Marina Gorbis's Blog, page 1590

Seven Questions to Ask Your Data Geeks

Using data to manage is nothing new. But using big data to manage IS new, offering unprecedented challenges, opportunity, and risk. Senior executives need to learn and learn quickly. They must be prepared to ask penetrating questions when their data scientists bring them a new idea. Those who ask the following questions will be better prepared to both exploit the valuable insights and avoid the disasters that can arise from big — but bad — data.

When your data scientists bring you an idea, ask them the following:

1. What problem are you trying to solve? It is far too easy for data scientists (and others for that matter) to go on extended "fishing expeditions," seeking "interesting insights" that aren't tethered to the business. While a certain amount of exploration is healthy, most innovation is of the small-scale, one improvement at a time, variety — even with data. Encourage your data scientists to focus initially on known issues and opportunities, as well as more tangible insights. As confidence grows, your data scientists should be less restrained. And you should develop a keen eye for the difference between "exploring a difficult path" and "wallowing around."

2. Do you have a deep understanding of what the data really mean? Too often people gather data without complete understanding of the wider context in which the data were created, and misunderstandings find ways to hide themselves until it is too late. All data, even well-known quantities like "force" are subtle and nuanced. NASA (which truly has "Rocket Scientists") crashed a Mars lander because one team used the English measurement "foot pounds " and another used the Metric measurement "Newtons." The potential for such problems only grows the less familiar the data — especially through social media, automatic measurement devices, etc. — and the more intermediaries that touch the data.

3. Should we trust the data? Untrustworthy, inaccurate data is all too common. Just as a car can be no better than its parts, so too analytics can be no better than the data. Some data is inherently inaccurate (GDP forecasts); other data becomes inaccurate through processing errors. All too often, data collection is just not up to snuff. For example, far too many credit reports contain inaccuracies. Unless there is a solid quality program in place, expect the data to be poor!

4. Are there "big factors," preconceived notions, hidden assumptions, or conflicting data that could compromise your analyses? There is a lot going on here. First, it's natural to expect a return from our investment in data and analytics, but there's a sneaky side effect. People will "find" what they think you want. Saying upfront that you expect a 10% uptick in revenue can cause people to find a short-term 10% growth that's not there for the long-term, to be so busy looking for the 10% that they'll miss a potential 100% gain, or miss negative correlations entirely.

Second, advanced data analytics involves considerable judgment. Data scientists may have included some data sets and excluded others, from their analyses. You need to make sure they've not done so in unfair ways. The clarity and completeness of the answer correlates with the weight you should give to their conclusions.

Third, analytics is essentially about developing a deeper understanding of how the world works. False assumptions are crippling. For example, the assumption that home prices were uncorrelated across markets was a major contributor to the financial crisis.

5. Will your conclusions stand up to the scrutiny of our markets, moderately changing conditions, and a "worst-case scenario?" Don't confuse data science with classical physics. Verifying your conclusions is not as simple as repeatedly dropping uneven weights from a tower. You want your data scientists to be skeptical; to challenge each other; to test, test, and test again ; and to quantify, or at least fully describe, the uncertainty in their conclusions under normal situations and to make clear when uncertainty explodes! This is critical because your data factory will almost certainly operate outside the purview of the data scientist.

6. Who will be impacted and how? The increasing volume of data available about people makes privacy a very touchy subject — both inside and outside of your organization. The line between helpful and creepy is gray and very thin. Data scientists can produce startling insights, but they are not fully-equipped to think through the implications. It's important to ask this question not only to the data scientists you're working with, but your executives as well.

Be careful that you are not the affected party by making your operations highly sensitive to uncontrolled data. You need look no further than the NYSE 'flash crash' or the drop caused by hacked Twitter messages. Imagine what that type of error could do to your business.

7. What can I do to help? Quite obviously, there is no need to ask this question if the answers to the first six questions don't satisfy. Bear in mind here that any important discovery will have implications across the organization. We're particularly concerned about change management. All change is difficult and resistance to counterintuitive results will prove far too strong for most data scientists.

Except for the first and last questions, we are well aware that simply asking these questions will stretch most executives' thinking. But take heart — they will stretch all but the most experienced data scientists' thinking also! Demanding cogent answers will help executives and data scientists deliver real benefits from Big Data.

What "The Internship" Gets Right (and Wrong) about Mid-Career Internships

Let's establish up front that Billy and Nick, the two 40ish, out-of-work salesmen played by Vince Vaughn and Owen Wilson in the new movie The Internship, are unlikely to land a coveted summer spot at Google. But in Hollywood, the place where Cameron Diaz can be an orthopedic surgeon (in There's Something About Mary), anything is possible. So let's suspend disbelief (and put aside the question of whether or not it's funny) and consider what the movie gets right about midcareer internships.

