Chapter 7: Tracking a Mobile Marketing Campaign

In This Chapter

Understanding the measurement and analysis of mobile marketing campaigns

Setting up databases, profiles, and CRM systems

Collecting data and analyzing reports

Calculating return on investment (ROI)

Measuring and tracking the results of your marketing programs, including both your direct and mobile-enabled traditional marketing programs, are an essential part of your job as a marketer. Over the last few years, marketers have been put under a significant amount of pressure to demonstrate a return on investment for the organizational resources they consume. Return on investment (ROI) is the measurement of dollars received for every dollar invested. Moreover, lately, marketers are starting to measure return on engagement, that is, how much value is created with each interaction with a person. In other words, you need to show the value of your effort and demonstrate how those efforts contribute to meeting the company’s goals and objectives.

Mobile analytics provide valuable insights into the performance of your mobile campaigns. Mobile analytics refers to the process of measuring, monitoring, and tracking your mobile marketing campaigns. With mobile marketing analytics, you can

• Track individual user participation in all your programs by time, frequency, location, and other measurements

• Measure and compare all your mobile marketing campaigns, in some cases in real time, so that you can make immediate adjustments to your programs

• Use data to calculate your ROI to see whether you’re making more than you spend (a positive ROI) or whether you’re losing money on your programs (a negative ROI)

This chapter shows you a number of ways you can collect information and analyze it to improve your mobile marketing programs.

Determining What to Track and Analyze

Mobile analytics is the process of collecting data from your mobile marketing efforts so that you can compare a change in a single metric over time and compare multiple sets of data to each other. Mobile analytics impacts your mobile strategy in two ways:

• Analytics help you plan your mobile campaign strategy

• Analytics help you evaluate the success of your strategy

To gauge your mobile marketing success, you need to analyze which parts of your strategy work and which parts don’t. Measurable data collected for the purpose of analysis are often called metrics. A metric is any number expressed in a scale, used to quantify how much you have of something. (Think temperature in degrees Fahrenheit, or speed limits in miles per hour, for example.)

Marketers use metrics all the time in other media, especially digital, direct, and retail. Although you could track literally hundreds of metrics, Table 7-1 shows you a list of the main mobile campaign interactions and the metrics you should be prepared to track and analyze.

Table 7-1 Mobile Interactions and Associated Metrics

Interaction	Metrics Worth Paying Attention To
SMS	Unique participants in a campaign Number of respondents, opt-ins, opt-outs, and total churn (percent of people who opt out compared to the total list) Successfully and unsuccessfully delivered messages Carrier participation and which mobile carriers drove the most participation Participation by geography Count of embedded URLs visited Average revenue per user (ARPU) Total revenue per campaign
MMS	All the SMS metrics listed previously Number of unsupported devices Mobile operator MMS policy rejections
Content downloads	Unique downloads Total downloads Failed downloads Total revenue
Mobile websites	Unique visitors Repeat visitors Entry and exit points Bounce rates (number of visitors who see only one page) Number of page views Geographic location of visitors Devices used by visitors Mobile operators used by visitors Average time spent on a site, on a page Conversion rate
Mobile advertising	Cost per impression (CPI) Cost per click (CPC) Cost per conversion (CPC) or cost per acquisition (CPA) Total number of click-throughs (the number of times an ad is clicked on)
Voice campaigns	Date and time of incoming calls Number of calls Total minutes per call Billable minutes Average call length Number of transactions per call Revenue per call Error messages processed
Mobile applications	Number of downloads Download count by handset type Unique active users Failed downloads Time to download Total revenue Location of download User reviews of your application Time spent using application
Mobile e-mail	Click-through rate (CTR) Conversion rate Opt-ins and opt-outs Open rate Unsubscribe rate Forwards and shares
Bluetooth and Wi-Fi	Total delivered sessions Total content downloads Frequency and time spent per location The type of phone interacting with the system The mobile carrier used Location access point by location

The metrics in Table 7-1 are by no means an exhaustive list; they simply highlight some of the more prominent measures tracked.

