So now that we understand what data analytics is and why it's important and what we need to survive, let's talk about some simple use cases to get you started.

Before we go there, I think it's important to understand “the why” of our members. Not too long ago, I was watching a TED Talk by Simon Sinek. He was talking about understanding the why of what you do and he describes a golden circle. In the center of the circle is the word “why.” The next ring of the circle has “how.” And the final ring says “what.” Sinek talks about how most people communicate from the outside in. They'll start with the what, tell you the how, and then get to the more fuzzy why. Then, he gave an example of what would happen if Apple were to communicate this way. Apple would start with the what: “We sell great computers.” Then, they'd move onto the how: “We use only the best parts and are really careful to make them user friendly so people want to buy them.” But the why is where things get murky in this process.

Sinek says he's noticed that really effective communicators and really effective companies start with the why and work from there to the what. In the case of Apple, the why would be: “We think differently at Apple.” Then, the how would be tied to how they think differently. Perhaps the response would be: “We challenge the status quo. We use only the best parts. We have the idea that our software has to be intricately connected to our hardware.” And the what is: “We make people want to buy a computer.” As you can see, inverting the circle changes the whole conversation. That makes me think what our members and customers get out of bed in the morning. There are a few things that you can do to focus on the why—and they all come down to analytics.

Look Ahead

In 2008, the world crumbled around the financial institution I was working for. It was a horrible time. We lost lots of money and we began to realize that our collections department was going to have to grow. Our leadership at the time, who was absolutely brilliant, had the foresight to go out and bring in 80 collectors, increasing the collector staff about 16 times. As you might imagine, this was a shock to the system.

In 2008, the mortgage bubble finally popped when people discovered that homes weren't worth whatever they had paid for them at the market peak. There was moral corruption in the system, people were packaging bad loans with good loans and selling them out, and everything finally burst. People were losing their homes and their jobs.

Well, one of the things that we were trying to understand at the time was how to get ahead of people who were losing their loans or were not paying their loans. How do we determine that someone is in a bad financial position before it happens? And finally, one day, I came up with a simple solution.

Most people prefer to use direct deposit for their paychecks. I don't know about you, but I haven't cashed a check at a bank in a long, long, long time. So when that direct deposit comes in, it's usually going to be deposited into the place where a person also has loans, because then it's easier to make those payments.

In 2008, direct deposit rates started to drop, and I began to wonder why. I thought, “If someone's direct deposit stops, it's likely, in this particular environment, that they have lost their job.” There are other reasons, but this is a good one for us to start with. This is important information for the bank to have. It is one of the indicators that I added to our list of data analytics to be able to identify and help customers who need it. And if you think about it, it's a great opportunity to make a phone call.

Imagine how that call would go. Bear in mind that there are other reasons direct deposit might stop, so it's important to approach this tactfully and respectfully. At the bank I worked at, employees said something like this: “Hello, Mr. or Mrs. Consumer, I noticed that we're no longer receiving your direct deposit and we were concerned that either there was an error or perhaps that you have found another financial institution, and we wanted to see if there was anything we could make things right.” If this person has lost his or her job, you'll probably get a vague response, something like, “Well, I'm no longer with that job.” It's possible that the person might be getting a new job, in which case you have the opportunity to solicit their direct deposit so you could make it easy for them, send them a direct deposit form, or send it directly to their employer.

It could also be that this person has lost a job and doesn't have a plan for a new job yet. You might look at the loans and say, “You know, we have a program called Skip a Payment that allows you to skip a loan payment. We will reorganize the loan or we'll recalculate it, reset the rate, and give you some time so you can get through this.” When a bank calls in the midst of a stressful time and says, “I'm here to help you,” customers will remember it for the rest of their lives.

We've covered two reasons why customers might stop direct deposit: (1) a job loss, or (2) a job change. There are other reasons, such as that they are moving to a different bank. That could be a sign that they will be moving other accounts as well. That's also worth a phone call, because it's worth understanding why they're leaving and maybe even recovering the business. If a bank doesn't recognize this trend and act on it, the customers will be proven right.

