Technology
Basis Platform
Advertising Operating System
Communication Portal
Intelligent Collaboration
All Channel Activation
Activate Media on Every Channel
Consulting
Resources
The customer journey is changing quickly, and it’s not all because of technology. Big data and the necessary legal regulations that follow have a huge impact as well. The General Data Protection Regulation (GDPR) is a new regulation that’s changing the way marketers can advertise to EU citizens. It’s the biggest regulatory change in data privacy in decades, and the deadline for compliance is fast approaching. Companies that don’t comply by May 25th, 2018 can face fines of up to 20 million Euros or 4% of annual global turnover.
Despite these threats, many businesses have done little or nothing to comply to GDPR. Hubspot did a survey in November and found that only 15% of companies had done anything to become compliant.
That’s likely because they don’t know anything about GDPR (36% of marketers hadn’t even heard of it in November) or they think it doesn’t apply to them and their business. That’s why we created this guide, to detail exactly what GDPR is, how it affects advertisers, and what changes you need to make to be in compliance.
|
Disclaimer: The GDPR is a 200-page document that covers data privacy reform for companies in a variety of contexts. Our guide is meant to illustrate how it can impact advertisers, but it’s not meant to be a complete resource on GDPR compliance. Use this guide as a starting point, then enlist the help of legal professionals to ensure your business is in full compliance. |
The GDPR is a regulatory act adopted by the European Parliament in April 2016. It’s aimed at protecting data and privacy for all individuals within the European Union, and addresses the export of personal data outside the EU.
Many advertisers make the mistake of thinking the GDPR doesn’t apply to them if they don’t have a business presence within the EU. But if your business processes any personal data of European residents, the GDPR will affect you.
Here are some of the main points businesses will need to address when managing the data of EU citizens:
Businesses must comply to strict record keeping requirements when handling personal data
Businesses must conduct privacy impact assessments
Personal data can only be used with the express (not implied) consent of consumers
Consumers have the right to be forgotten and a right of data portability
Businesses must adhere to communication protocols if consumer data is breached
Essentially, businesses need to get consent to collect personal data from their audiences, then take steps to ensure they handle it properly.
GDPR’s definition of personal data is very broad, including a number of online identifiers for profiling and identification that advertisers use. Here’s their own wording on the topic:
“Natural persons may be associated with online identifiers provided by their devices, applications, tools and protocols, such as internet protocol addresses, cookie identifiers or other identifiers such as radio frequency identification tags. 2This may leave traces which, in particular when combined with unique identifiers and other information received by the servers, may be used to create profiles of the natural persons and identify them.”
For your purposes, you can assume personal data to include:
Broadly, any information that can tie to a person’s identity is personal data. This includes information advertisers use to segment audiences based on interests, political leanings, ethnicity, etc.
Marketers are no longer allowed to use passive methods that imply consent for data use. Now consumers must take an action indicating that they are okay with their data being collected and used.
That means:
You can’t bury consent in the Terms & Conditions. It must be up front and center so consumers understand they’re giving consent.
No more pre-checked opt-in boxes that consumers can overlook. They must take action to opt in.
No more “By continuing to browse our site you accept our use of cookies” notices.
Consumers must be able to quickly and easily revoke their consent at any time.
Businesses must also be prepared to provide individuals with their personal data upon request.
For advertisers, GDPR impacts the personal data you can collect on consumers for ad targeting, how you store and use that data, and how you get permission to use it in the first place. Advertisers who use data science for search engine marketing need to take special care to ensure they’re in compliance.
The GDPR assigns responsibility for compliance to three main roles:
Data controller — Responsible for defining how and for what purpose personal data is processed.
Data processor — Groups that maintain and process personal data.
Data protection officer (DPO)
The data controller is responsible for ensuring outside contractors comply with GDPR when handing data, while the data processor is also liable for non-compliance with GDPR guidelines. Essentially, both you and the advertising platform you work with have obligations for data protection. For companies that store and process large amounts of personal data (e.g. banks or hospitals), a DPO is also necessary within the organization. This doesn’t apply for most advertisers.
Whether your organization is considered a data controller or processor depends on where the data came from. It’s an important distinction to make as it affects your responsibilities under GDPR.
If you’re advertising through Adwords or Facebook using data they collected from consumers for ad targeting, then they’re both the data controller and processor, and you have no additional obligations to protect data under GDPR. What changes is when you use your own consumer data with these platforms for ad targeting. When using conversion tracking cookies, remarketing ids, Customer Match and other data collected from your site, the responsibility lies on you to obtain consent and explain what the data will be used for.
Let’s look at an example for Adwords advertising. If you tag your site to build remarketing lists for search ads (RLSA), then Google’s the data controller, not you. If you upload a email list to run a customer match, then you’re the data controller and all responsibility lies on you to comply with GDPR requirements.
If you use Facebook for advertising, your obligations for GDPR will also depend on the kinds of ads and features you use. Facebook has its own guide for GDPR consent you can refer to. If you use the Facebook Pixel to collect additional consumer data for ad targeting, Facebook is both the data controller and processor. They’re responsible for protecting the data, all you need to do is clearly explain your cookie policy and gain permission from visitors to use their data.
If you upload a Custom Audience to Facebook, you have responsibilities to properly collect and protect data. They’re actually in the process of developing a Custom Audiences permission tool so advertisers can provide proof that they obtained proper consent.
Luckily, if you’re already GDPR compliant for Facebook ads, you won’t need to do anything additional for Instagram, since Facebook owns it.
YouTube ads work in a similar way. If you’re using remarketing ads, affinity audiences, in-market audiences, similar audience, etc., then you need to get consent to use consumer data. If you’re using YouTube’s internal targeting features, the responsibility is on them.
Even with an explanation, your obligations can be confusing. Just remember that if you collect any data from your audience for advertising or otherwise, you need to get permission, fully disclose what you plan to use it for, and take proper steps to ensure data management and security. Adwords, Facebook and other advertising platforms will do the same.
