to the account.
The results of expanded text ads
It’s too early for me to share findings of how expanded text ads have performed for us but Merkle RKG has announced some mixed preliminary results. This is not to say that you should stay away from expanded text ads. Getting your own experiments up-and-running is an important first step in being able to A/B test the results.
I recommend that you launch expanded text ads for all your ad groups as soon as possible but leave regular ads running just in case those still perform better. Advertisers will still be able to create legacy ads until October 26, 2016.
Here’s the AdWords Script that you can copy and paste into AdWords. As you run it, keep a few things in mind:
- AdWords Scripts can only run for up to 30 minutes. If you have too many unique final URLs, the script may time out before it is able to process all of them. If this happens, you can update the query we use to fetch ads to only include a subset of all your campaigns.
- The script uses the UrlFetch() function, and this has a Google-imposed daily limit of fetching a maximum of 20k to 100k URLs and 100 Mb of data (depending on your version of Google Apps). If you need more quota, you could split the script by campaign and then wait a day to start on the next one.
I’m excited to see the launch of the first of many announced new capabilities for AdWords. Google is strategically launching this in the dog days of summer to minimize any possible disruption to advertisers, but that doesn’t mean you shouldn’t start converting your existing ads to the new expanded format as soon as possible so that you can get your own data about what works best.
And while this script should help most advertisers make the transition quickly, let’s not forget that the real benefit should come when we take the time to craft new messages that take full advantage of the longer ad length.
Thanks to Google, a new artificial intelligence systemis outperforming humans in spotting the origins of images.
Google has unveiled a new system to identify where photos are taken. The task, simple when images contain famous landmarks or unique architecture, goes beyond the overt to examine small clues hidden in the pixels.
The program, named PlaNet, uses a deep-learning neural network, which means the more images PlaNet sees, the smarter it gets.
“PlaNet is able to localize 3.6% of the images at street-level accuracy and 10.1% at city-level accuracy. 28.4% of the photos are correctly localized at country level and 48.0% at continent level,” wrote the research team.
That’s still a long way from a reliable level of accuracy – but PlaNet already outperforms even the most well-traveled humans.
To compare PlaNet to human accuracy, the researchers matched their program against 10 well-traveled people in the game Geoguessr, a game providing a random street-view photo and requiring players to identify where they believe the photo was taken.
PlaNet and its human challengers played 50 rounds in total.
“PlaNet won 28 of the 50 rounds with a median localization error of 1131.7 km, while the median human localization error was 2320.75 km,” according to the paper.
Other computer programs are tackling image location as well. Im2GPS has achieved high accuracy by relying on image retrieval to identify location. For example, if im2GPS was trying to identify where a picture of a forest was taken, it would browse the internet’s millions of forest photos. When it found one that looked almost identical, it would conclude they were taken in the same place. With enough data, this method can achieve high accuracy, according to the paper.
The researchers trained the neural network using 29.7 million public photos from Google+. The neural network relies on clues and features from photos it has already seen to help identify the most likely whereabouts of a new image.
The program has some limitations. Because it depends on internet images, PlaNet is at a disadvantage when confronted with rural countrysides and other rarely photographed locales. The team also left out large swaths of the Earth, including oceans and the polar caps.
Tobias Weyland, the lead author on the project, noted that supplementing internet photos with satellite images resolved some of these weaknesses. PlaNet also focuses on landscapes and other factors besides landmarks, making it more accurate at identifying non-city images than other programs.
Structuring campaigns based on personas is can effective, but what happens when you have keyword overlap that dilutes your messaging?
Keywords and search queries can mean different things to different people. That’s where intent comes in. You might, for example, have one keyword that serves multiple personas.
So the work that you need to do to qualify those leads in a PPC environment typically happens at the ad creative and landing page level, not necessarily with the PPC campaign structure.
My agency recently inherited a PPC account that was building campaigns based on personas, and the strategy didn’t prove itself. (By the way, if you’re interested in the cabinet of curiosities we discovered when we got into the account, check out my column from last month.)
Using this account as a case study, I’ll share with you some important lessons on understanding keywords, what to do when they serve more than one audience type in PPC and what results you can see when you reorganize based on the moneymakers.
The Situation: Misguided Campaign Structure
The business in question runs on licensing and continuing education for a particular industry. So the company wanted to target two distinct personas based on those two different groups.
The previous PPC account managers had separated the campaign structure based on audience personas: continuing education seekers and new licensees.
That sounds okay at the outset, except the keywords that were in the continuing education campaign were some of the same keywords that were in the new licenses campaign — and frankly, any of them could cater to either audience.
Here’s an example of how it was structured:
Continuing Education Campaign
dog walking continuing ed
dog walking renewals
dog walking licensing
New Licenses Campaign
dog walking licenses
dog walking courses
dog walking license tests
For the continuing education ads, searchers landed on a page that catered to that side of the business, and for the new license ads, they landed on a page with info specific to that.
But the thing is, it was a crap shoot. Any new licensee or a person seeking continuing education could come in via “dog walking courses,” for example.
So in the off chance they did convert on a particular landing page, in my opinion, it was pure luck.
The Fix: Follow The Money
When we dug into the account, we rolled up our sleeves; we had work to do. And we did what we always do: Follow the keywords that are making the business money.
The client was at first hesitant and wanted to continue the way they had been: campaigns based on personas. (We did get past that.)
Once we restructured the campaigns with the keywords that drove traffic and conversions, we refocused the ad creative and the landing pages (our messaging strategy catered to both possible personas), and let those do the work of qualifying the personas:
The Results: 123% Lift In Revenue
With a little love, the account experienced a huge lift in conversion rates, transactions and revenue for the client year over year. We suspect this will only get better, as we are still testing our strategy and adjusting it as we go.
From the report, we see:
- A 39-percent lift in conversion rates.
- An 84-percent lift in transactions.
- A 123-percent lift in revenue coming from PPC.
The moral of the story is this: PPC managers and online advertisers need to do the work to understand the intent behind the keywords, and then work that insight into various steps in the funnel.
That starts with ensuring the account structure is sound, following the keywords that are showing the most ROI, and then using marketing insights to make the ad messaging and landing pages guide the audience down the path to conversion.