Metrics aren’t just about measuring success. In fact, just as often, metrics help you recognize that parts of your mobile marketing are not working so that you can fix them.

Preparing Your Database to Collect Information

A mobile marketing database (see Figure 7-1) is simply a database that houses all the information about each and every mobile customer you have and all the campaigns and programs you run. A mobile marketing database allows you to link your mobile marketing campaign interactions with the people in your database and apply metrics to individuals and groups of individuals.

You’ll probably end up working with a number of databases, depending on how many mobile marketing companies you work with. Many companies support multiple mobile media paths, but no one supports them all. For a specific setup of programs, a database is typically built and hosted by your mobile marketing application provider. Or, you may already have your own customer relationship management (CRM) system or database that you work with.

In any event, whether your mobile marketing database is hosted by your application provider, is integrated with your CRM solution, or is some hybrid model, you need to set up your database files to accept the data you need to collect and analyze from each type of interaction.

The following sections show you the types of information to consider adding to your database for each customer or prospect who interacts with your mobile marketing campaigns.

Outlining demographic data

Demography is all about studying populations; consequently, demographic data are the data points that detail a population’s characteristics. You want to think about the various demographic data points you’ll find helpful for your business (every business is different). Data you might consider including in your database include

• Age (either in the form of a birth date or age range, for example, 14–24, and so on)

• Sex (male/female)

• Marital status (single, married, divorced, domestic partnership, and so on)

• Number of children (zero, one, two, three, more than you can handle)

• Level of education (high school, some college, college graduate, doctorate, life experience)

• Occupation (there simply are too many to list — isn’t that a great thing?)

• Income (typically presented in monetary ranges, for example, $50,000–$75,000 per year)

• Nationality (American, French, British, Chinese, and so on)

• Geography (residence, place of work, or if you’re a road warrior, American Airlines, seat B17)

In tech speak, all this data is commonly referred to as metadata, or “data about” something. When your engineer asks you, “What kind of metadata do you want to capture?” you can say, “I need demographic metadata — age, geography, and psychographic data (that is, information about your preferences, attitude, intention, and so on).” Now you’re cooking!

Organizing psychographic data

You may want to capture and organize psychographic data about members of your audience. Psychographic data is qualitative data used to measure an individual’s lifestyle. Examples could include

• Behaviors such as frequent travel

• Attitudes, interests, and beliefs

• Values and opinions

• Personality

• Purchasing motivation

• Usage behavior

You can include psychographic information in your database to help you describe and identify customers and prospective customers in more detail and to aid in developing promotion strategies designed to appeal to specific psychographic segments of the market for a product or service.

Planning for preferences data

Preference data is data volunteered by a member of your audience regarding his preferences, for example, and his likes and dislikes (favorite food, music, and so on). Other preference criteria may include the days of the week and the times the consumer prefers to be messaged or called, or how many times the consumer is willing to let you contact him within a particular timeframe (for example, ten times a month, but no more than three times per week).

By setting up your database to include preferences information and using it appropriately, you have a much better chance of pleasing the customers and prospects who interact with your mobile marketing campaigns.

Planning for behavioral data

Behavioral data includes information about past behavior or expected behavior based on past interactions or predictions based on other data. Including behavioral data in your database can tell you whether someone is likely or unlikely to respond to a specific campaign. For example, people with a behavioral profile that indicates procrastination might not respond to a limited-time offer.

You can get some great mobile subscriber behavior data from comScore (www.comscore.com) and Nielsen (www.nielsen.com), two research firms that track user behavior on mobile phones.

Behavioral data is an emerging field in marketing. Talk to your application provider and connection aggregator to get a sense of what they’re doing in this area or with whom they’re working.