College internships are well established, but the strategy of opening up internship programs to non-traditional candidates is gaining currency. It provides a cost effective and low-risk way to engage with mid-career professionals seeking to return to work. (I wrote about this trend in the November 2012 issue of HBR.) Sometimes, as in The Internship, the non-traditional candidate turns out to be the best choice.



No matter the age of the intern, internship programs give both employer and intern an opportunity to try each other out before committing. In the movie, only the best team will be offered full-time employment, giving Google a unique chance to see its future employees in action.



Despite Billy's Flashdance references, the movie underscores that the "older" generation can collaborate constructively with millennials. In fact, as the movie progresses, their differing perspectives and skills help them form an effective team, with each age group relying on the other to provide what it lacks. My own experience bears this out. When I returned to work at an investment firm at age 42, after 11 years out of the full-time workforce, I was the grandmother of the operation. Humor and a willingness to admit what I didn't understand enabled me to establish productive working relationships with my younger colleagues. They in turn would seek me out for advice that only someone "more seasoned" could offer.



Mid-career professionals returning to the workforce or transitioning, whether as interns or not, will have a better chance to succeed if they are "coachable." In the movie, Nick is initially more coachable than Billy, and he excels earlier because of it.



Mid-career professionals should not make assumptions about who may be "on their side." (Warning: Spoiler Alert.) In The Internship, the manager overseeing the internship program appears hostile to Billy and Nick until the very end, when he discloses that he cast the deciding vote to admit them. Lacking the fancy education of so many Googlers, he had prevailed through hard work and saw the same tenacity in Nick and Billy. Career transitioners and returning professionals may be surprised by who will relate to their backgrounds. The unapproachable 64-year-old senior partner may have a daughter returning to the workforce after a career break, or the 45-year-old middle manager's husband may be transitioning to a new career.



The movie caricatures middle-aged technophobes, and Billy's repeated references to being "on the line" (rather than online) are hilarious. But in fact, technological ignorance among mid-career professionals is rare today, and most understand that employers expect them to come to the table with basic technology skills in Excel, PowerPoint and Word, and some familiarity with social media. Serious candidates for employment make it a priority to update skills before applying for jobs. Yet the movie also shows Billy and Nick moving quickly up the technology learning curve, demonstrating that mid-career professionals can achieve a level of technological proficiency necessary to survive even at Google (or at least the Hollywood version).



One skill that is sometimes undervalued among younger professionals is sales. In The Internship, Billy's exceptional sales prowess saves the day for his team.

When I left the theater after seeing The Internship my initial reaction was how unrealistic it was. The idea that two watch salesmen who had recently enrolled at the University of Phoenix would even make it as far as the Skype interview for the Google internship is pretty far-fetched. And how convenient that Billy and Nick are both single with no other commitments and could relocate at a moment's notice. Yet, setting all that aside, the movie offers a number of lessons.

Will it inspire companies to hire more mid-career transitioners or returnees? Hard to say. But just maybe some employers will get the message. Maybe even Google.

Five Reasons Social Media Won't Consolidate

"I wish there was just this one-stop tool for all things social."

"It would be great if we had one tool that would do everything."

"There is no one good tool."

Talk to the head of social media at any Fortune 500 company, and you're likely to hear comments like these. It's not that social media professionals are lacking software options: on the contrary, "there are a hundred different solutions for aggregating discussion, tweets, et cetera," Miguel Moreno Toscano told me.

Toscano is marketing communications director at Coca-Cola, and he's one of half a dozen enterprise social media leaders I've spoken with in recent months, beginning with several at the annual SXSW conference. In those conversations, one concern was repeated by all: the embarrassment of riches (in terms of social media software options) has given marketing, communications and customer relations pros a major integration headache.

"We are almost having tool fatigue," said Jay Bartlett, vice president of global social marketing for Xerox. "These tools are developed in a very fragmented way; I wish there were some integration. You can see how they could work together, but they are all very niche-y."

"I'm using one tool for monitoring and listening, another for monitoring and reporting, another for engagement," said Carla Saavedra Kochalski, manager of social media and digital for Samsung Mobile USA. "Then I've got one for case management, one for content management and another one for reporting. I wish those companies would focus on making a more cohesive ecosystem."

Nor is the fragmentation problem limited to the tools: the ecosystem of social media platforms is fragmented, too, and growing more so.

