Global mobile trends all point to the same conclusion – operating in channel-specific silos no longer works, and now’s the time for marketers to implement a strong cross-channel marketing strategy.
If you subscribe to this blog (and if you don’t, see that second little box on the right), you already know we’ve been evangelizing the message of “cross-device, cross-channel.” There’s a good reason for that.
As we approach the halfway point of 2016, it’s more important than ever that marketers not only use data to understand customer behavior, but also to act on that behavior to deliver engaging, personalized experiences.
On May 25, Nitin Rabadia – our Director of Audience Marketing EMEA, APAC – will explain how to use data to win the online battle for attention and revenue. Gleaning insights from our 2016 Global Mobile Report (available with webinar registration), Nitin will field your questions and discuss:
Register for the webinar today.
When we looked at performance marketing data from the first quarter of 2016, one thing became clear: cross-channel, cross-device targeting remains the most powerful differentiator for profitable marketing strategies.
To create our quarterly benchmark reports, we sample the Marin Global Online Advertising Index, composed of advertisers who invest more than $7 billion in annualized ad spend on the Marin platform. We analyze data from around the world to create our report. For Q1 2016, key findings include:
For detailed information on Q1 2016 search, social, and display mobile performance – including detailed data charts with YoY performance and up-to-date recommendations – download our Performance Marketer’s Benchmark Report Q2 2016 – Vital Search, Social, and Display Performance Data by Device.
Mother’s Day is almost here! With flowers, cards, and family visits close at hand, many brick and mortar retailers are gearing up for the shopping spike. The season of maternal appreciation extends to online retailers, who are also gussying up their search, social, and display campaigns to attract consumers around the world.
How did online retailers do in 2015, and what to expect this year?
In the week leading up to Mother’s Day 2015 (May 10th), clicks increased an average of 15% across retailers as click-through rates rose 6%. In addition, spend increased 9% during the same time period, peaking a few days before Mother’s Day.
Most notably, conversions saw a bump of 12%, peaking on the 5th at 18% above the monthly average. This noticeable bump for all retailers was more pronounced among those specialty retailers that Mother’s Day particularly impacts.
CPCs actually dropped slightly during this period, except for two days where they spiked, the 4th and 5th. The 5th proved to be a particularly important day for consumers and advertisers, showing abnormal surges along all metrics.
Perhaps consumers took account delivery times and the looming holiday date into account, giving themselves a few buffer days in case of delays in delivery and arrival.
These numbers dropped dramatically on Mother’s Day itself, and returned slowly to roughly average afterwards. Click-through rates remained elevated for Mother’s Day and a few days afterwards before returning to seasonal norms.
For retailers looking to maximize their Mother’s Day sales, here are a few key takeaways:
This is the first in a series of posts on transparency. In today’s post, we lay out the many ways transparency is elusive in digital marketing today. We also include some best practices for stamping out the fuzziness prevalent in the programmatic landscape.
Most marketers will admit transparency in media buys sounds like a good idea. So why don’t we have it all the time? Inertia, circumstances, or legacy business practices are the usual culprits. Knowing about the types of programmatic transparency is a good place to start.
You may have read about the recent survey on programmatic buying by Forrester and the ANA. Although we know intermediaries carve up a media dollar along the ad delivery path, a surprising 33 percent of survey respondents in this study have turned a blind eye while knowingly opting into an undisclosed programmatic model.
Not knowing the true value of your media obscures your true ROI. This buyer/seller blindness stands in the way of programmatic growth and success.
Let’s dive in and take a look at the three types of transparency: intermediaries, environmental, and data.
According to the ANA/Forrester study, 55 percent of marketers are concerned with the opaqueness of the intermediaries along the supply chain, up from 21 percent two years ago. No advertiser is immune to the supply chain realities, but seeing how the budget is allocated should be as natural as homebuyers scrutinizing loan origination fees from their mortgage broker.