Looking out for location data

Location is a dynamic data point that can be used to adjust your interaction with members of your audience so that those interactions are contextually relevant to their location. For example, if they’re browsing a mobile Internet site and your database knows a user’s location, you can display location-relevant advertising or guide her to the nearest location. You can also track location in real time and use that information to deliver location-based messages when your customers give you permission to use their location data.

Mining syndicated data

Syndicated data is consumer purchasing data compiled from individually scanned consumer transactions from thousands of purchase locations. The data is collected by market research firms, cleansed, and put into aggregate to protect the privacy of individuals. It’s then used by marketers to better understand the purchasing behavior of their target audience. Marketers use syndicated data to

• Improve sales tracking of their products and their competitors

• Monitor marketing promotions and merchandising for both their own products as well as those of their competitors

• Determine which distributors are using their products

• Segment customers effectively, identifying their best and worst prospects

• Perform market basket analysis, understanding what other products are typically purchased at the same time as theirs

Syndicated data can be very useful. In fact, it can often be better than data you collect yourself because it amasses information not just about your own client interactions but also those of your competitors as well.

Although data and information collection may give you knowledge and power, with this knowledge and power come great responsibility and liability. You must protect the data you collect. If you don’t protect it (if you misuse it, or if it gets in the hands of someone else), you may face many commercial and legal challenges. Seriously think through all the liabilities before you collect and use data. Be sure to consult your legal counsel. You can read more about the laws governing mobile marketing in Chapter 1 of this minibook.

Populating a Mobile Database

After you’ve identified the types of customer and prospect data that you’re going to collect and you’ve worked with your own internal team or mobile marketing application provider to build a database, you’re ready to start populating the database, that is, loading it up with all the information about your campaigns and your customers.

The following sections show you how to collect profile information about your customers when they interact with your campaigns. Then in the next section, you track the details of each interaction with your campaigns.

Collecting data through SMS

With SMS, you can collect some data automatically as well as request that participants of your campaigns provide the data. When a consumer opts in to your campaign via SMS, the mobile marketing application captures the participant’s mobile phone number. From this mobile phone number, your application provider can identify the following:

• Previous participation in other programs you’ve run: You can match the number to see whether the number has been used in other campaigns.

• Wireless carrier: Discover which wireless carrier the subscriber is using.

• Crude location: Using the country code and area code of the number, you can make a crude estimate of location (country, state, city, and time zone). This method should not be used for real-time location detection because it does not tell you where the person is, just where their phone was registered.

• Porting history: Find out whether the number has ever been moved from one wireless carrier to another.

• Technical bits: Discover whether the phone supports binary data (for example, pictures and video), text messaging, or WAP push. (WAP push is a method of delivering content and mobile Internet sites. Ask your mobile marketing application provider about this.)

In addition to the preceding data, you can request the user to submit any number of different types of data points, including demographic, psychographic, and preference data via SMS. For example, you may ask the user to submit her birth date as an opt-in challenge if you’re marketing a program not suitable for children. You simply need to make sure that your text messaging application supports the ability to collect the data you’re asking for and in the manner you’re asking for it.

Collecting through the mobile Internet and installed applications

You can use Internet and mobile Internet browser forms or forms in an application that you’ve downloaded to your mobile device to collect information from your prospects and customers by requesting that the visitor to the site complete the form. Upon submission, you capture the requested data.

For example, you can use web forms that can capture different consumer preference attributes, as shown in Figure 7-2.

You can use forms on mobile Internet sites, but be sure to keep them short. Mobile subscribers do not have the patience to complete long forms via their phone, so ask only the basics. If you need more info, augment the experience with another mobile (for example, voice) or traditional (for example, the Internet) path. But be careful. Make sure that you understand your audience. For example, not all demographics use all mobile technology.

Be sure to ask your application provider whether you can create your data schema and forms. Companies like iLoop Mobile (www.iloopmobile.com) have robust mobile customer relationship management elements that give you this capability.