"The critical mass of consumers is on Facebook, Twitter and Pinterest, but there is a whole other range of networks with different types of content, and it a huge challenge for brands to engage there, too." Coke's Moreno said. "The long tail of networks is starting to be very important."

"The question is, how do you let customers get the content they want from you in the place where they want it?" asked Sam Weston, vice president of communications for Huge, a digital agency. "Companies have embraced Facebook and Twitter, but now Facebook and Twitter own those relationships instead of the business itself. Now, if you want to get your post to the people who want the brand then you have to pay premiums to do it. Ideally you are able to get this content to individuals, but it's about how efficiently you can get your content to the people who follow your brand."

If companies are experiencing so much pain from the proliferation of social tools and platforms, why is there no one-stop option that can deliver the full range of functionality required? After all, we're now a full decade into the social media revolution: there's an established market for social solutions, particularly those that solve pain points for corporate CMOs.

Andy Levey, senior manager of new media and analytics at Cirque de Soleil, sees integration just over the horizon. "We're starting to see those all-in-one tools that are more encompassing," he told me. "I know that is the path we are going down."

Big players like Salesforce and Oracle are indeed beginning to roll out software tools that offer something closer to one-stop shopping. For companies that have gradually consolidated their purchases of other kinds of enterprise software needs like enterprise resource planning (ERP) and database systems it may feel intuitive to expect a similar consolidation in social media purchasing.

But there are good reasons to doubt that the proliferation of social media software and platforms will give way to integration. Among them:

Low barriers to entry: If I want to sell you my new financial software platform, I've got to dummy up some data for that demo, with enough quantity to convince you that I can support an enterprise of your magnitude. But with a social platform, I can aggregate much of your social data as easily as you can, simply by pulling in your social mentions across platforms. That makes it easy for me to show you exactly how my platform will work in tracking your social universe...and significantly reduces the competitive barriers to later entry.

Ease of migration: Barrier to migration is similarly low. Yes, there are always costs associated with changing platforms (especially in terms of the learning curve of your employees). But unlike ERP or database migration, changing social platforms often requires no data migration, but simply the effort to move a bunch of keyword searches and social network authentications.

Fragmented needs: Social media tools are fragmented in part because businesses' social media needs are fragmented. You've got marketing people trying to push out campaigns, customer relations people trying to say on top of customer issues, business analysts trying to gather appropriate metrics, human resources teams trying to recruit off of social...and that's just the tip of the social iceberg. Any integrated tool is likely to represent an imperfect compromise between these units' very different needs.

Consumer demand: You don't need your customers' buy-in to push different divisions of your company onto the same ERP system. But your customers, not your staff, decide which social networks they'll use...and you'll have to go there to meet them. So far, consumers (goaded by the tech and business press) seem prepared to embrace at least one new social platform a year (this year, it's Vine...last year, it was Pinterest). Enterprise social media managers must prepare to do the same.

"Startup" mentality: The reliable emergence of a new platform or two each year means that working in social media will continue to attract people who have not just a tolerance for novelty, but a hunger for it. These are the kind of people who will happily sign up for a dozen new social web platforms or services each month, just to see what they're about (I'll raise my hand if you will). Toscano called this a "startup mentality" — a willingness to "use a combination of tools, and use new tools as they pop up."

So for that desired "integrated solution" to take hold, it has to undertake continuous and aggressive innovation so that it can support each major new social network and compete with newly emergent social tools. Tall task, to say the least. While there will surely be enterprise clients who value the efficiencies of integration over the features offered by having just the right tool for each job, the companies and professionals who have clear and high expectations for their social media performance will continue to explore and adopt the most effective platforms and services. Since these are likely to be the companies who will be most innovative and effective in their use of social media, expect their example — including their fragmented toolkit — to be the imperfect best practice in social media.

The Imperatives of an Organization Built for Speed

In Greek mythology, Hydra, an ancient water-serpent had many heads. If one head was cut off, two rapidly grew in its place before another head could be cut off — an energy-sapping disappointment for any opponent trying to overcome it. Regenerative speed made the Hydra formidable. Even Hercules, the legendary Greco-Roman hero, needed his nephew's assistance to win. To sustain a competitive edge, your company's new business development engines must similarly fire on all cylinders at supersonic speed.

As a CEO or a leader of a business, how do you build this competency? Measure, motivate and model.