Even if you’re not currently using personal data for advertising, it’s best to be proactive and create a framework for compliance with GDPR standards. After all, transparency with your audience is a good business practice all around, whether they’re located in the EU are elsewhere.
Here are 5 steps you can take to start on the road to compliance before the May 25th deadline:
1. Audit your existing data
The first thing you should do is perform an analysis of your existing data to see how it’s already being used. Use this audit to develop processes to gain compliance for existing data and establish new practices to capture data.
Your audit should answer questions like:
Who are our data subjects?
Where do we keep their data? Who has access to it?
For what legal purpose do we have their data?
How are we processing their data?
You’re going to need to disclose to your customers how you plan to use their data and what third parties you might share it with. Therefore it’s important to start by mapping out where their personal data is held so you can be as transparent as possible.
2. Establish new practices for data collection
Next you need to change how you collect personal data from your audience. Start by changing the way you collect cookie data for site visitors. Be very transparent about what the cookies are used for, ensure your site visitors must take action to approve the use of cookies, and make it easy for them to opt out.
Do this the wrong way and you can end up ruining cookies as a data source for advertising. Many visitors will blanketly opt out if you present them with a binary consent option (cookies or no cookies). Instead, illustrate how cookies are helpful for their user experience and give them options for what kind of cookie data you can use. Here’s a good example:
There are tools available, such as Cookiebot, that can help you create custom GDPR compliant cookies.
You may also need to rework the opt-in forms you use on your PPC landing pages for lead generation. Say for example you use display remarketing to direct users to a gated lead magnet on your site. You may want to put an un-ticked opt-in checkbox at the bottom of the form allowing users to choose if they want to receive future marketing emails from your business. Or you could explain at the bottom of the form that by signing up, they agree with your privacy policy.
3. Create a data protection plan
If you’re collecting consumer data (not your advertising platform), then GDPR requires that you have a data protection plan for internal business processes. Draw out a clear data security plan for your business that’s in line with GDPR requirements. Here are some important points to consider:
The right to be forgotten
Under GDPR, users have the “right to be forgotten,” which means they can request their personal data to be removed from your databases or cookie pool at any time. Procedures should be in place to properly purge data if and when users want to be removed from your databases.
Requests for personal data
Users also have the right to request personal data they’ve provided to your company. You’ll need procedures in place so you can easily provide personal data “in a structured, commonly used and machine-readable format.” A CSV file should be sufficient.
Procedures for handling data breaches
Include procedures for addressing data breaches, so you can report necessary information to involved parties in a timely manner. GDPR mandates that data breaches should be reported to all consumers and respective bodies within 72 hours.
4. Review/revise your privacy policy
Your existing privacy policy may include clauses or language that aren’t in line with GDPR standards. Review your privacy policy and look for changes you can make, such as eliminating language related to implied consent.
In order to be fully compliant with GDPR requirements, your privacy policy should fully disclose to consumers what you plan to do with their personal data in a clear, concise, and transparent manner. It should address important areas like:
Who is collecting data?
What kind of data are you collecting?
Your legal basis for processing the data.
Who you plan to share the data with.
How you will use the information.
How long you plan to store the data.
What rights the data subject has.
How data subjects can raise complaints, request their data or request its deletion.
To get an idea of how to word your privacy policy for GDPR compliance, you can refer to templates provided by the EU GDPR Documentation Toolkit.
5. Seek Privacy Shield certification
Privacy Shield is a framework developed by the European Commission, Swiss Administration and the US Department of Commerce to develop a mechanism to comply with data protection requirements when transferring personal data for transatlantic commerce. Seek out and obtain certification under their standards. Do this before GDPR comes into effect to ensure you’re in compliance.
On the surface, GDPR can seem like a marketing challenge that hinders your ability to collect and drive advertising insights from consumer data. But fully complying with GDPR mandates actually helps ensure you’re marketing to the highest value leads when they do opt-in.
It may be by force of law, but advertising platforms and advertisers alike are making positive changes to ensure they market to people who really want to be targeted with advertisements. While these changes might reduce your pool of marketable leads, it does improve the quality and effectiveness of your advertisements immensely.
Now that you’re targeting a smaller base of engaged users, ad costs are bound to go up. That leaves even less wiggle room in advertising budgets for wasted ad spend. Artificial intelligence and bid automation tools to accurately allocate ad spend will become even more essential as advertising becomes more expensive.
There’s an endless number of options for creating a digital media campaign, but rarely does your client’s budget afford the luxury of an endless bankroll. So driving performance is key and the key to driving performance is optimization. Optimization is defined as the action of making the best or most effective use of a situation or resource and in the case of your campaign that resource equates to money.
In the generation of evolving technology, making guesses in order to optimize your campaign is no longer a viable option. There are too many risks involved, too much time required with manual analysis and the chances of success cannot be guaranteed. You want to achieve your campaign KPI as efficiently and effectively as possible. Enter machine learning optimization.
Our model-based technology leverages machine learning and a unique algorithm to maximize a tactic's performance. Basis collects data from more than 30 tactics parameters, at the brand level, to dynamically create models in real time to optimize towards a desired CPC goal. And it is constantly refining its model to best achieve that goal. By determining a unique price per impression based on the likelihood of meeting goals, the optimizer maximizes spend and saves time for buyers.
The optimization capabilities in Basis started with algorithmic optimization and with the introduction of machine learning optimization, media buyers can now opt for optimization with granular control or they can put intelligence in the driver’s seat.
With two cool optimization options in Basis, how do you know which to use?
| Algorithmic Optimization: Want the option for granular control? |
Machine Learning Optimization: Want to put intelligence in the driver’s seat? |
| Ideal to help you monitor the inventory in which your campaigns are buying. | Ideal for campaigns where you need multiple variables to be evaluated. |
| Choose from five different options to optimize against: CTR, eCPC, eCPA, eCPCV, and VCR. | Let the technology evaluate over 30 different targeting parameters to create an ideal optimization model for your campaigns. |
| Determine how the algorithm will work for you by setting the objectives and advanced control values. | Allow the optimizer to submit smart bids based on the probability of the impressions won, giving you the results you expect. |
| Work with the optimizer by modifying individual domain, exchange, or placement bids or status. | Permit your campaigns to learn from each other and aggregate success results to increase performance. |
There is an incredible amount of data available to media buyers and this level intelligence wouldn’t be possible for humans to produce on their own, let alone in real-time. Machine learning delivers a greater depth of knowledge and the ability to make buying and placement decisions instantly. The most significant benefit of machine learning for media buyers is its ability to continuously learn from data produced by any campaign. This allows the programmatic technology to react to changes in campaign performance and continuously improve results.