Here's a fourth one, one that's not commonly considered. What if a horrible tragedy has occurred and the customer is deceased? That's also worth a call, to help the surviving family get the finances in order.

Finally, the fifth reason why a direct deposit might stop: human or technical error. What if the person at payroll at the customer's job forgot to process the payment? What if something went wrong with an employer's payroll processing system? What if the incorrect hours were logged? What if there was an error at the bank that prevented the account from being credited? It would be fantastic to get a call from the financial institution that said, “I notice your direct deposit didn't hit this month or this week and it usually does.” And when an organization does, it can help its clients.

No matter the reason, if a direct deposit stops, it's worth a call to the customer. This is such a simple thing to track. It's easy predictive analytics because the outcomes are so easy to understand. I can predict that if someone's direct deposit stops, one of these five things has probably happened.

To do this, someone working on strategy and data analytics has to say, “Hey, I need to identify all the direct deposit accounts,” and then the bank begins to track them. The names that are identified should go on an outbound list for your call center along with an outline of how to approach the conversation. This little thing could be the difference between maintaining a loyal customer and losing a loyal customer. So there's one simple thing that banks can implement—when they know the why.

Credit Card Usage

Credit card usage by vendor can also be incredibly useful for actionable data analysis. I'm fond of asking banking and credit union CEOs, “If I were using your financial institution's credit card or debit card to buy all of my airline tickets, and all of a sudden I stopped, would anyone from your institution call me?” The answer is almost always no. No one would call; no one would even notice. This sort of income, which we call interchange, is very valuable to financial institutions. This is just one piece of data that is tied to a larger data-analysis strategy that tracks customer trends and behavior. When steady customer behavior falls off, banks have an opportunity to find out why and make some changes.

Tracking travel is very easy because the consumers actually tell their banks when they're going places because they want their credit cards to work at their destination. When they're leaving the country or traveling out of state, they tend to let financial institutions know because our neural networking, otherwise known as artificial intelligence fraud algorithms, will prevent their credit cards from working if they are doing something that is out of the normal behavior of them, which would include traveling to a foreign country. Why wouldn't FIs also track what customers use their cards for when they travel and encourage them to use their bank cards? Provide them with personalized rewards.

This is another simple use of analytics. But more importantly, it is a chance to pick up the phone and talk to this person. I'm betting that, more often than not, banks would hear that this particular consumer has decided to use an airline credit card or some other rewards-based card that gives them a perceived value that, in their mind, is better than the FI card they were previously using. However, usually these cards come with fees and a higher interest rate. If you were to just do the math and say, “That's great that you've got a new card that gives you two free bags and extra miles. But the $400 a year fee combined with the higher interest rate, will cost you more than our card, which has a 10 percent interest rate. With the money you save, you could buy another three plane tickets.” Unfortunately, most banks don't have a way tell these stories—and as a result, they let the business escape. Data analytics lets banks identify these customers and keep their business.

General tracking of credit card usage by vendor can also be useful. It can help a bank pinpoint vendors that everyone in an organization or demographic relies on. For example, Amazon, iTunes, and Netflix are hugely popular retail and services sites that are pervasive in almost every demographic. It doesn't take heavy-duty data analysis to identify these organizations and determine the penetration by measuring how many of your FI's cards are being used by these retailers. The next time you are at one of those big meetings where all the managers are sitting around circular tables, stand up and ask everyone in the room who has an Amazon account to raise their hands.

Based on my experience, that at least 90 percent of the room will raise their hands. Then ask them to raise their hands again if they have a credit card or debit card from the institution that employs them in the Amazon account as a form of payment. Unless you work at one of the major players (e.g., Chase, Citibank, Capital One, Bank of America), it's likely that only a small percentage of your own staff are using your credit cards in these digital providers. The question to ask is, “Why?” Are they motivated by rewards points? Are they motivated by security? Are they reacting out of habit because they started using this card for online purchases a long time ago and now it's just muscle memory? If so, we have to understand the why to break the habit or find the right motivation. It's always important to track down the why.