Measure: Measure your company's "heart rate" and optimize for speed

Every team, business unit and/or company as a whole, has an underlying execution rhythm. At the most basic level this may be an individual's task completion rate (TCR). Setting a TCR of 2 weeks would mean any task you give to another or take from another needs to be done in 2 weeks. Imagine every employee, putting a red sticky on a company-wide virtual whiteboard, when an assigned task isn't completed in the allotted two weeks. With an explosion of stickies, you know that either the task allocator (a project manager) is not breaking down the task into a meaningful two-week chunk, or the doer (a low rung employee or a high rung decision maker etc) is not able to complete the task, or perhaps there are other dependencies, etc. While this is a crude example, it illustrates the importance of tracking, doer-allocator transparency and an implicit service level agreement across team members, which encourages "good enough" instead of perfect, thus optimizing for speed.

Just as agile product development methods, such as Scrum, use process and tracking tools to set and track execution rhythm, so must the organization's leader measure and monitor to ensure useful output. After all, you can't improve what you can't measure.

Motivate: Instill the sense of urgency

The best way is to expose employees to "the jungle." Too often, front-line sales people, but not necessarily the engineer or financial analyst deep in the organization, can "feel" the competition. Simple steps such as sending them to a conference dominated by a competitor, or having them listen to a tough sales/customer service call can get their emotional investment. Some may be motivated by threats, others by solutions and the impact they can have on the world. In case of the latter, define competition as the worsening of a current problem statement. Regardless, one needs to "feel the jungle" to adopt a sense of urgency.

Model: Lead the way

You must role model to lead the way. First, don't be the bottleneck. Empower and delegate decisions so people aren't waiting for your decisions or resource allocation requests, any longer than the desired TCR. When the stakes are high and you need to decide, lead, even when in doubt. Innovation by nature is uncertain and your job is to realize what is knowable, what is not, and how to move forward to eliminate critical unknowns. So, stop looking for data that doesn't add to your decision and stop using the lack of data to procrastinate on hard decisions. Speed must be a factor in your consideration.

Finally, as this I Love Lucy video illustrates very aptly, you can't speed up the belt forever. When moving faster would result in over-utilization or amplifying skill gaps, find new ways. Can you buy instead of build? Form partnerships and alliances for mutual benefit? Fail-fast to enter a white space with a higher probability of success?

Ultimately, every organization — whether a nimble start-up or a large, established firm — needs to find ways to speed up or be left behind. These three simple rules can help you move faster.

Quality vs Frequency: What's Your Mobile Strategy?

Your customers are in the midst of a mind shift. First they get a smartphone. Then they learn that they can ask for any information and get an instant result. What's the weather forecast? Is this dishwasher highly rated, and is it available more cheaply elsewhere? What's the song playing on the radio in the deli? Mobile answers all these requests and reinforces, in a Pavolovian way, the concept that we call the mobile mind shift: the expectation that any desired information or service is available, on any appropriate device, in context, at a person's moment of need.

Based on an analysis of data from over 4,000 online American consumers, we estimate that 22% of consumers have made this shift already. They are among your best customers: typically younger and more affluent than other Americans. And they are demanding mobile utility. Your job is reduce the distance between what they want and what they get. Meet their demands and they'll be loyal to you. Fail, and they'll switch to a competitor, or a digital disruptor like Uber or Mint or Google News.

When attempting to deliver on mobile, we've recognized that company strategies vary based on where they fall on two dimensions: quality of experiences and frequency of experiences.

The vertical axis reflects the quality of experiences, as measured by tools like Net Promoter or the Customer Experience Index. The horizontal axis reflects the frequency of experiences — companies whose customers interact with them weekly or more frequently are on the right, those with less frequent interactions on the left. What strategy should you use for mobile? That depends on what quadrant you fall into.

Companies in the upper right deliver frequent, high-quality experiences. Their customers will expect those frequent experiences to include mobile elements. A good example: Tesco stores in Korea. They put graphics up on the walls of subway tunnels that look similar to store displays. Customers can use their smartphones to photograph the items they want, and have them delivered as soon as their subway gets them home. Result — demanding customers get what they want, and remain loyal to Tesco.

Companies in the upper left deliver great experiences, but infrequently — they need to use mobile to increase the frequency. Think of brands like Nike, with which a customer might interact only two or three times a year, when they buy running shoes. Nike realized that on mobile, people were interacting with mobile brands more often, so they created a whole line of free mobile apps, like the Nike Plus Running app that runners interact with every day. Now the Nike customer thinks of Nike in a favorable light far more often.