No matter how we’ve described machine learning optimization and Basis here, the best experience is to see it in action.
Today’s marketing and advertising leaders have more data than they know how to manage. Over the last five years, advancements in machine learning have equipped marketing teams with the ability to reach customers with on-target, precise messaging. But these same strides have the potential to bring chaos to the martech industry. Around the world, data regulations are becoming more stringent. In response to heightened needs for security and privacy, ad platforms are revamping their core technologies. Rising retail fragmentation makes it tougher for brands to strengthen and fortify customer relationships.
With this challenge comes the need for disruptive leadership. It is this reason why 50% of CEOs see their CMO as the primary driver of growth for their organizations, according to Accenture Strategy: “If CMOs focus on disruptive growth, rather than on traditional growth avenues, they have a chance to impact the bottom line in a heretofore unseen way—earning the key to that corner Chief Growth Officer (CGO) office.”
But only 36% of marketing leaders surveyed rank launching new business models as important priorities. This stat likely stems from a general trend, according to Accenture, that marketing teams are also likely to take blame if targets aren’t met. Even though marketers have a wide open door to unlock new growth channels for their organizations, they’re gridlocked. One way to overcome this challenge is to capitalize on deep funnel data which can help better predict a buyer’s intent.
Reach the right customer with the right message at the right time in the buyer journey. This goal is fundamental to all marketing—and the primary driver for how companies choose effective advertising channels. Deep funnel intent data tells marketers what their buyers are searching for and how to develop tailored, high-performing campaigns. At the same time, intent data “causes a great deal of perplexity for many marketers,” writes Sean Zinsmeister for MarketingLand. “Simply put, intent data is information collected about a person’s or company’s activity.”
Deep funnel intent data, also known as first-party data, “contains highly predictive buying signals, since the content is relevant to the purchase decision — for instance, which pages a prospect touched, links they clicked on and how long they spent on each page.”
External intent data, also known as third-party data, comes from publishers, social networks, and other data aggregators such as Bombora, The Big Willow, IDG, Connexity, and TechTarget. This information helps marketers gain contextual information about their audiences.
Deep-funnel intent data enables enhanced prioritization and outreach. With an understanding of an audience’s needs, marketers can develop systems for prospect prioritization, outreach, nurture programs, and more effective measurement.
But these frameworks only scratch the surface of what is possible in marketing. Intent data also enables personalization and targeted advertising. According to Viktor Mayer-Schönberger and Thomas Range in their work in the Harvard Business Review: “To compete against digital champions, [most business leaders] will have to overcome not just scale and network effects but especially new, data-driven feedback effects.” In other words, deep funnel intent data is more than a targeting tool. It provides a mechanism for building in-depth customer profiles through data-driven attribution. Intent data helps marketers distinguish between rudimentary clicks and the buying decisions that those clicks represent.
While it’s easy to collect a variety of data points from different sources to inform your bid strategy, deriving actionable insights from them is the real challenge. Online advertisers and search marketers need to be able to quickly merge and analyze downfunnel data from various sources, such as call tracking or CRM.
Despite these challenges, the value of deep funnel data to identify and better target high quality leads is simply too high for online advertisers to continue ignoring. Using a blanket strategy to bid equally on all kinds of keywords using a simple CPA model can lead to lost opportunities and wasted ad spend. A simple portfolio strategy can cause search marketers to overspend on cheaper keywords or overspend on expensive keywords. Instead, they should be using a customized strategy to reallocate funds to target the most valuable leads.
The information leads provide after they click through to your site can tell a lot about their value and better inform keyword bidding strategies. For example, businesses typically capture good indicators of quality in their lead capture form that can be used to identify trends in where the best qualified customers come from. Businesses can derive insights from metrics like company size, address, etc. Finance or insurance companies can use credit rating, income, address, and other indicators of lead quality, for example.
While most marketers capture this information to inform lead nurturing strategies, the same data can be used to improve ad targeting. Every data point represents an action from a human or machine. The future of marketing precision is built on an ability to make sense of it all, tying signals to user behavior. In this way, deep funnel intent data can help bridge connections between digital and physical touchpoints, allowing marketers to better understand their customers and increase their relevance so they can market to them better.
AdExchanger’s PROGRAMMATIC I/O conference was held in San Francisco on April 10th – 11th this year. It’s a time that advertising, media, and tech companies come together to learn about buying, selling, and executing all things programmatic.
My favorite topics and themes from my two days were:
Along with these themes were a handful of enlightening sessions, below:
According to a study done by Ad Perceptions, DMPs are having a moment where advertisers can’t recognize who they are. When shown logos of popular and the most recognizable DMPs in the industry, only 30% of respondents were able to identify them. What’s surprising is the same study showed agencies and brands work with an average of 4.3 DMPs. The larger the client, the greater number of DMPs they tend to use. This was rationalized by advertisers feeling they had to do their due diligence, or because of dissatisfaction with their current partners.
The top 5 criteria that advertisers want most out of a DMP are:
Presenter Joanna O’Connell from Forrester brought together the data to build a story that drives insight. According to Joanna, if you find your consumer one-to-one and don’t understand the context of which you are finding them, you’re not doing a good job. It’s no longer about finding your consumer one-to-one but rather about ‘one-to-moment.’ Every moment of a consumer is different than the one that came before. It’s not simply about finding the consumer, but rather about finding them in the moment AND understanding them. What that looks like for one person will look different for another. Find your consumers wherever they need you – whether that be online or in store.