Recently, my team and I were working on statistical analysis of purchases from a credit union, when we discovered a cache of $40 transactions from Netflix. These transactions fell out of our analysis as outliers. After all, how would one even incur a $40 charge from Netflix? The usual charge is around $12. I could understand one or two such charges out of a few thousand transactions, but this file had hundreds.

After reviewing the file with the credit union, we determined that it was fraud. I started to wonder what other outlier transactions from subscription services we could find by using a simple routine to track charges that should be consistent. My list included Spotify, Apple Music, Hulu, DirectTV, and many others. Tracking and understanding outliers will be a new feature for financial institutions in the future. In the same way that neural networking is used to stop credit fraud by watching for transactions that don't match your patterns, new techniques using the same technology will be used to determine if customers are at risk of defaulting on a loan or are currently involved in a crime, or are in the midst of a divorce. The outlier approach is another simple win that could be implemented at any financial institution to reduce fraud and provide a service to their customers.

Usage Monitoring

Not long ago, I enrolled for an account on a website of a web-based productivity tool. Let's call it Happy Go Lucky. I was visiting the site pretty regularly for a while, and then one day I finished the project and didn't really have a need to continue to go there anymore. About a month went by, and I got an email from Happy Go Lucky, and it said, “Dear John, we miss you. When are you coming back?” And I thought, “Aw, they miss me.” Then I realized that the folks who run this website have realized that my usage had dropped off.

I frequently asked CEOs and CTOs if they monitor usage of the digital channels, and if someone stops using it, whether they contact their customers to determine why they stopped using the site. The message shouldn't be cheesy or out of sync with your bank's messaging, but it should encourage them to come back or, more importantly, to try to find out the why of why they stopped visiting. After all, if we can't figure that out, then we're in bigger trouble than we thought.

These are great use cases for analytics. Don't be intimidated; data analytics can be an easy way for you to start looking at your data and maybe even seeing some direct outcomes from putting these easy use cases into use. People don't realize how simple it is to make a phone call to just ask a question to understand the behavior of other people.

The key to understanding why is to be willing to challenge the conventional wisdom of the industry and your organization. There are so many common tropes in the financial industry, but data analytics can let you know how many of these are true. Think of these assumptions:

• Billpay is a sticky feature.
• Digital customers are cheaper than branch customers.
• Mobile members are made up of mostly millennials.
• Only 25 percent of your consumers use a branch.
• Customers that have more than two products are a better credit risk.

Each of these cases involves understanding a customer's why. Why is billpay a sticky feature? Is it because customers have a difficult time getting the data back out once they spend time setting up their payees? This reason is actually a risk because it is only a matter of time before someone (a browser plugin, maybe) devises a way to automatically populate payee information. Understanding the why would allow a financial institution to pivot to a more beneficial scenario for the customer, such as implementing a rewards system for using the service or allowing customers to export and import their payees as contacts similar to a mail client like Outlook or Gmail.

Digital “Why” 101

Here are some common things I look for when examining an organizations digital data:

• More than 15 percent of the population of digital users being subjected to multifactor authentication. If more than 15 percent of your customer base must answer additional questions to access digital services, one of two things is wrong.
  • Your multifactor authentication is not working correctly. Here is a good example: I often access sites that have a click box on that says, “Don't ask me for the code again for 30 days when I use this computer.” I click the box, and yet these sites still ask me for the code on the next time I log in. Since I have a background in coding online services, I know I am not clearing my cookies, changing my IP, or any other part of my digital footprint that would break the way a click box like that works. On many sites it simply doesn't work, and this is very frustrating for customers. It is important that the security for your site works as advertised, because if it doesn't, customers will lose trust in the services behind the login.
  • There is a possibility you are under attack. There is a chance that your organization is being slowly trolled by cybercriminals testing the limits of your security services. These are usually easy to spot in the data. However, without a trend to follow, its hard to spot these sorts of things in the sea of data that a typical digital service provides.
• Less than a third of digital base is using mobile. If your digital mobile users are under a third of your total digital membership (the digital membership consists of customers that have actively used your digital services in the last three months), then something is wrong with your mobile application. Here are some common things I find:
  • Customers cannot log in. Many organizations gave up access to their credentials to their home banking provider, and as a result had to create new credentials for the mobile application. This is unacceptable to customers in 2017, and many will not try very hard to get into your mobile application.
  • Your mobile application is missing a critical feature that customers expect. For example you cannot pay a mortgage, or see your commercial accounts. Credit card accounts are not visible or you cannot pay an auto loan on the mobile application. A lack of utility in the mobile application can spell certain death for the mobile platform in the organization as it is hard to recover from a perceived lack of services.
  • You have horrible reviews on iTunes and in Google play. I have seen this many times. Organizations don't monitor their reviews in these services and as a result they have a horrible reputation. Customers do look at the feedback from others, they decide if the application is worth loading based on the amount and quality of the feedback. Many organizations are reputationally damaged by one bad version of the mobile application in the application stores and never recover. The key to recovering is to work with your customers to have them review the applications. Honest reviews are important. I am not saying to “buy” or “encourage” good reviews, but try to get your customers to tell you how they feel about the application. When you discover there is bad feedback address it immediately.
• Online ACH debits are more than twice as large as the ACH credits. This is definitely a sign of fraud and should be looked into immediately. Any form of transfer that allows funds to be moved out of the accounts should be reviewed regularly.
  • On more than a few occasions I have discovered ACH fraud that goes back months. This is because the criminals have targeted the accounts of the elderly who perhaps are not watching their money as closely as others or lack the technical capability to set up alerts or check online. The fraudsters often socially engineer their way into these accounts, and as a result, they are able to sneak money out of accounts and into ACH mule accounts without garnering any attention from organizations security systems.
• Digital active users are less than a third of the total customers. This is a big red flag because it either means that age demographic of the customer is so high that digital services are not enticing to them or that the digital services are subpar and as a result aren't being used. Worse yet, the digital services are fine but not being marketed correctly.

So these are simple why tools and criteria that you can implement at your financial institution. One, track your direct deposit. If it stops, find out why. Two, track people's usages of your credit cards in major vendor sites like Netflix, Amazon, and airline sites. Watch for trends and watch for drop-offs. If it stops for some reason, contact them. Find out the why. Watch for drop-off trends and digital channels, particularly ones that you want your consumers to use, like home banking and mobile. If there are drop-offs, track down those members and ask them why, because at the end of the day when we learn the why, then we can deal with the how and the what.

Digital Marketing

The time is coming when financial institutions will no longer be able to rely on the marketing techniques they have used in the past. Most financial institutions are still stuck in the digital dark ages when it comes to marketing. They rely on print media, billboards, signage at their branches, and digital banners in their digital channels.

When the digital channels evolve to be more personalized platforms, customers are going to be able to control their advertising and most will likely opt out of marketing all together. Your digital transformation strategy should include radically rethinking your marketing strategies to take advantage of your new digital platforms.

The first marketing technique that needs to be mastered is inbound marketing. Inbound marketing is a marketing method that attracts customers to your products and services by providing rich content that pertains to your products and services. Financial institutions have been doing this for some time. For example, I can remember helping to set up seminars at Suncoast Credit Union on getting a mortgage or planning for retirement. Print media with articles relating to financial health were sent to the customers on a regular basis. The idea is to provide valuable content that doesn't overtly sell your products or services, but rather, educates a customer about the various aspects of a product service. The goal has always been to become the customer's trusted advisor for financial services, and as such, the first place they will turn to when they need financial help or new products.