There is a host of intermediaries in today’s programmatic supply chain including:
Not surprisingly, there are also several cost models:
The advertiser pays most of the fees, while in some cases the publisher, or both the advertiser and publisher, pay them.
It’s common to have an agent buy media on the advertiser’s behalf, only revealing the final price of a campaign, total margin, and fees. Just as common is the masking of the closing or winning bid prices.
This lack of bidding transparency is precisely what’s needed for optimization. This practice is especially prevalent among black box vendors, as is straight-ahead arbitrage. Without transparent insights into what improves targeting and conversion, marketers are flying blind.
So, what’s the average take rate of each partner? It varies of course, depending mostly on targeting strategies and pricing/profit models. But asking your supply chain partners exactly what they’re charging you is the first step in achieving total transparency.
Certainly one of the hottest issues in ad tech today, environmental transparency of an ad is as important as the campaign’s message or who’s being targeted. There are more mysteries than answers focused on who sees your ad, how much was seen, how long they see it, and where the ad showed up, but help is on the way.
In the early days of RTB, fraudulent or unviewable inventory was a common problem. Although challenges remain, there is an increasing number of new tools available for advertisers, publishers, and ad servers to detect bot fraud, fraudulent inventory, or unviewable ads.
Still, there’s no consensus on how viewability is defined. Standard bodies like the IAB and MRC are driving clarity on this issue. Many new vendors are trying to monetize viewability. Large holding companies have their own standards as well.
Advertisers are increasingly demanding that publishers bear the burden of proof by complying with imposed measurement of viewability-centric campaigns. Viewability-tracking fees, brand safety-tracking fees, and brand lift study fees are paid by either side in an effort to run cleaner campaigns. Although far from being solved, the use of ad verification and brand safety tools goes a long way in solving environmental transparency.
It seems logical that any data used in an ad campaign that you paid for would be accessible to you. But that isn’t always the case. Publishers could block the intent data or other data sets you would normally have access to with more transparent partners.
You may prefer to pay a black box provider because your only KPI is sales – this can work for some who don’t insist on understanding their true ROI. However, for data-driven marketing to work, seeing all your data for future learnings or to calculate your true ROI is essential.
Irresistible pricing models are as tempting as a timeshare in Tahiti. We get that. But regardless of whether you use a DSP or publisher tools for your programmatic buys, the more you know, the more you can improve outcomes – that is, if you want to know exactly how to improve outcomes rather than relying on your black box vendor to give you numbers devoid of margins or analysis.
Data are collected at every turn, every segment of the customer journey. CPC, CTR, and impressions are table stakes. For more intelligence, you need the eCPM and in-view impressions. Getting site-level reporting helps you blacklist/whitelist and improve targeting.
If you’re striving to get to your true ROI, knowing how the data points were calculated is certainly also part of the equation. Since we’re talking numbers, understanding the logic, math, and algorithms behind a bidding process is another must-have.
You should be able to decide exactly what success looks like for your brand. This means choosing your own KPIs, publishers, and the data you want to bring, buy, optimize, or analyze. Here are some best practices for how to bring more transparency to your programmatic initiatives.
Next time, we’ll dive deeper into the programmatic supply chain and how it affects cost.
2015 was a banner year for mobile.
Continuing its ascent into the status of omnipresent being, global smartphone adoption reached an all-time high last year and shows no signs of slowing down. Thanks to this rapid expansion of smartphone usage around the world, advertisers now have an opportunity to reach consumers even more easily.
We sampled the Marin Global Online Advertising Index, composed of advertisers who invest more than $7 billion in annualized ad spend on the Marin platform, to analyze data from around the world to create our latest annual benchmark report.
We uncovered three key findings:
For detailed information on 2015 search, social, and display mobile performance – including detailed data charts with YoY performance and further recommendations for 2016 – download our Mobile Advertising Around the Globe: 2016 Annual Report.