Make your DSP work for you. Jounce Media says, “Your DSP does what you tell it to do, so tell it to do what you want it to do.” Its philosophy for training your DSP were around two main ideas of choreography and coaching. The core concepts of each are:
It’s no secret that Connected TV usage is growing per day. As an industry, we have seen connected TV grow by 10x in 2017. Traditional viewership has declined and people are not watching cable TV as much as they used to. The time for Connected TV is already here as 1 in 5 hours of TV consumption is done on a streaming device, and by 2020, half of all viewing will be via a Connected TV. But, it’s not as simple as saying that people are moving from one device to another, as many people are consuming on Connected TV and cable because while we are seeing a rise in Connected TV usage, 80%+ of US households still have cable (and many people are using both). The bigger story here is that for many advertisers, it justifies a bigger spend investment in this category.
There were many more insightful topics from the conference – some with a more immediate application (PMPs, issues with third-party data), and others more conceptual (blockchain). If you are a digital media professional, I highly recommend attending the Programmatic IO conference – the next one is in New York! It’s always valuable for me to hear from other industry experts and get to talk programmatic advertising.
Learn more about the Connected TV advertising opportunity with Centro.
Every business that engages in digital advertising wants to create engaging PPC landing pages that will connect deeply with their customers and drive conversion rate optimization (CRO). Of course, great landing pages are dependent on a myriad of different factors, including design choices, how and when you convey product value, relevance, and many more. All of these have to be in balance in order for your paid search ads to stand out as a pillar of your overall marketing strategy.
Of course, it’s not actually possible to distill the entirety of paid search marketing into a single sentence, but if you could it wouldn’t be too far off from this: CRO is all about guiding the lead through a compelling and value-added journey. Your search ads, landing pages, and CTAs must illuminate the path for your audience, otherwise they are likely to stray off it, try and build their own path, or travel down it in a way that wasn’t intended.
We all like to think of ourselves as mavericks from time to time, but in reality, people crave and appreciate guidance. This is not just true for consumer marketing; most of our relationships and interactions involve us being led through a narrative. What marketers can do is take advantage of this fact by structuring landing pages around familiar and recognizable touchpoints.
This is why you fulfill the promise of value at the beginning, because buyers are attuned to recognize when someone is giving them something useful and continue the relationship. It’s why you contextualize their pain point in the content, because people recognize and respond to that familiar yearning of wanting something more. And finally, it’s why you introduce that CTA at exactly the right time, as a way of saying, “Hey, now it’s your turn to make a move and continue this partnership.”
No matter how interesting and innovative your product is, people won’t just buy it for its own sake. You have to explain to them why they should listen to you, why it makes sense for them to buy it, and then exactly how they can buy it. That’s what guiding the customer through a buying journey is all about, and it’s what every successful landing page has to accomplish.
Generic marketing has its time and place: think about highly subjective video ads that succeed wildly at creating a certain mood, but don’t do very much to speak to particular buying situations. However, there’s really no room for generality when it comes to PPC landing pages.
For landing pages, specificity is always preferred, whether you’re talking about product descriptions, your CTA, links, or anything else on the page. When you get your targeting correct, then the specific messaging you employ is highly relevant to the users it reaches in a way that more generic material never can be.
The reason is that paid search visitors don’t arrive on a landing page just for the fun of it, except in rare occurrences. They navigate to your page because they have a specific goal or need in mind. Since generality isn’t driving their actions at this stage of their journey, generalized responses from a brand won’t be able to engage them in any meaningful way.
Yes, this means you will have to create different landing pages for various campaigns and categories of buyers. Utilize headlines that let the user know immediately that they’ve arrived at a page that speaks directly to what they were searching for. If hyper locality is a part of your marketing strategy, then put specific geographic information at the top so that it hits the reader’s triggers.
This landing page from IMPACT provides us with a succinct and pitch-perfect example of specificity in action. The headline speaks directly to professionals who want to increase the ROI of their blog and guides them into a solution for doing so.
URLs may live forever as long as the domain remains hosted, but that doesn’t mean that the information contained on them is always up to par. Numerous parts of your landing pages can change over time: Information becomes obsolete, new statistics are published, and links and images can become broken. If you direct a user to a landing page that displays any of these, you are putting up a serious red flag of unprofessionalism.
If you’re utilizing a landing page for a long-term marketing campaign, then checking in on the coding and the content at regular intervals is crucial to maintaining its effectiveness. Test links, make sure images load and are formatted properly, and verify the accuracy of any claims you have made. It’s the only way to ensure that every lead is presented with the optimal experience when they click on your ad.
Landing pages created for shorter-term campaigns also need to be monitored carefully. Once it has ended, you need to diligently remove any links to the customized landing page from AdWords as well as your internal site navigation. Few things will mar the customer journey more quickly than clicking on a broken link or being redirected to a landing page containing an offer that is no longer valid.
Almost every company now uses a site platform that features responsive web design, so there’s not much more to say about mobile, right? Not so fast. Employing responsive web design is an excellent start, but there is more to the mobile experience than simply formatting your landing pages to be readable on a mobile screen.
To understand why it’s so important to tailor the landing page for mobile users, just consider that mobile devices now account for approximately 53% of all paid-search clicks. These users now likely represent the majority of traffic on your PPC landing pages, no matter what industry you are in, and they deserve the same curated journey on your site that desktop users do.
The ubiquity of mobile screens has changed the way our brains respond to the browsing experience. People are now accustomed to content that unfolds vertically, and they expect high-quality images that pop on a mobile screen and fill up most of its space. They don’t want to have to pinch and zoom in order to access the value that was promised by the search ad. According to research from Adobe, companies with landing pages optimized for mobile triple their chances of increasing their mobile conversion rate to a minimum of 5%.
Check out this fantastic mobile landing page from Squarespace, a company you would expect to be on the leading edge of mobile optimization. Beautiful, well-formatted images, a relevant headline, and a clear CTA jump out at you as soon as your eye lands on the page, and they do a great job of layering buyer-specific value vertically.