In the digital age, this process has evolved from sending marketing to someone's house to emailing a customer targeted articles that are relevant to where they are in their financial life. The key to accomplishing this task is executing on the data analytics section of this book. Data analytics will reveal your customers' specific needs and allow you to engage them with topics that will cut through the inbox clutter and spam and grab their attention. The content for these articles can come in many forms. For instance, if you are rolling out a new product feature, it may be useful to create a video that demonstrates the feature. If you are seeking to grow your investment portfolio, your investment team might do a podcast or blog about investments. In Kirk Drake's book CU 2.0, he describes a marketing message as asking more questions than telling someone what to do. He states:

If the consumer bites and follows the topic, instead of finding a product description page with the features in bullet points, the consumer would find a well-written article on the value of a new free checking account without overtly selling the financial institution's product. The article would also be written in such a way that it would be picked up by search engines such as Google and Bing. This would also bring new traffic from internet searches, and the traffic would be people looking for information about free checking accounts. If your article is useful to them, they will be back for other financial information. In this way, you can attract new customers. You may choose to write more than one article on the topic with different perspectives or details. Each article can be used to collect data from the readers and determine which articles' message resonates with customers.

Today, your organization can hold webinars with hundreds of attendees. You can use webinar tools to provide free content and financial advice. The webinars can also be recorded and then reused as content on your YouTube or Vimeo channel. If a picture is worth a thousand words, a video must be worth millions. I have watched my millennial son and his friends look for instructional YouTube videos for everything from fixing their cars to how to finish a level in a video game. A quick search for the word “mortgage” in YouTube brings back 1,790,000 results. While there are many videos on how to get a mortgage, very few are professionally done by a financial institution. A complex process like a mortgage is exactly the kind of topic that a customer would like to see simplified.

Inbound marketing also includes using technology to track your progress across all of your channels. One of the most prolific tools in the marketing space is a product called HubSpot. In fact, HubSpot's CEO, Brian Halligan, coined the term inbound marketing to describe what his company does. Inbound marketing is about your ability to harness your data and use it to determine who your customers are individually and using the information to provide them valuable information.

Get outside of your four walls. As mentioned before, I foresee a day when your customers may not need to visit your home banking, website, or mobile banking platform to get their balances or history. When this happens, it will impact the banner-based digital marketing that FIs have depended on for so long. New marketing techniques will need to be employed that include email, social media, and targeted website marketing. Your organization will need to partner, hire, or learn the ability to market in the same ways the big players do like Google, Amazon, and Apple. I can almost hear the groaning from here about this. I realize that you believe that what the big players do is out of your reach, but I have learned from running my own business that these tools are well within reach of the average financial institution. In the past, the large search engines would give more weight to larger content providers; however, the new search engine algorithms give more weight to content that is regionally related, meaning that if your organization can write articles that include relevant financial points about the local community, then your articles can appear above the major banks' articles. If you need help to determine what to write about, I have a simple solution:

• Step 1: Find a millennial. They are easy to find—look at the local hipster cafe, coffee shop, or local restaurant (not a chain).
• Step 2: Ask him or her to tell you what he/she knows about buying a house.
• Chances are, this person will have more questions than answers.
• Step 3: Write down all of his/her questions (make sure you capture all of the details).
• Step 4: Have your staff, or hire someone to, write articles that answer these questions.
• Step 5: Pick another topic such as checking accounts, or credit cards, and go back to step 1.

In this way you will become an educator, and all people (even nonmillennials) value educators. If you can become an educator, then you become an advisor.

Now that you have some great content, it's time to get the word out. Tools like HubSpot or Marketo can help you create a plan to deliver these articles to the right people, at the right time and on the right channel. There is an art to when to send out a tweet, post, or email and it is not a one-size-fits-all approach. Once you get to this level, it's time to stop marketing to the composite member persona that you have been using for all of these years. The composite customer persona represents a homogenized profile of all your customers. This approach is no longer effective because the average person is smart enough to see through an email subject, tweet, or Facebook post that has been homogenized. It's the difference between sending out an email with the subject line: “Save money by transferring your balance to our lower Credit Card,” and sending out an email with the subject line that says, “Learn how you can save $1,398 a year by transferring the balance of your Citi credit card to our new low-rate credit card.” The specificity of the second subject line makes it far more likely to be opened by the intended recipient. The key here is that the subject line has to be true, and to do this you will need to leverage data about your customers. The good news is that we have the technology to individualize offers in the manner I have described above; it's a matter of working with your data governance committee to get access to the credit reports of your membership. Each customer must be individualized and the marketing must be tailored to the customer's specific needs.

