Post has attachment
Mobile data is more than hyper-local marketing

Hyper-local marketing offers advantages in targeting the right audience through mobile devices, that's why many brands concentrate their data-efforts on it and connect the client at a time when he is in physical proximity to their offices. Starbucks, for example, sends a message to customers, if the system detects their location next to a particular coffee shop.

On the one hand, this is a very useful and efficient method, on the other - mobile data-campaigns should not be limited with hyper-local targeting. Craze for smartphones, coupled with modern technologies, opens the endless sea of possibilities for marketers. So if all you do in this area is reduced to the invitation of customers to come to your store when they are close by - you overlooked pretty much.

Return to the madness. We all know (by personal example as well) that people today can not live without their gadgets. According to an independent research, 91% of US adults do not unhand their phones throughout the day, or keep them in the immediate neighborhood. On average, every smartphone owner looks at the screen (thus performing an action) 150 times per day. By tracking these actions, data-marketers can learn a lot about their customers / potential customers.

Let's consider the possibilities of mobile targeting on the example of the same Starbucks. You can simply keep track of customers who find themselves near your coffee shop, and send them messages with invitations and special offers. Also you can go a little further into the data and find out that the same device was fixed at Starbucks coffee shops in 4 different cities over the past 30 days. From this we can conclude that the device owner is an often traveling businessman, and target this user accordingly. After all, the approach to this client must be distinct from the consumer, who constantly comes in the same coffee shop. Not to mention the fact that the user will be interesting for the airlines and the companies leasing cars.

Smartphone owners use their mobile devices anytime and anywhere - on planes (over 80%), on the couch watching TV (86%), and even sitting on the toilet (75%). This means that Mobile Data gives marketers incomparably more information about target audiences than Web could ever make. The heyday of mobile Internet has changed the game, mainly because it allowed to learn everything or almost everything about life and interests of consumers. Thanks to mobile data brands can go beyond the online behavior of users and get some sort of a D-3 picture of their behavior in real life.

In desktop reality we used cookie-files in order to associate any actions of users with a relevant brand, product or service. Accordingly, behavioral segments and lookalike-models were formed basing on these actions. Nowadays marketers have the opportunity to go far beyond elementary targeting and predict the relevance of brands and advertising, using the huge amount of data in addition to the purchase history. It is also an opportunity for advertising agencies to reach a new level and start building brands.

So, do not forget that making efforts to be "hyper-local" for your mobile users, you can lose a huge amount of data-possibilities given by the mobile channel.

Post has attachment

Today, there are many misconceptions regarding big data and the things that make projects related to it, successful. To clarify and promote a better understanding of what makes the campaigns involved in data studying, profitable, let's take a look at some key moments.

Technologies. The biggest misconception is based on the fact that the main big data projects are tied exclusively to specific technologies - Hadoop, Python, Pig, Hive, etc. Of course, it's hard to argue with that, but they all are as important as useful for large projects. But if your company is not a startup, you probably already have the skills and technologies, which can also be useful. Recent studies of major projects related to the research data platforms such as Aster by Teradata corporation, made it clear that the companies can generate large amounts of data applications on existing programming languages, for example, SQL. In addition, the majority of players on the market, that have own data stores, as a rule, are more successful in working with big data, than those who have no such domain-specific information databases. Such analytical tools as SAS, SPSS, and R still did not loose their relevance in work with large volumes of data.

People. Similarly as you can use some of your old technologies, you do not need to recruit new people into the project. Representatives of large companies say that they do not seek to recruit only professionals of the highest level. Instead, personnel departments form teams of educated and experienced people with a high level of business expertise. Of course, companies have to constantly upgrade the skills of their employees in the use of such technologies as Hadoop and scripting languages. However, professionals working with big data are always in short supply.

Proper change management. This type of control is the key for the success of any project. Sometimes it seems that technical problems in handling large volumes of data, outweigh the scales of staffing problems. But this is not the case. Big data projects often include so-called "prescriptive analytics" - algorithms and analytical systems that point to the company's employees how to perform their tasks. As an example, we can take the largest US shipping company UPS that included Orion app in its service using collected data for creating efficient routes, and Schneider National application using sensors and GPS data to remind drivers to refuel. Both applications have made significant changes in the rhythm of the drivers, who can not ignore the accurate and reliable guidance.

Clear goal. It is common knowledge that work with the data is, first of all, sieving much information for finding promising solutions. This is a very important task, but its effectiveness will be reduced to zero if the company has no clear business goals. For example, telecommunication companies T-Mobile and Vodafone use big data technologies to get information about clients and the network operation. It would be a Sisyphean task, if companies did not set a clear goal to prevent customer churn. Having set a goal, the Australian subsidiary of Vodafone was able to resolve network problems, due to which customers go to competitors, in a few weeks.