Integrating CRM with mobile campaigns

A time may come when you want to merge your mobile campaign data with that data stored in your company’s customer relationship management (CRM) system. This is easy to do and typically can be handled in one of three ways:

• Manually: You can ask your mobile marketing application provider to give you a manual report (in Microsoft Excel or in an XML data structure) so that you can combine your data with that of the mobile marketing campaign database manually.

• Via data feed: Your mobile marketing application provider should be able to give you access to an XML data feed that you can pull from on a regular basis (once a day, every five minutes — the timing depends on your application needs) so that you can combine your data with that of the mobile marketing campaign database automatically per the set schedule.

• In real time: You can also ask your mobile marketing application provider whether it can pass you data in real time as your participants interact with the system (see Figure 7-3). For example, maybe you need to know immediately if someone opts out of your mobile marketing campaign so that you can update the other permission marketing management system in other parts of your company.

Courtesy of iLoop Mobile, Inc.

Tracking Interactions: Clicks, Calls, Votes, and More

Your mobile marketing application is capable of tracking most, if not all, interactions a mobile subscriber has with it. That includes any calls, text clicks, snaps, scans, StarStars (**), pounds (#), presses, pictures taken, acceptances, submits, votes, requests, replies, and more.

How this information is tracked and reported on, however, varies significantly by the type of mobile method you’re using. Moreover, you should consult with your mobile application provider prior to running your campaigns(s) to ensure that the data you need (that is, the data you want to use to measure and report on the efficacy of your program(s)) is being tracked and retained.

Using third-party tracking tools

You have your choice of a number of mobile analytics providers to track consumer interactions on your mobile websites and applications. For example, if you want to track interactions on your mobile websites, you can consider using Google Analytics (www.google.com/analytics). Google Analytics allows users to get rich website information, and the account creation process is simple:

See Figure 7-4 for mobile web report from another tracking tool, Bango (http://bango.com).

Courtesy of Bango

You can choose among a number of application provider reporting tools, including Flurry (www.flurry.com), Medialets (www.medialets.com), Motally (www.motally.com), Bango (www.bango.com), Adobe Omniture’s SiteCatalyst (www.omniture.com/en/products/online_analytics/sitecatalyst), Webtrends (www.webtrends.com), and others.

If your mobile application provider won’t let you integrate a third-party reporting tool, find another vendor. Third-party validation is incredibly important.

Obtaining metrics from partners and service providers

You should be able to obtain from your mobile application service provider a wealth of operational and business metrics from your campaigns. In general, obtaining metrics is a straightforward process, and as a marketer, you have to assess what mobile metrics are appropriate for you to measure.

Here is a list of the common metrics, tools, and questions you should be asking your provider about so that you can develop your mobile analytics strategy:

• Frequency of data available: How often are the reports updated? (Many providers only provide previous-day snapshots, not real-time data.)

• Summary levels of data (in addition to detailed transactional data from the mobile campaign): For example, does it provide you with a total SMS message count or mobile web page view count in a summary report?

• Specific data elements available: Ask your provider what data elements it can provide you, like those listed in Table 7-1.

• Sorting and selecting data capability: Can you sort and filter the data online, or do you need to export it and do it in a spreadsheet program like Microsoft Excel?

• Availability of campaign dashboard tools: Campaign dashboards are graphical web-based tools provided to marketers for manipulating and visualizing data.

• Charts and graphs availability: Similar to summary level data, can you get bar charts, pie charts, timeline graphs, and other visual representations of your campaign data?

• Customization options (alarms, triggers, views): In other words, can the system send you an e-mail notifying you whether your opt-out rate exceeds a threshold that concerns you, or if your text-message campaigns hit previously unreached heights?

• Integration capability: Do you have the ability to integrate with databases or other proprietary tools for data delivery?

• Help support availability: Is both business and technical help available?

• Privacy controls: Can you administer privacy controls, such as masking a user’s phone number out of reports?