Conventional wisdom says shorter sign-up forms are better at converting, because you want to get them in and get them out as quickly as possible. It makes sense when you think about it abstractly, because short forms are easy to fill out, and you want to make signing up as easy as possible. However, it doesn’t always play out that way in reality.
Multi-step forms have actually been found to convert more effectively than short forms, by up to 300%. There’s actually some very simple psychology at play that contributes to this dynamic. The first few questions each appear less daunting to answer one-by-one compared to filling out the entirety of a form in order to get what you need. By the time the respondent gets to the final questions, they will already be invested in the outcome, and you can move on to questions that have more substance behind them.
Keep in mind, however, that you can definitely go too far with multi-step forms. Try to stick to seven questions or fewer; when you go higher than that buyers start to feel like they are the target of an inquisition, and that your experience isn’t living up to its promised value.
~
To learn more about how you can build more effective, higher-performing PPC landing pages, connect with our digital media experts today.
Ask the Expert is a blog series from Basis where we break down the complicated tools, tech, and trends you’ve been hearing about in the trade pubs and around the office. We reach out to our in-house experts to ask the tough questions and turn them into bite-sized, palatable Q&As for your reading pleasure. This month’s topic: GDPR. We talked to Chris Coupland, Basis' platform operations manager, for the break down.
In the simplest terms, what is GDPR?
GDPR, which stands for General Data Protection Regulation, is a new law in the European Union governing the collection and processing of personal data of European member state citizens (data subjects). Under the GDPR, personal data that is used to offer goods and services, or to profile users, can only be collected for explicit, specified purposes, and the processing of that data must be compatible with those same purposes. There are only a few very specific legal bases for processing, most notably, through the consent of the data subject. In addition, data subjects have very broad rights, including the right to transparent information about the data collection and processing, the right to be forgotten (erasure of data), the right to object, and others. The intention of the regulation is to give data subjects more control over their personal data: who can use it, how it is used, who it can be shared with, etc. All companies that interact with European end users are obligated to comply with the law after May 24, 2018, regardless of said companies’ geographic location. Those that don't will be vulnerable to harsh monetary penalties.
Is this strictly about programmatic ad-buying?
No. The GDPR is designed to cover personal data regardless of industry.
Who is responsible for ensuring consumer privacy?
All companies that handle personal data should be responsible for ensuring consumer privacy. While the GDPR only relates to EU data subjects, other jurisdictions have their own privacy laws that should be taken into account as well.
How are advertisers going to be affected?
Every company that operates in digital media is unique because of business models, partners, customers, country operations, and many other factors. Basis recommends advertisers review the GDPR and seek legal advice applicable to their unique business model. In general terms, advertisers will need to ensure that their advertising activities are lawful under the GDPR when targeting EU member states in their campaigns. Advertisers that are collecting and processing personal data, and have determined that their activities fall within the GDPR's scope, will need to be certain they have a valid legal basis (such as user consent) for doing so. In regards to personal data shared with advertisers by Basis, we will be making changes to our terms governing the transfer of personal data in accordance with the new law.
How would the Internet user experience change in the E.U. member states?
End users may see an increase in solicitations of consent from companies that are actively collecting data. This may be a publisher, an Internet service provider, a device manufacturer or an app creator. The regulation covers a wide net. We’ll also likely see a variety in the ways for which this consent is asked.
Is this coming for U.S. Internet users?
The GDPR is now considered to be the gold standard in privacy legislation world-wide. It is expected that its principles will be emulated by other jurisdictions. As an example, amid the recent Facebook hearings in the U.S., two senators introduced the 'CONSENT Act' bill, which has very similar requirements to GDPR. Whether or not it is passed, the general industry sentiment points to momentum behind the idea of increased privacy protections. I think there will be a lot more evolution in this area.
What is Basis doing to meet GDPR requirements?
Basis takes privacy seriously and intends to fully comply with the GDPR. All processing activities are under review and we have engaged professional privacy consultants and legal experts to assist with the effort. Existing agreements, terms and our privacy policy will be revised to ensure compliance with the new regulations. Basis is also pursuing membership in the Privacy Shield framework -- a program founded by the U.S. Department of Commerce, and the European Commission and Swiss Administration to help companies facilitate transfers of personal data with their transatlantic partners.
What is my role in all of this as a media professional?
Get educated and get used to operating with transparency and consent-driven advertising. Know what your company practices are in handling data. Regardless of regulation, companies who are collecting data and are serving targeted advertising should be responsible for keeping user data safe and secure.
Interested in other Basis resources that will help you understand GDPR? Reach out to [email protected].
How will that sofa look in your living room? You could just measure it, check the color, and hope for the best – or you could use an augmented reality app to see exactly what your space would look like. Retailers like IKEA are hoping you'll opt for the second choice, and use a digital app that incorporates Augmented Reality, or AR, to see exactly what their furniture and accessories will look like in your living space. By using this emerging technology, the furniture giant and other brands hope to tempt consumers into purchasing items based on how they integrate into the buyer’s existing home and setup.
Augmented reality is not new; the AR-based Pokemon Go had legions of fans and players and allowed users to spot virtual Pokémon in real-life settings. While the game is no longer as popular as it once was (newer AR apps and games have edged this one out of the prime market share it once enjoyed), it is an excellent example of how easily consumers accept and adopt this emerging technology.
As brands feel pressure to be omnipresent, innovative forms of advertising and marketing that incorporate the always-connected consumers' own device continue to be in demand. Traditional outlets like television advertising continue to decline, pacing the way for brands to interact with consumers in new ways and to provide increasingly personalized experiences.
Augmented Reality is a set of technologies and tools that incorporate the real world with an overlay or enhancement of another item or image. From seeing how a new pair of glasses or new hairstyle would look to determining which chair works best in your dining room, AR provides a new way for consumers to interact with brands and items in their own familiar setting. By viewing the home or even themselves on a phone or other device, users can add an Augmented Reality overlay and picture a new item in a familiar setting.