Those in the lower right deliver frequent experiences, but often disappoint customers. Most telecom operators and banks fall in this quadrant. Fixing customer experience problems is slow and expensive. In the meantime, companies in this quadrant can add good mobile experiences to help counteract the poor ones. Take Verizon, whose tablet app delivers access to 75 channels of TV when viewers are connected to their home networks. This gives customers reasons to think favorably of Verizon many times a day. This app was popular enough to get a three-and-a-half star rating.

If you're in the lower left, mobile probably won't solve your customer problems. But you can use mobile utility to remind customers that you're on their side, perhaps by partnering with a digital disruptor in your space. Many health insurers are in this spot, since they typically rate at the bottom of scores like the Forrester Customer Experience Index. One of those insurers, Cigna, helped its subscribers by providing free access to an app that helps with meditation, an app that would otherwise cost subscribers money.

As more and more customers make the mobile mind shift, companies need to make mobile strategy decisions that match their customers' experience. Failing to deliver mobile utility could drive these customers away. But choosing the wrong strategy could be costly. Choose based on the quadrant your company lands in, and you're more likely to create mobile experiences that match your customers' expectations.

Innovations in Digital and Mobile Marketing
An HBR Insight Center





Being Digital Demands You Be More Human
CMO's: Build Digital Relationships or Die
Measurement in a Constantly Connected World
Brick and Mortars (Still) Can't Beat the Web on Price

CEOs Are Different from Us (and from CFOs)

Just 9.8% of chief executives can be categorized as risk-averse, compared with 64% of the (similarly aged) general population, according to a survey of about 1,000 top leaders in the U.S. by John R. Graham, Campbell R. Harvey, and Manju Puri of Duke University. Moreover, 80% of CEOs are very optimistic, well above the mean. CEOs are also much more optimistic than CFOs; only 65% of CFOs can be classified as very optimistic.

Beware Trading Privacy for Convenience

We pay for things with the swipe of a finger. We ask Siri how to get to a restaurant. Our friends can track exactly when we'll show up. We can monitor our heart rates and calories burned — and compare our results with friends and strangers. We're in the early days of a digital, mobile transformation. The benefits can seem limitless. And as a society, we are already becoming accustomed to the convenience, the connectivity, and the new insights that surface.

This democratization of technology impacts all races, incomes, cultures, and geographies. For example, the Supplemental Nutrition Assistance Program in the U.S. now distributes all its benefits — what used to be called food stamps — electronically via Electronic Benefit Transfer debit cards. These can be used like any other debit card for food purchases at grocery stores, local farmers markets, and some restaurants. Benefit recipients do not need to have a bank or deposit account. The cost savings are huge, with the federal government spending $200 million less on paperwork and administration. It's also more convenient for beneficiaries than carrying around a stack of actual food stamps.

We also see savings and other benefits as tollbooths go away and electronic payments take charge. Just recently, San Francisco's Golden Gate Bridge got rid of all its toll collectors. In Texas, the six toll roads around Austin have also stopped taking cash. Drivers using the TxTag can whiz through the 52 toll points and 45 ramp toll plazas, and the state will save an estimated $8.5 million each year.

The slow trade of privacy for convenience

The benefits of this digitization for consumers, companies, and governments are obvious and exciting: Instant gratification, seamless experiences, cost savings, greater insight. However, the digital exhaust we leave behind while using these convenient and often free services reveals much about us. Social networks, mobile apps, self-quantification, and sensor networks are creating a massive and very public mesh network of data to be mined.

Repeated check-ins can reveal your home location. Personal information shared in public social networks is being used by hackers to crack passwords. Marketers now have rich profiles that can predict when you are more likely to order sushi during the week or whether you prefer wine over beer. The electronic toll way system and the cellular towers know where you are. Your smartphone's location-based services mean you can never disappear.

Even worse, a global push to digital transactions means that we no longer have anonymity when we make a purchase. The rapid adoption of electronic payments may render cash obsolete in our lifetimes. The new payment methods might be more convenient, yet every transaction will be tied back to us.

Our longing for convenience means we've created a matrix that can and will be used against us. Most of us just don't know it yet — although last week's revelations about the large-scale surveillance programs the U.S. National Security Agency has been conducting in cooperation with telephone and Internet companies has raised the awareness level a bit. This digital trail isn't protected by Fourth Amendment protections against unlawful search and seizure — law enforcement in the U.S. doesn't need a warrant for many digital searches; a simple subpoena will do. This applies to cell-phone location data, information you store in the cloud, you name it.

The battle for digital identity

Meanwhile, social networking sites, financial service companies, smartphone manufacturers, governments, insurance companies, and telecom carriers all want customers to trust them with their identities. With that trust, they hope to secure revenue streams from identity to orchestrate payments, manage access, validate reputation, and ensure security.