General conversion metrics about your visitors only tell part of the story. In reality, there are many steps a visitor might have taken before converting on your site. How do you measure the value of your upper-funnel prospecting campaigns, and determine whether they’re providing incremental benefit and driving last-touch attribution and conversion?
Assisted conversions help give you better insight for how other campaigns may have contributed to your final conversion. This insight is important, since it helps you make better decisions on your campaigns and immediately illustrates the value of your top-of-funnel marketing efforts.
Suppose you’re running a campaign where you’re targeting people who visited your website. You have another campaign that targets people who looked at a specific product page on your website, a much more focused group. You’re probably measuring how well you’re targeting website visitors, but you may not be crediting this campaign with any conversions that come from your product page.
In other words, your website targeting campaign alone looks like it’s not providing any value, although it’s pushing customers along
Here’s another example: Suppose your visitor sees or clicks a Facebook News Feed ad, and then clicks a web ad to convert. With general standard conversion metrics, the web ad gets the credit for the final conversion. But, in this scenario, your Facebook News Feed ad should get an assisted conversion credit, since it contributed to the “slam dunk,” as it were.
To read more about assisted conversions and how they contribute to accurate attribution, see Understanding Assisted Conversions.
“It makes my job a lot easier, and now I don’t have to spend all day combing through spreadsheet after spreadsheet, trying to figure out where a booking value came from because it’s nowhere in
– Kevin High / Digital Marketing Manager, IBC Hotels
IBC Hotels had a retargeting problem. Not only were they unable to easily attribute conversions – they were having a hard time even implementing their existing solution’s dynamic tracking code, and considered their vendor’s service team “unknowledgeable and nonexistent.”
IBC Hotels prides itself in introducing travelers to unique, locally owned hotels all over the world. Since IBC makes commission on each acquired booking, it’s crucial for them to accurately attribute the source of their conversions and revenue.
If they were going to lower cost and increase ROI, they needed a platform that would make their jobs easier, not more burdensome and clunky.
IBC implemented Marin Display – with its Site Tracking Tag – to build audiences for retargeting across channels and devices. IBC found Marin Display’s tracking solution worked flawlessly and was easier to implement than their previous retargeting solution.
The Site Tracking Tag allowed IBC to automatically collect important information such as order ID and revenue, and to easily attribute conversions. IBC could then effortlessly access this data and
From here, they were able to optimize their retargeting funnel, attribute conversions accurately back to their own internal reporting, and ultimately lower CPM and improve ROI.
Learn more and see full results in our IBC Hotels case study.
With the steady rise in remarketing as a digital advertising strategy, audience segmentation and activation has become a key tactic for digital marketers. What are some things that display advertisers should take into account when defining and streamlining their strategy?
Audience segmentation can be defined as a process of dividing people into homogeneous subgroups based on defined criteria such as product usage, demographics, psychographics, communication behaviours, and media use. Audience segmentation is now a major tool advertisers can use to tailor messages, improve targeting accuracy, and drive performance.
For display remarketing, a sound audience strategy is the foundation for a successful campaign, and has three elements:
To create a truly meaningful audience segmentation strategy, advertisers need flexibility in the tools they use to segment their audience. Segmentation methods also offer increased flexibility in what an advertiser can count as a user conversion, creating an extra dimension to audience creation.
Let’s explore four key segmentation methods that allow advisers to go beyond path-based segmentation or a one-size-fits-all remarketing vendor approach.
Query string is part of a URL that contains data that doesn’t fit conveniently into a hierarchical path structure. The query string commonly includes fields added to a base URL by a web browser or other application. This opens up a huge number of possibilities when it comes to audience segmentation. For example, here’s a query string generated after a user searched on a fictitious travel comparison website.
http://www.example.com/searchresults.html?checkin_monthday=13& &checkout_monthday=27& year_month= current -2&dest_id=United%20Kingdom& group_adults=2&group_children=2&no_rooms=1
Looking at this query string, we know the user is:
1. Looking for a two-week holiday from February 13to 27, 2016
checkin_monthday=13& &checkout_monthday=27& year_month=current-2
2. Interested in a UK holiday
3. In a party of two adults and two children
4. Looking for one room
Based on this information, we can now create audience lists based, grouping users based on urgency, demographics, and interests. And, our 1st party data set is fresh and reliable.