Augmented Reality vs. Virtual Reality
Where Augmented Reality enhances the real, existing environment, a virtual reality setting creates a new world to engage and interact with entirely. For users of AR, only a few select elements are enhanced and added to the physical world or setting. Virtual reality plunges the user into an altogether different setting. While both AR and VR impact the way a user sees and interacts with the environment, AR does so in a more realistic and more seamless way.
Augmented reality can add a virtual overlay to the real world – whether that item is a Pikachu, an adorable Star Wars Porg, or a dining room chair. This allows consumers to see what owning the virtual item would be like or imagine that character or piece right in their own personal setting.
While the best-understood use of AR is to add items to the familiar environment and real world, it can also help eliminate unwanted items from view. AR-equipped glasses or phones can be used to eliminate items from view that do not meet your specifications. A trip through the grocery store is totally changed when only those foods that fit perfectly into your low carb, vegan or pre-diabetes diet can be seen. Looking for gifts for someone specific? AR can be used to filter out those items that do not meet the correct criteria, from price range to target audience or demographic. In addition to selectively highlighting those items that match a pre-specified data set (low-calorie foods, STEM gifts for teens, baby boy toys), AR can also integrate personal shopping data and history, highlighting those items that are most likely to be purchased and even generating promo or discount offers based on the preferences the buyer has exhibited in the past.
As more brands experiment with incorporating their own marketing materials, characters, and products into the existing environment, the ability to highlight specific brands and make others fade into the background is a more complex, but more powerful way to use AR to impact the consumer shopping experience.
What does the ability to highlight a specific feature, product, or character mean for digital marketing? For most brands, it is an additional opportunity to connect with consumers in an original and highly personalized way.
AR products that identify items by sight are making it easier than ever for consumers to get information. One of Google Lens’ most recent innovations allows users to identify items simply by looking at them or snapping an image. Want to know the breed of that cute pup you passed on the sidewalk? Look at it with your AR-equipped glasses and the information will be there waiting for you. Spot a cute pair of shoes? A quick snap or look will allow you to source them instantly. As brands like Google continue to evolve and use Augmented and Virtual Reality in innovative ways, consumers will have more and more options.
Making items instantly identifiable is just the beginning. Once the item is identified, the consumer can be directed to the right place to buy. In the retail setting, a custom incentive offer can be generated to accompany the information. Google is not alone – Apple’s new ARKit software for developers offers the same sort of functionality, providing everything from shopping history to real-time support as shoppers browse the retail store or setting. About 24 million AR-equipped devices are expected to be sold in 2018; this number is expected to increase to over 500 million by 2025, according to Bloomberg.
While AR requires a smartphone or other device to display, brands are launching the AR experience from printed media, online media, and in-app experiences. According to experts at AdAge, Augmented Reality could help revitalize the print marketing industry and further personalize the shopping and buying experience. From offering additional information, pop-up style to the integration of popular or branded characters into the media itself, print and augmented reality technology can be used to create an immersive and cohesive consumer experience.
The applications for using Augmented Reality are limited only by a brand’s imagination and willingness to invest in this still-emerging technology. The display of items and products in the consumer’s own space in the correct scale makes it far more likely that a prospective customer can proceed with confidence. Since they have already seen the item in place or in action, that prospect can purchase without trepidation, knowing they have made a good decision.
Augmented reality removes the barriers imposed by time and geography. If a buyer wants to see what a dress would look like “on” before purchasing from an online retailer, augmented reality allows them to do so. Once the item is seen, it can be purchased, worry-free. For furniture and home goods retailers, the ability to offer consumers a way to see items in their own homes can remove barriers and increase sales. Once the accessory, chair, or table is seen in place, the consumer is far more likely to purchase it with confidence.
Home retailer Lowes is already incorporating AR into its in-store navigation and remodeling programs. By offering consumers a way to explore what an upgraded bath, kitchen, or floor would look like in their own homes, Lowes is making it easier than ever for their in-store and online teams to close the deal.
Retailers like Rebecca Minkoff and Ralph Lauren are already incorporating "magic mirrors" directly into the consumer experience. For shoppers, a different size, color, or item can be viewed in the mirror without having to actually head out to the sales floor and pick out another piece. The clothing itself connects with the mirror and interacts with the image of the customer, expanding their options and in some cases, upselling additional items. This allows for a fully personalized experience – – and keeps consumers in the store for longer periods of time for each shopping session.
Augmented Reality is a natural match for social media, and brands are boosting awareness of characters, entire product lines and specific items by making them available for consumer use. Adding a favorite character to a photo, creating a branded filter that shows off the user’s own image with the branded item overlay ensures that both parties, the sender and the recipient or viewer, interact with a specific brand every time an image is viewed.
Augmented reality also elevates a branded social media image from a static one way experience to something more. When a consumer integrates a branded overlay into an image, the result used to be a singular image that definitely increased awareness but did not do much more than that. By incorporating AR for marketing, a brand can not only have a presence, but offer additional information or options for the viewer to engage with, creating a more interactive experience and allowing that passive image to become a strong call to action.
Incorporating Augmented Reality into an established digital marketing strategy gives brands some powerful new tools and innovative new ways to establish a connection with customers. By engaging in new ways, maintaining a presence right on a consumer’s own device, and offering convenient, try before you buy options, brands can harness the power of augmented reality for marketing.
Big data is like teenage sex; Everyone talks about it, nobody really knows how to do it, everyone thinks everyone else is doing it, so everyone claims they are doing it.
— Dan Ariely, Professor of Psychology and Behavioral Economics, Duke University
Information is the oil of the 21st century, and analytics is the combustion engine.
— Peter Sondergaard, Gartner Research
Everywhere you look these days, machine learning is in the news. A familiar buzzword, most people have heard it enough by now to know it has something to do with computers and algorithms, but that’s about it. But as we noted on this very blog, 97% of marketing influencers are predicting machine learning is the future of marketing. In fact, according to Google Trends, interest in the phrase has been increasing steadily over the last year, and right now machine learning has never been more popular. Other data supports this point. According to LinkedIn's 2017 U.S. Emerging Jobs Report, machine learning engineer is now the fastest-growing position. Furthermore, annual conferences dedicated to either machine learning or AI have now swelled to 243. And if we are to believe Fortune magazine, machine learning is no longer merely a trend, but a veritable revolution “electrifying the computing industry.”