• Process analysis (for user flows and consumer acquisition): Can you evaluate consumer engagement across different types of campaigns, such as moving from initial acquisition into a loyalty or customer service program?

If your mobile service provider cannot comply or provide the majority of these items, either as a standard offering or by developing a custom report or tools for you, you may want to consider finding another vendor who can help you.

Before you set up a custom report with a service provider or agency, be sure that you’re clear about who owns the information gathered. Also, be sure that you understand and are comfortable with its process for data security. You want to make sure that you own the data and that you can take the information with you if you choose to move on to another service, even if the information is collected through its platform.

Understanding analytics reports

An analytics report is a collection of mobile metrics presented to you in a single report so that you can see all your data as a whole and begin to gather insights from it. You can then use these insights to make decisions on how to best move forward with your business (see Figure 7-5). In other words, reporting and analyzing the data generated from your programs are absolutely critical components to your mobile marketing efforts. What good is the best mobile marketing campaign if you do not have strong insights in exactly how it performed or is performing?

Many of the companies in the mobile marketing space today can provide a high level of reporting sophistication. Campaign dashboards are often configurable so that you can see the metrics that matter to you and see them in a variety of formats.

Reports typically are summary views of a mobile campaign on a daily, weekly, or monthly basis, but they can also be unprocessed and unformatted data files that require additional work to arrange and present the data.

Every mobile application provider has a different user interface for pulling its reports. Ask about the reporting feature in the application as well as the mobile website reports available when you’re selecting a mobile service provider.

Courtesy of Mobile Behavior

You’d be amazed by how quickly data can be amassed as mobile subscribers interact with a mobile marketing application. Every click, snap, and call can be, for the most part, tracked.

Your application provider(s) no doubt has a lot of experience with collecting and analyzing transaction data. However, more often than not, the application provider is not in your business and doesn’t know exactly how you want to look at the data or what analysis of the data is meaningful. Rather than assuming that it knows what you need, tell it the type of results and data you want to see in the transaction reports. Have this discussion before your campaign launches; otherwise, you may have some late nights slogging through log files to get the answers you need. If you plan and coordinate with your application provider prior to the launch of your program(s), the application provider can give you the answers you seek in a perfectly packaged report, often in real time.

Mobile marketing reports can often contain both graphs and data for the metrics and be made available in a variety of formats (such as Microsoft Excel spreadsheets, Adobe Acrobat PDFs, or other formats). A current trend is to provide a visual representation, typically in graph form, so that a user can quickly see how his campaign is doing. Figure 7-6 shows a sample trend graph on the effectiveness of mobile advertising.

Calculating Your Return on Mobile Marketing Investment

Return on mobile marketing investment (ROMMI) is an important measure to keep track of because it lets you know whether you’re achieving your objectives and getting more out of your mobile marketing than you’re putting into it. You’ll find two basic ROMMI measures: a ROMMI against direct revenue–generating programs and a ROMMI against indirect revenue– generating programs. Both are discussed in the following sections.

Calculating expected ROMMI for direct revenue programs

Calculating your ROMMI for revenue-generating mobile marketing programs is a simple math exercise. The ROMMI for revenue-generating programs compares the sales you generated, or expect to generate, in revenue terms against your mobile marketing spending, or expected spending, that helped generate those revenues.

Here’s the formula to calculate your ROMMI for direct revenue programs:

[Incremental revenue attributed to mobile marketing (in dollars) * Sales contribution margin – Mobile marketing spending (in dollars)] / Mobile marketing spending (in dollars)

Here’s an explanation of each variable:

• Incremental revenue attributed to mobile marketing (IRAMM) is the total additional revenue, in monetary terms (for example, in dollars) you generate by running the mobile marketing program as opposed to the revenue you’d receive if you did not run the mobile marketing program. For example, if your revenues were going to be $22,000 before running the mobile marketing program and you estimate they’ll be $30,000 by adding the mobile marketing element to your marketing program, your IRAMM is the difference, $8,000.