We can also count a conversion anytime someone visits a page with a specific URL query string: http://www.example.com?page=thank-you-new-user. In this case, we only count conversions from new users.
For example, suppose a user filters to view products from high price to low. It’s normal for these users to have a higher average order value per product than a user who filters from low to high. This may affect not only the amount we’re willing to pay to acquire these users, but also the type of creative we want to show them and which publishers we might want to target.
Recency refers to how recently a user last left your website or app. Creating remarketing lists based on recency enables a range of remarketing tactics.
It’s common for conversion rates to be high when a user sees an ad in the first few minutes after they leave your website, so make sure you’re highly visible during this time. Recency segmentation also allows different creative, offers, or calls to action based on how long it’s been since someone last engaged with your website.
Recency also allows for interesting cross-sell tactics. Say a travel agent knows that certain users are most likely to purchase travel insurance 30 days after they’ve booked their flights. Advertisers could use recency targeting to show travel insurance ads around this time.
Regular expression (regex)
A regular expression is a special text string for describing a search pattern. This allows advertisers to set up complex audience lists, such as one that matches multiple web pages, query strings, or products. Regular expressions also allow you to set up complex conversions, for instance, ones that match multiple-goal pages.
Say for example you want to create a list for users that go to the Caribbean section of your website as long as the subdirectory is in the second position. You can’t use ends with, or starts with, or contains; however, you can create this list with a regular expression.
^ A caret in a regular expression forces the expression to match only strings that start exactly the same way your regular expression does.
.* The dot could match any letter or digit. The star right after it matches the ability of the dot to match any single character, and keep on going so that it ends up matching everything.
Combining segmentation methods allows you to create sophisticated audiences that matter. By combining numerous segmentation methods, you can create an almost endless number of audiences to activate through remarketing.
To run the most successful remarketing campaigns, advertisers need segmentation tools that allow them to slice their audience in an almost unlimited number of ways. Currently, the number of advertisers using simple, path-based audience segmentation or a remarketing vendor’s standard segmentation approach is surprising. With tools that create and activate a meaningful audience segmentation strategy, you can build the foundation of a truly successful remarketing campaign.
This is a guest post from Dionte Pounds, Account Manager at
When building out a fully functional PPC account, it’s important to utilize remarketing lists in addition to your standard campaigns. Remarketing lists allow you to target individuals with ads that are already familiar with your brand because of a past interaction, generally an ad click leading to a visit.
These visitors are valuable because they’re usually further down the sales funnel. Remarketing is a great way to retain these past visitors, capture incremental volume, and shorten the gap between time of click and time of purchase.
If you’re advertising on a pay-per-click network (Google, Bing, Facebook, etc.), you’ve more than likely utilized remarketing lists to improve account performance. You can also improve your remarketing lists, specifically your Google and Bing lists, by segmenting your audience based on time of last interaction.
There are a few benefits to segmenting your audience by time. The first is that it breaks apart a very large audience into multiple audiences of very manageable sizes. This then allows you to bid more or less aggressively depending on the audience.
For example, you may want to bid very aggressively to get an audience of users that last interacted with your website one to three days ago back to the website. You may not want to bid as high for the people that last touched the site 25-30 days ago.
Using this method, you can place a bid on each audience that’s most appropriate. However, be conscious of the size of the main audience you’re trying to split. This practice is usually a better fit for more general touchpoints that generate larger audience lists. It isn’t always the best to break apart a very small audience pool because at that point, the lists can become too small to employ.
1. Create a new remarketing list
2. Select who to add to your list
Generally, I select page visitors. But there are options to select page visitors who did/did not visit another page, visitors of a page during specific dates, and visitors of a page with a specific tag.