So what’s all the buzz about? Why has machine learning become so popular? More importantly, why now? What is it about this point in time that makes it particularly ripe for a machine learning revolution? In this post, we demystify machine learning not only by defining it, but also illustrating how it evolved over time to fuel some of the most innovative technology today. In short, we show that all the buzz around machine learning isn’t just hype and that a revolution is happening in the computing industry being driven by machine learning.
Almost six decades ago, Arthur Samuel, widely regarded as the father of machine learning, first defined machine learning as a subfield of computer science that “gives computers the ability to learn without being specifically programmed.” While this definition is a good start, Alex Tellez, a self-described “machine learner” and author of Mastering Machine Learning with Spark 2.0, provides a more accessible definition. Alex defines machine learning as the “development, analysis, and application of algorithms that enable machines to make predictions and/or better understand data.”
But let’s unpack this a bit. What do we mean by make predictions or better understand data?
Normally, to solve a problem on a computer, we need an algorithm, which is simply a set of instructions that will transform an input into an output. But for some tasks, we actually lack an algorithm.
Take for example the problem of trying to be able to tell spam emails from legitimate emails. In this scenario, we know what the input is—an email—and we have a good idea as to what the output should be–a yes or no indicating whether the email is spam or not. But this is exactly where it gets more complicated: we still don’t know how to transform the input into the output because spam is constantly changing over time and from individual to individual. More broadly, as Ian Goodfellow, Yoshua Bengio and Aaron Courville suggest in their work Deep Learning, “The true challenge to artificial intelligence proved to be solving the tasks that are easy for people to perform but hard for people to describe formally—problems that we solve intuitively, that feel automatic, like recognizing spoken words or faces in images.” In other words, all of us intuitively know what spam is when we see it in our email inbox, but would we be able to describe it, i.e., draft a set of instructions for how to detect it?
This is where machine learning steps in. At root, machine learning allows us over time to transform input into outputs for tasks and applications for which there is no set of instructions, such as the spam detection task. Specifically, machine learning simply allows computers to learn without being programmed to do so and machine learning focuses on developing computer programs that can self-teach to change and grow when new data is introduced.
To see how this works in practice, let’s go back to the spam detection task. We can safely say that although we may lack algorithms for many tasks, we do however have massive amounts of data to help computers learn. Using data, we can begin to identify patterns or regularities that can help us to generate a useful approximation of the process. Assuming the near future doesn't change drastically, these patterns in data can help us to understand a process or even help us to make predictions that have a higher probability of being correct.
In the case of the spam email detection example, we can solve the problem with machine learning: a simple machine learning algorithm called Naive Bayes can distinguish between legitimate email and spam.
Machines can now perform complicated cognitive tasks that until recently only humans were capable of performing–such as driving cars, beating professional chess players, and even judging a song competition. In this sense, they’ve come a long way since the factory floors and manufacturing plants of the Industrial Revolution. But the history of a complex subject like machine learning, ironically enough, actually begins in many ways begins with a simple game of checkers.
Although Alan Turing had already created the Turing Test to determine if a computer has real intelligence in 1950, it wasn’t until 1952 that the first machine learning program was developed by Arthur Samuel. And by 1957 the first neural net was created for computers that simulated the thought process of the human brain. However, with the publication of his landmark study in 1959, “Some Studies in Machine Learning Using the Game of Checkers” Samuel introduced machine learning as a subfield of computer science.
While there was some progress between 1960 and 1989–such as the “nearest neighbor” algorithm that allowed computers to recognize patterns as well as Gerald Dejong’s Explanation Based Learning which enabled computers to analyze training data and create a general rule from it in 1981–things didn’t start to really heat up until the 1990s, when a paradigm shift occurred in the field, shifting focus away from knowledge and onto data. This is when scientists essentially started creating programs for computers to analyze large amounts of data and “learn” from the results and modern machine learning was born. Indeed, one of the most significant developments occurred in 1997 when IBM’s Deep Blue became the first chess-playing program to beat a reigning world chess champion at both a game and a chess match.
Fast forward to 2006, and Geoffrey Hinton, the man now credited with helping Google make AI a reality, coined the term deep learning to describe new algorithms that finally enable computers to “see” objects and texts in images and videos. Hinton went on to become a lead scientist at the Google Brain AI team and in 2011 developed a neural network that can learn to categorize objects.
Because Hinton’s postdoc students have all gone on to lead AI labs at Apple, Facebook and OpenAi, it should come as no surprise that in 2014, Facebook developed DeepFace, an algorithm capable of recognizing or verifying individuals on photos the same way humans can. And in 2015, Amazon not only launched its own machine learning platform, but Microsoft also launched its open-source Deep Learning Toolkit, which efficiently distributes machine learning problems across multiple computers. This tool kit allows almost anyone with a laptop and an Internet connection to become a machine learning expert, moving us one step closer to the democratization of machine learning.
Interest in machine learning took a dark side in 2015 when the Future of Life Institute published an open letter implicitly suggesting existential risk from advanced artificial intelligence–signed by both Stephen Hawking and Elon Musk along with 8,000 other AI and robotics researchers. However, a second open letter drafted in August of 2017 by Musk and 115 other experts to the U.N. explicitly warned of the dangers of using lethal autonomous weapons that are threatening “to become the third revolution in warfare.”
As Google Trends tells us, machine learning is at peak popularity and has virtually exploded in 2016 and 2017. But why is it so popular now, given that the field of study is at least 60 years old? What factors are converging to make this historical moment especially good for machine learning?
1. Mature Field: Both the identity and the methodologies of the field have matured in the last decade and that maturation has accelerated in the last few years. Although machine learning was once primarily a methodology under the larger discipline of artificial intelligence, it's now become a discipline in its own right because it's come to rely more heavily on the field of statistics. Moreover, the tools and methods used in the field of machine learning have also been maturing for the last 20 years.