This is the paradox. The companies contending to win our trust to manage our digital identities all seem to have complementary (or competing) business models that breach that trust by selling our data.

Customers must take back control of their data

Something has to give. My belief is that we won't be able to build sustainable digital business models until we agree on some limits to how customer data can be used. A compact must be reached on the balance between privacy and convenience.

What exactly that compact will look like is anybody's guess. But here are seven basic protections that consumers ought to demand:

Make "opt-in" the default. Basic profile information should require an affirmative permission to share information, use for offer creation, or even suggest next best action. Opt-ins should also apply to user-generated information such as messages, photos, audio, and video.
Be transparent in how personal information is used. Organizations should detail what information will be shared. Users should know if their information will be sold and if so to whom.
Give advance notice of privacy changes. Organizations should provide adequate warning when new features impact a user's privacy preferences.
Require "opt-in" for privacy changes. The default option should be to keep privacy preferences the same. The recent Electronic Privacy Information Center FTC complaint and settlement with Facebook reinforces this principal.
Prevent access to user's data upon account deletion. Information about a user should be locked down when an account is deleted. It should not be used in aggregate statistics or data.
Allow users to export their data. Customers should own their data and be able to take it with them as needed. Doc Searls and the Project VRM community have been advocating Personal Data Stores for quite some time. This may be the necessary requirement for social business to make it to the next level.
Give users a "hard delete" option. Users should be able to request and receive a permanent deletion of their data, with all information removed from all files.

Many would like us to believe that privacy is dead. Yet, privacy is a societal choice — it is only dead if we allow it to be. We should insist that businesses and government agencies offer choices to engage in both offline and online models. This may result in a rebalance of how much privacy we are willing to trade for convenience and lower cost. For example, we may decide that some inconveniences — such as a mandatory option to be able to conduct business in cash — are worth it. We should not resign ourselves to thinking that we cannot defend our individual freedoms. We must have an open dialog about where we will draw the line. Let's get started before it's too late.

Data Under Siege
An HBR Insight Center





Welcome to the "Data Under Siege" Insight Center
Four Things the Private Sector Must Demand on Cyber Security
Does Your CEO Really Get Data Security?
The Companies and Countries Losing Their Data

When Digital Marketing Gets Too Creepy

The digital marketer who effectively runs Qantas Airlines' highly regarded — and very successful — loyalty program has an unusual iPad problem. Flight attendants on Australia's flagship carrier can now get up-to-the-minute data on the airline's most elite and valued frequent flyers displayed on their onboard tablets. The information is useful, helpful and the app was a digital innovation actually sought by Qantas staff.

The unhappy catch? Too many flight attendants sounded like they were reading from a script when using this information with these valued customers. They couldn't smoothly incorporate the customized data to authentically connect with their frequent flyers. Instead of making their best customers feel special, the data-driven app too often creeped them out.

"We learned that staff needs training to take full advantage of the data we can provide them," said Rob Colwell, Qantas Loyalty's Chief Commercial Officer at a recent analytics workshop hosted by Ernst & Young in Sydney. Colwell observed that the rate of analytics innovation consistently outpaces interpersonal skills. Creating better customer experiences from hard data required a softer touch.

I've personally and professionally experienced similar pathologies experiencing "customer support" from credit card companies, financial services firms, airlines and telecoms operators. One can practically hear their all-too-human staff struggle to utilize the bespoke information they're (presumably) reading off their screens to mollify, humor and/or resolve my issues in ways reflecting how much they "know" what a great customer I am. The variances in customer experience quality are enormous. Like those Qantas frequent flyers, I occasionally feel more creepy than comfortable with these "interactions." Not good for those net promoter scores.

So should technology-empowered "customer touch" personnel attend improvisational acting workshops? Or teleprompter training? After all, as the joke goes, the key to success in acting is "sincerity" because if you can fake sincerity, you've got it made.

The better question, however, doesn't revolve around performance art skills; it asks whether organizations really want to build data-based relationships around their people. After all, Amazon doesn't have these problems. Neither does Google nor Twitter. (Intriguingly, Apple — even with its high-touch Genius Bars — appears to know, and care, far more about their machine(s) than their owners.) For many if not most organizations, an increasingly digitized self-service/DIY sensibility seems to make the most economic, organizational and cultural sense.