Even if after you manage to customize your marketing and start sending customers specific content, you may find that they still are not opening your email. The new tools for email and other channels such as a social media allow us to see who is opening our mail or reading the articles. When you discover customers that are not responding to your marketing, it's time engage “Plan B,” as in A/B testing.

A/B testing or split testing is the process of creating two different versions of your content for marketing campaigns. The marketing audience is then split in two, and each group is sent a different version of the content. Each version of the content can be monitored to determine which one is more successful in terms of getting customers' attention. The better of the two is then used for a future marketing campaign. Many organizations will use A/B testing in succession to dial in their marketing message and make sure it is resonating with everyone. Others will use the process to overcome users who are not responding to the individualized content.

The content you are providing should have a place for people to comment on the articles. This is critical because it will allow your customers to provide feedback about your content, and as an added bonus, it can improve your search engine placement. However, you must also be ready to receive negative content in these forums. It is very possible that a frustrated customer may choose to use a comment section of a piece of content that might be related to the service or product that they had a bad experience with to voice their frustrations with the financial institution. The first instinct of most organizations is to delete content such as this. I would argue that, unless the content is vulgar or inappropriate in some way, it is a bad idea to delete this content. It would be far better to address the problem right there in the public arena. If customers believe the forum is being moderated or censored, they will not trust the legitimate recommendations and praise that are also often offered in the comments. The same is true of social media feeds, iTunes application reviews, and other public digital forums where customers can publicly display their discontent with your organization. What's better than a happy customer? An unhappy customer that has been converted to a happy customer, right in front of your very eyes.

Another key to success in digital marketing is overcoming some of the cultural issues that have kept your organization from doing digital marketing in the past. In the new world you must be ready to market in what some FIs and their boards would consider questionable places such as Reddit, Twitch.tv, Facebook, Google, and community blogs. This allows your organization to be seen in common places that customers have come to expect to see advertising. It's no different than having a billboard along a major highway.

Another key component is to make sure the marketing messages are consistent across every channel. Wherever the customer chooses to engage the organization, he or she should see or hear the same message. Also if the customer opts out of an offer in one channel, this should carry over to all channels.

So what happens after customers read an article or several articles? Then it is time to convert them from students to business prospects. This can be done in many different ways—for instance, you can engage the customer in a chat right on the website screen, you can also display a banner for your product on the content page, and finally, you can do something insanely crazy. You could (and this seems weird) call the member by cell, Skype, or regular phone line. That's right, you could engage the outbound mode of your call center, create a queue of opportunities, and when the call center representatives have free time, they can use the time to make outbound scripted sales calls to prospects. These conversations are very different than a cold sales call. This is because you know, based on the customers' online behaviors, that they have an interest in the product or service you are trying to offer them. The conversation is usually far easier than a normal sales call.

Another evolving digital marketing method is to write an e-book about a subject. The e-book is usually longer than an article and is done in a format that provides quick value to the reader. The e-book offer is presented in trade for the user's email address or other contact information. Once the users provide their email address, the system can start to target market them with products that will be of interest to them.

Another note: No one wants your calendars, foam fingers, and refrigerator magnets. In fact, most stainless-steel refrigerator front doors are not magnetic anymore. Here are some better items to put your logo on and give to your customers: USB sticks (at least 8 GB), USB chargers (these are gold), USB hubs, screen cleaners, and smartphone credit card holders. If your marketing is going to be digital, your marketing materials should be digital as well.

Developing a digital marketing strategy will be critical to your digital transformation. Using digital marketing to educate your customers and your prospects on your digital services is a perfect match.