If you’re more advanced, definitely utilize the custom combination option. I’ve used this capability to refine my segmented lists even further in the past and to block past converters from my lists.
3. Set the rule
Enter the page URL that you want to build your audience around.
4. Set the membership duration
Here’s where you can get creative. Go to the Tools drop down, then select Conversions and take a look at your attribution data. How long is the time lag from click to conversion? Use this information to set your membership duration for your audiences.
If you’re unsure, just use common sense to create reasonable durations. For this example, let’s assign the first audience a five-day membership duration.
After creating the first audience, repeat the process and extend the membership duration with each additional audience. Using the five-day example above as a starting reference, we can create three more audiences with membership durations of 10, 20, and 30 days.
In the end, instead of one very large audience, we have one broken up into chunks based on the account’s specific conversion history, which gives us more control over bidding and ultimately better performance. Using this method, we don’t bid the same amount for someone that last interacted with the website 30 days ago as a person who last interacted with the website one day ago. Try it out and see how it performs!
This is a guest post from Dionte Pounds, Account Manager at
Customer Match is an exciting new feature that Google recently unveiled that can greatly strengthen your ability to connect with an existing customer base. You now have the option to upload the email addresses of past customers or email subscribers directly to AdWords. You can then target that audience through Google Search, Gmail, or YouTube. Similar features are already available through AdRoll and Facebook. With Google’s newest addition, you can now leverage 1st-party data across yet another network.
This is fantastic news for all advertisers, particularly those in possession of large lists of customer emails who are looking for new ways to utilize that data to improve marketing efforts. Every marketer I know is looking for a better way to increase marketing efficiency, so this should really benefit all of us.
Google is allowing you to take what you know about your customers and use that to drive messaging across devices and platforms. This, in turn, allows you to build loyalty and repeat purchases among an existing customer base.
1) The first, and most obvious, way to use Customer Match is to stay in front of your customers. If someone has made a purchase from your business, these audiences can be used to target existing customers and keep your brand fresh in their mind. This encourages repeat purchases and leads to incremental gains.
2) The second is to re-engage past buyers who haven’t interacted with your brand in an extended period of time. Imagine Jane Doe bought a stereo in January and hasn’t purchased from your brand since. You can now create an audience specifically to target her, and individuals like her, when they’re logged into the Google network.
3) The third method is to create a negative audience. This audience is made from a group of people whom you do not want to see your ads. (Maybe you don’t want to risk overexposure, or you wouldn’t benefit from re-engaging this audience.) Businesses focusing on generating leads fit into this category. Customer Match allows you to create and exclude that audience from your advertising campaigns. As a result, you only capture new leads.
Setup for Customer Match is simple. Upload a .csv file containing hashed email addresses directly into AdWords. The larger the list the better, since audiences with fewer than 1,000 members won’t be targeted through any of Google’s networks for privacy reasons. Once processed, you have a new audience to target across devices and channels like any other remarketing audience.
The one exception here is the Display Network, since this feature is not yet compatible. For YouTube and Gmail, Google also creates a “Similar Audience” when eligible. This can increase overall lead volume by allowing you to target audiences made of new users who exhibit characteristics similar to your Customer Match lists.
Google goes to great lengths to protect user privacy, and this feature is no different. All data uploaded to AdWords must be 1st-party data. All email addresses must be hashed before uploading. Once they’ve been processed and matched to Google users, all data is discarded. This process ensures that all user information remains safe and protected throughout the entire matching process.
To summarize, if you have a large amount of 1st-party data, Customer Match is a feature you should definitely test. It’s simple to implement and can be used in a variety of ways across Search, Gmail, and YouTube. Since Google makes privacy a top priority, you don’t need to worry about putting any of your customer base at risk. Overall, the AdWords team has made a great improvement that makes it easier for businesses to enhance consumer relationships and brand loyalty.