But if you're a Qantas or Singapore Air — or a luxury brand like a Zegna or Lord &Taylor — technology forces you to revisit and rewrite the relationships narratives you want your customers to experience. Should ever-more data and individualized insight be an essential ingredient or simply alluring spice in keeping the relationship going? Should ever-greater service intimacy become an integral expectation of top-tier customers and clients? Does your organization have serious situational awareness around when useful knowledge acquires the stench of "creepiness?"

Sitting on that Sydney panel with Qantas' Colwell made it crystal clear that even the most sophisticated and successful data-driven customer-care organizations have only begun grappling with these kinds of questions. (The Qantas executive recited, with nary a hint of schadenfreude, to help identify elite flyers waiting in airport queues).

Before the decade's end, even minimum wage customer service personnel will have real-time access to remarkable amounts of personal data of customers walking into Starbucks, McDonalds, Walmarts and/or Walgreens. Should customer experience be better defined by employees who enjoy greater familiarity or a studied distance? Who owns the answer to that vital human capital and customer care concern?

The more customer data an enterprise has, the more that kind of accountability matters.

Innovations in Digital and Mobile Marketing
An HBR Insight Center





Being Digital Demands You Be More Human
CMO's: Build Digital Relationships or Die
Measurement in a Constantly Connected World
Brick and Mortars (Still) Can't Beat the Web on Price

So You Gave Your Customers' Data to the Government...

Rang, Recorded, Delivered, I'm Yours


When the news broke about the U.S. government's PRISM program, which collects data from phone companies like Verizon and online sites like Facebook, the reactions and reporting mostly focused on the NSA's broad ability to know exactly what many Americans (as well as foreigners) are doing, one keystroke or call at a time. For telecom companies like Verizon, Sprint, and AT&T, complying with government requests for information isn't uncommon: In 2011, Verizon received about 260,000 requests for customer data; AT&T got 49,700 (965 of which it rejected). Essentially, "complying with requests for such data has been part of doing business for much of the past decade," though Verizon in particular notes that it takes pains to ensure customers' privacy. And Denny Strigl, Verizon's former president, says that the company is "between a rock and a hard place here. If people are going to make an issue of this, the issue is with the government — not with the corporate citizen who complies with the law." Tech companies, meanwhile, are denying any knowledge of PRISM. While public anger does seem to be aimed mainly at the government right now, that can change on a dime, and companies may well suffer from customers’ wrath. On a lighter note, something else any good businessperson should know: Never, ever use the type of graphics the U.S. government did in a slideshow about the program.

"This Isn't Harvard Grad Stuff"

Costco CEO Craig Jelinek Leads the Cheapest, Happiest Company in the World Bloomberg Businessweek

Circuit City fell first. Best Buy and JCPenney are struggling. Then there's Costco, the membership warehouse that's doing incredibly well and paying its employees a living wage (not to mention providing health care and retirement benefits). In this in-depth profile featuring the company's new CEO, Craig Jelinek, Brad Stone explores the history of Costco's way of doing business and the challenges the company faces going forward (let's hope Amazon can't figure out how to deliver groceries in bulk over the internet). Two things stand out about Costco’s corporate culture: First, people stay at for decades, and the company doesn't hire business school grads into its upper echelons. This means senior executives frequently talk about succession planning, something that's often lacking at other companies. The other? There's no PR department, an omission that echoes the company's insistence on simplicity. "This isn't Harvard grad stuff," Jelinek explains. "We sell quality stuff at the best possible price. If you treat consumers with respect and treat employees with respect, good things are going to happen to you." Touché? Or naïve?

Who's That Crossing My Bridge?

Trying to Stem the Rising Tide of Patent Lawsuits Planet Money

So who are these patent trolls anyway? The U.S. government says it's cracking down on purveyors of patent-infringement lawsuits that are only about squeezing money out of companies. In a close look at the problem, NPR follows the case of a technology innovator who took out patents on his and others’ ideas in the 1990s — ideas that never turned into products or companies but that have become a source of revenue through patent suits. Lots of companies that are targeted by this kind of lawsuit end up settling, because defending a patent case requires explaining technology to a jury. Just try that sometime. The patent-suit business can be very lucrative, at the targets' expense — one company was so severely hurt by a settlement that it had to lay off employees. —Andy O'Connell

Good Business Sense

Three Reasons Why I'm Voting for Gay Marriage Financial Times

Former BP chief executive John Browne resigned from his position in 2007 after he lied to a judge about a former boyfriend. Now he would like you to know why he's supporting an upcoming gay marriage vote in the UK. One: History shows that marriage is an evolving institution, so it’s not true that gay marriage represents a disruption of something eternal. Two: It makes business sense: "People are happier and more productive and make more money for their company when they feel they are included and can be themselves." Three: If Browne had seen other gay men in stable, legally recognized relationships when he was growing up, it would have been easier for him to come out. According to Quartz, his piece is "the first such forceful defense of gay rights by a major figure in the global oil industry, in which a homophobic environment persists."