• Marketing spending is the total cost for running the mobile marketing program, including all strategy, creative, tactical execution, and technology elements.

• Contribution margin is the calculation of the marginal profit (a percentage) generated from the product or service being sold. To calculate this, you need to know the selling price of the product(s) or service(s) that is contributing to the revenues being tracked and the variable cost it takes to produce it. With this information, you can use the following formulas to calculate your contribution margin. The calculation is a two-step process. You first must calculate the contribution per unit, and then you can calculate the contribution margin:

• Contribution per unit (in dollars): Selling price per unit (in dollars) – Variable cost per unit (in dollars)

• Contribution margin (percentage): Contribution per unit (in dollars) / Selling price per unit (in dollars)

Here is an example: Say that you’re selling T-shirts. The shirt sells for $10 and it costs you $1.50 to make it.

The contribution per unit (in dollars) is $10.00 – $1.50, which equals $8.50. The contribution margin is $8.50/$10.00, which equals 85%.

If you plan to use mobile marketing for the promotion and sales of multiple products, you can use the same model. You simply have to run the math for each product and add things up.

For example, say that you own a chain of retail outlets and you want to use mobile marketing to drive traffic to your store to generate T-shirt sales. You’re already spending money on a direct-mail campaign, and you want to include a mobile component by adding a mobile coupon call to action. The total cost of the mobile marketing element to enhance your existing marketing campaign is $5,000. You expect the mobile marketing element of the campaign to increase revenues from $22,000 to $30,000. You previously calculated that the contribution margin of the T-shirts sales averages out at 85%.

In this case, your ROMMI would be

[($8,000 * 85%) – $5,000] / $5,000 = 36%

In this case, your mobile marketing program is estimated to generate $0.36 per every dollar of spending. Not bad!

Calculating expected ROMMI for indirect revenue programs

Calculating your ROMMI for indirect revenue programs can be a bit harder than doing so for direct programs, but it is not impossible. The same premise applies as for direct revenue programs: You want to estimate the overall value of the mobile marketing program to your achievement of the company’s objectives.

Your goal for indirect revenue objectives is not sales, but rather some other trackable measure (opt-ins, impressions, clicks, traffic, redemptions, and so on). For example, maybe you want to measure the number of new leads in the opt-in database (opt-ins), total mobile advertising impressions generated (impressions), increased traffic in the store or on Internet or mobile Internet sites (traffic), the number of loyalty points redeemed by customers (redemptions), and so on. To calculate your ROMMI, you need to know what each of these measures means to you. Look at your historical data and calculate how much additional revenue an opt-in, impression, redemption, or traffic means to you in terms of real revenue. You can then compare the value of an indirect measure, like an impression, against the estimated/actual costs of running the marketing program to estimate your indirect ROMMI and see whether the program is worth doing.

Here’s the formula to calculate your ROMMI for indirect programs:

[Estimated value of measure – Mobile marketing spending (in dollars)] / Mobile marketing spending (in dollars)

Here’s an explanation for each variable:

• Estimated value of measure is the total revenue you’d expect from the metric. For example, say historically that you’ve calculated that every new impression on your website, on average, is worth $0.20 in expected revenue.

• Mobile marketing spending is the total cost for running the mobile marketing program, including all strategy, creative, tactical execution, and technology elements.

As an example, say that a travel agency wants to sponsor a live event. The sponsorship costs $35,000 (including the mobile marketing program costs). The agency assumes that the sponsorships will generate 250,000 impressions, which results in a cost of $0.14/impression ($35,000/250,000). The agency knows that historically an impression is worth, on average, $0.20 in future revenues.

The ROMMI for these indirect programs is

($0.20 – $0.14) / $0.14 = 43%

Based on these estimates, the sponsorship is worth the expense.