Effective project management. Should the manager maintain a dialogue with all interested parties? Undoubtedly. It is undeniable, even despite the fact that the technical difficulties associated with the use of big data and the dominance of special terminology may cause some difficulties in establishing contacts between the management, customers and employees.

Of course, in addition to all of the above you still need luck. The projects that involve the use of large amounts of data, based on new technologies and new approaches, always entail a certain risk. Working with big data you will fail from time to time, and that is not really a problem, because mistakes help learn and not to stumble in the future. Today, work with big data is more applicable to the field of R & D - research and development. But those companies that are willing to experiment and combine traditional management techniques with the processing of large amounts of data, obviously, will be more successful than their competitors.

Post has attachment
A big-big secret: what do you pay for?

In the world of advertising technologies there is one not very pleasant and not quite secret: the cost of media buying includes a number of additional charges besides the inventory price.

Inhabitants of the ecosystem - from agencies to technology start-ups and even more advanced platforms - provide services on preparation of campaigns, creativity and data analysis, for which it is difficult to make a bill. They compensate for these costs, including them into the price of the sold inventory or charging a certain percentage.

Take, for example, advertising sellers. For them, preparation of a campaign for the brand is a complex process consisting of creativity, campaign setting on their system, placing of advertising, monitoring and optimization. And all this must be somehow taken into account for billing.

Agencies can sell the inventory on their own. If this does not happen, then they compensate their costs by charging interest on CPM or CPC model. This scheme helps brands to accept the need to pay for all service of the provider. They know that they need to pay for inventory, and they will get excellent functionality and measurable results with it.

The problem is that, together with the costs of services, which are officially hidden due to economic reasons, useful insights obtained during intermediate operations of the campaign, are left behind the scenes, too. That is, a kind of "media wall" is formed, which hides a huge amount of useful information, such as most effective types of combinations of creativity and messages, the audiences, time and geographic locations they fit best, etc. It doesn't mean that sellers deliberately keep this information secret, they have no malice. Just the "media wall" does not let the brands to see the complete picture.

A new theory of evolution

The need to overcome these obstacles will lead to either evolution or revolution, the purpose of which will be destruction of the wall and release of information resources. Here is a possible scenario scheme:

• New business models will provide advertising sellers with compensation of costs and at the same time will be completely transparent, allowing access to the data. This will be possible due to "consumerization" of platforms (when the functionality and effectiveness of technology platforms are at such a high level that brands buy them or buy access to the accounts from which they work by themselves). This scenario is beneficial for both sides: the supplier of technology receives money from the brands, and the brands have full viewability and control, thus solving the problem of the "media wall".

• Data providers will sell information without being tied to a particular inventory. If they do not sell audience and advertising, they can sell insights for search, display and social media advertising campaigns. Media providers know everything about the audiences on the channels they work with. But marketers need a full review, they want to know what kinds of content and creativity cause the greatest impact, regardless of the digital channel. Cross-channel, independent of media information will expand the list of key competences of both brands and agencies.

• Brands will appreciate technologies of advertising platforms much more. While data-insights are hidden behind the "media wall", a platform has tactical, not strategic value. In the current model of analyzing, only specialists responsible for operations with advertising and campaign management, are the main users of technological advertising platforms. But as soon as the new model provides access to more detailed information about customers in all channels, the value of the platform and information it helps to get, will increase. With the new model, even the analytic directors, strategic planners, vice presidents and directors of marketing will want to get access to information about customers and use the information obtained.

However, classical media model will not be completely out of use, but digital industry players will need to search for and find a solution that provides access to interactive, practically valuable information and control of campaigns. Most likely, technological companies will become the main driving force of this evolution / revolution.

Post has attachment
Who owns your data?

A luxury sports car with missing parts in the engine seems to be perfect until an experienced mechanic looks under the hood. The same thing happens with marketing metrics.

According to a recent study by Duke University, only a third part of the companies managers is satisfied with marketing analytics services and see their contribution to the overall profit.

Despite the increase in the amount of data collected across all marketing channels (and perhaps just because of it), companies can not make a complete data-image, which could help them in making advertising decisions.

"Marketers have a lot of data, but very few really effective insights," says Jennifer Zeszut, CEO of Beckon, a provider of software solutions for marketing analysis.

We should search for a solution to the problem in the interaction between the data acquisition channels. According Zeszut, marketers, having the huge amount of data, have a very vague idea of how to use it.

Teradata survey, conducted among 2 200 marketers, found that 65% of respondents see the problem in the data fragmentation, which does not reveal clear patterns in the behavior of people on the consumer funnel stage.

"A user has not converted, simply by seeing a link on Google page," says Ran Sarig, CEO of Datorama, a cross-channel analysis platform for advertising buying. "In reality, there is a variety of other interactions with the brand, which leads to customer engagement."

The problem of data transparency from different channels is of particular relevance in view of today's fragmented digital marketing. It turns out that the most relevant information is often stored in the private data bases of agencies, DSP, Saas-providers and other third-party systems.