2. Volumes of Stored Data: Arguably the single greatest factor contributing to machine learning’s mainstream appeal is the sheer abundance of stored data, which is rapidly growing. Machine learners simply have more data to “play” with, i.e., learn from. You often hear people complain now of “information overload” or “data exhaustion” largely because the systems and tools we use almost every day are generating data. And we’ve never collected data for individuals on this scale before. There are now groups such as QuantifiedSelf who are exploring all the ways you can track the collection of everyday information, such as heartbeats and even breath. If this trend continues, by 2025, we’ll be creating 163 Zettabytes of data every year.
3. Computational Processing Power: The simple answer is that computation is now also abundant and cheap. While only corporations used to have access to large computers with powerful processing, that’s all changed. With hosted infrastructure, you can now rent powerful computers for a few dollars an hour to run large experiments on immense data sets that you could never perform on a workstation or home PC.
4. Affordable Data Storage: And one of the most obvious reasons is that it’s simply become cheaper to store data. Data has to “live” somewhere. And with large datasets, it used to be very expensive to store data. No longer. Cloud-based machine-learning solutions from the big three public cloud providers: Google, AWS, and Microsoft make it affordable for almost any enterprise now to get involved in machine learning.
Machine learning is essentially a solution to more intuitive problems that lack a specific set of instructions, allowing computers to learn from experience and understand the world in terms of a hierarchy of concepts. Computers come to understand the world around them by defining concepts through their relation to simpler concepts. When computers accumulate knowledge from experience, rather than through a set of instructions, human operators no longer need to formally specify all the knowledge a computer needs.
In each case, the computer learns complicated concepts because of their nested relationship to simpler concepts. A graph drawn that would illustrate exactly how these concepts are built on top of each other would have many layers.
The image above, taken from Ian Goodfellow, Yoshua Bengio and Aaron Courville’s work on deep learning, illustrates just how difficult it is for a computer to understand raw sensory input data. In actuality, although humans are able to easily identify people and their faces, function mapping from a set of pixels to an object identity is very complicated: “Learning this mapping would be almost impossible if handled directly, but deep learning resolves this by breaking the desired complicated mapping into a series of nested simple mappings, each described via different layer in the model.” In this model, the input is called the visible layer because it contains variables we can observe. In between the output layer and input layer are hidden layers that extract abstract features from the image. We label these layers “hidden” because their values aren’t present in the data but must be determined by the model when it seeks to uncover which concepts are best at explaining certain relationships in the data.
This example can help us to see why this particular approach to machine learning is often called “deep learning.”
Machine learning is used in Financial Services, Healthcare, Government, Marketing and Sales, Oil and Gas, and Transportation among others. While there are many applications for machine learning, below I’ve provided three of the most popular:
Learning Associations—Retail Cross-Selling
In retail, a common application of machine learning is basket analysis. Basket analysis involves finding associations between products bought by customers and developing an association rule from statistical probability to enable cross-selling.
For example, if customer X happens to frequently purchase product Y, and if a customer X can be identified who doesn’t yet purchase Y, he or she is an excellent candidate to cross-sell product Y. More plainly, if X happens to purchase beer and customers like X also happen to frequently purchase chips with beer, then an association rule—70% of customer who buy beer also buy chips– can be developed via machine learning to enable cross-selling to X.
Classification—Financial Risk Assessment
Classification in machine learning builds on associations and goes a step further to identify group membership. A common application of classification is when banks try to predict in advance the risk of a bank loan. What is the risk that the customer will default and not pay the loan back? Which applications belong to a high-risk group and which belong to a low risk group? The way the bank calculates the risk is by looking for patterns in data about past customer loans as well as information regarding a customer’s financial history—income, savings, collateral etc– in order to make a prediction about the future. The bank fits a model of past data in order to calculate the risk of a new application, making a decision to accept or refuse the risk.
In this example, two classes are established, low risk and high risk, and the job of the classifier is to assign the input (the customer) into one of two classes:
IF income > 01 and savings is >then low risk ELSE high risk
Once a classification rule has been established the primary application is prediction. Assuming the future is similar to the past, if we have a rule that fits past data, then we’re able to make predictions about new instances in the future. In the case of a bank loan, the classification rule will enable the loan qualifier to evaluate a loan application with certain income and savings and quickly decide if the loan is low or high-risk. As such, a classification rule in machine learning is a tremendously powerful risk assessment tool for financial institutions.
Regression—Medical Mortality Prediction
Unlike classification, regression involves estimating or predicting a response or output value not from a membership in a group, but from a continuous set of training data. In other words, given a set of data, find the best relationship that represents the set of data.One of the most exciting examples of regression being used is machine learning for the healthcare industry. Machine learning algorithms can help medical experts analyze data to identify trends or red flags that may lead to improved diagnoses and treatment.
By analyzing the data from past cases to understand the risk factors that contribute to a certain patient outcome or diagnosis, these algorithms can ingest the data of a new patient and compare it to the models developed with the training set to predict the likely outcome. In the case of predicting clinical outcomes for patients diagnosed with a stroke, clinicians can use machine learning for creating diagnostic scores that will more accurately predict an outcome. According to a recent study, strokes account for “5.54 million deaths worldwide” and are the second commonest cause of mortality. A quick and accurate diagnosis of a stroke is important for immediate resuscitation. Using a free and easy-to-use “exploratory regression technique’, researchers were able to predict a 30-day mortality rate following a stroke in the rural Indian population that was 14% more accurate than existing scores.
It’s no secret: Digital is more complex than ever before. The industry has seen an explosion of tech, vendors, tracking metrics, cost types, and devices in recent years.
Considering our relentless focus on minimizing industry chaos, we partnered with research firm Ad Perceptions to survey more than 150 digital media professionals. We talked to the marketers who are tasked with making sense of it all every day, and we worked to uncover their pain points, expectations, and wish lists for programmatic advertising in 2018.
Want the state of the (programmatic) state? Download our infographic.