"Brands want fast, personalized marketing solutions, and to find them, they need the appropriate data, which is in the hands of agencies," explains Justin Honaman, former manager of the sales and marketing department at Coca-Cola and the current commercial consultant at Teradata. "Agencies can not always provide the desired information quickly and do not always want to share it, if they consider themselves to be the owners," he adds.

That is why marketers need to specifically discuss the issue of data ownership with their partners. Mark Zagorski, CEO of DMP eXelate, says that he observed conflicts between agencies and data-platforms, working on the same client, many times.

"Especially in cases when data management takes place outside the purview of agencies, the latter see a threat in it," said Casey Carey, senior manager at Adometry, development company of attributive solutions. "Over time, marketers will gain the experience of working with data and learn to skillfully build relationships between all parties of the process."

The task of marketers is to make sure that agencies and service providers have weakened control over the data they collect, and not be afraid of information leakage.

"Many brands do not have access to information at platforms, on which their ads are placed because agencies with which they work, are afraid that brands will work directly with the sites or through in-house solutions. But transparency of data would be a prerequisite in any case, and those who resist it, eventually will be outsiders", adds Carey and tells how one of his clients - the largest cable company - put forward an ultimatum to its partner agencies, urging them to participate and support during analytical programs and in the process of adoption of certain marketing decisions.

This is the right approach, and it is best to reinforce such an agreement documentary, especially when it comes to long-term cooperation with the data-providers.

"Customers are often under control of agencies and other third parties that's why it is important to specify all the conditions for cooperation in advance," said Mike Lempner, executive director of customer data in the consulting company Infinitive.

Jay Stocki, vice president on online marketing at Experian Marketing Services, agrees that this is a key point. He says that many agencies believe marketing data to be their intellectual property, as creative processes.

"In such cases, you need to make a separate item in the partnership agreement, explaining that the customer retains ownership of all data, including data from a third party and the data collected in the course of the campaign," recommends Stocki.

However, it is the only one of the obstacles for marketers to overcome. Both Stocki and Lempner note that the technical limitations on the part of agencies can make the process of real-time transferring of the right data to marketers, very costly or simply impossible.

"Even if you are the owner of all the data, access to it is a different story," sums up Lempner.

Post has attachment
As part of the conference series ClickZ Live New York, which took place this year, there was a round table discussion which participants tried to get an answer to the burning question: Does Big Data work on marketing, and how exactly?

Big Data is not a new-fangled marketing term for a long time. In place of these once mysterious words other buzzwords came - "marketing content cloud", "second screen", "gamification" and others.

First we need to understand whether small pieces of data or any of their analysis work on marketing. Too many experts, despite the emergence of data technologies and integration of social networks and targeting with CRM solutions, are not able to optimize their local data processes, not to mention the processing of large arrays of multi-structured data in real time. Therefore, you should start with the basic volumes, gradually adding more data and dynamics into your marketing mix.

ClickZ Live round table experts identified 5 aspects of advertising campaigns which can be affected by correctly used big data.

1. Segmentation. With access to updated array of user data, you can abandon the classical list model and start working on a 1: 1 one (personalized, customer-oriented marketing). If you are psychologically ready for this step and have the technological basis for the collection and processing of relevant data in real time - go forward to 1-2-1 marketing! It will fundamentally change the rules of the game in general, and methods of Big Data usage in particular.

2. Prototypes. In digital marketing, this term means a specific set of user characteristics representing this or that group of the target audience. Despite the fact that marketers today often prefer placement based on visitor statistics and the interests of the audience instead of this principle, prototypes are still playing an important role in modern marketing. Big data helps generate more accurate and dynamic prototypes, which consequently increase the chances of hitting the target at audience buying for any targeted campaign.

3. Social Analytics. Social Networks are an ideal field of application of big data, even if you do not use it when dealing with other aspects of the campaign. Organize fragmented information from posts, comments and tweets to understand your customers better and apply to each maximum personally.

4. Optimization of the communication channel. Optimization of each individual channel, not to mention the multi-channel optimization at the consumer level, is a very difficult task. Big data can help marketers understand what is important at this particular moment. Big Data can actually be helpful, if a marketer wants to understand which customers are the most valuable in each channel. Platforms, through which Big Data is managed, can handle both structured and unstructured data. So market researchers really need to include such statistics as Webstream, Clickstream, and social data in their analysis.

5. Native advertising. Can Big Data help to quickly collect, organize and display content? Can it transmit to the "management center" information that would help to increase the efficiency of offers in the key points of interaction with customers? Yes, it does. Actual, updated data can undoubtedly play a key role in the formation and development of native advertising. Of course, working with big data and algorithms involves the risk of over-reliance on automation of processes, what, in theory, could hardly fit into the concept of content marketing. But you should not be afraid of errors, it is much more interesting to learn from them.
Wait while more posts are being loaded