Grin and Actually Mean It

Maximizing Happy Kellogg Insight

"How happy were you today?" This is one of the questions Kellogg School of Management professor Kelly Goldsmith asked in a series of studies that measured whether happiness is actually attainable – and, she found, this might very well be the wrong question to pose. While people who monitored themselves using this question as a baseline reported small increases in happiness, it was another question – "Did you do your best to be happy today?" — that helped people make real progress in feeling happier. Why? The latter question reminds us that we want to be happy, and because it implies we have the power to increase the feeling, it can lead to actual changes in behavior.

BONUS BITS:

Once You Start Me Up I'll Sometimes Stop

The Selling of Instagram (Vanity Fair)
The Misevaluation of Farmville (New Yorker)
The Echo Chamber of Silicon Valley (New York Times)

Defeat Hackers with Biomimicry

From denial of service attacks to server crashes to day-long disruptions of Google Drive, almost all organizations are familiar with threats to their information security. Given that digital information is more central than ever, it's worrisome that the history of data security is littered with failure. Organizations seeking to be better prepared for and more resilient in response to information threats may want to draw on a far larger and older source of lessons on information security — the 3.5 billion year history of life. Tapping into biology's security database — which was developed by millions of species in response to extremely complex natural security problems — gives us first a wakeup call, then some practical guidance on how to keep our information secure.

The wakeup call concerns our assumptions about the borders, barriers, and firewalls we construct in a valiant attempt to protect our data. In nature, barriers — between organic and inorganic chemicals, between land and sea, between species, between everything — have been built, tested, overcome, rebuilt, and overcome again with almost endless repetition. Barriers — be they cell walls, border walls, or firewalls — are at best a temporary imposition to an invader. In the same way that tightly controlled unicellular life eventually evolved into more open and distributed multicellular life, the rapid evolution of cyber threats has outpaced the evolution of defensive barriers.

The lesson is simply that modern organizations should work under the basic assumption that almost anything electronic is now open source. My colleagues in climate science learned this the hard way when politically motivated hackers stole and released thousands of emails sent among scientists. Not only did sensitive data and preliminary analyses methods leak out, but the petty interpersonal spats and behind-the-back sniping that probably appear in all email chains were revealed in all their unappealing light.

So how do we operate in an effectively open-source world without barriers? Here biology offers some get-off-your-ergonomic-chair-and-do-something advice.

The biological world is also open source in the sense that threats are always present, largely unpredictable, and always changing. Because of this, defensive measures that are perfectly designed for a particular threat leave you vulnerable to other ones. Imagine if our immune system were designed to deal only with a single strain of flu. In fact, our immune system works because it looks for the full spectrum of invaders — low-level viral infections, bacterial parasites, or virulent strains of a pandemic disease. Too often, we create security measures — such as the Department of Homeland Security's BioWatch program — that spend too many resources to deal specifically with a very narrow range of threats on the risk spectrum.

Advocates of full-spectrum approaches for biological and chemical weapons argue that weaponized agents are really a very small part of the risk and that we are better off developing strategies — like better public-health-response systems — that can deal with everything from natural mutations of viruses to lab accidents to acts of terrorism. Likewise, cyber crime is likely a small part of your digital-security risk spectrum.

A full-spectrum approach favors generalized health over specialized defenses, and redundancy over efficiency. Organisms in nature, despite being constrained by resources, have evolved multiply redundant layers of security. DNA has multiple ways to code for the same proteins so that viral parasites can't easily hack it and disrupt its structure. Multiple data-backup systems are a simple method that most sensible organizations employ, but you can get more clever than that. For example, redundancy in nature sometimes takes the form of leaving certain parts unsecure to ensure that essential parts can survive attack. Lizards easily shed their tails to predators to allow the rest of the body (with the critical reproductive machinery) to escape. There may be sacrificial systems or information you can offer up as a decoy for a cyber-predator, in which case an attack becomes an advantage, allowing your organization to see the nature of the attacker and giving you time to add further security in the critical part of your information infrastructure.

In the end, we are only vulnerable to digital information threats because we are so dependent on digital information. We have, by choice and not, become enmeshed in an escalation toward ever more technological reliance. Yet sometimes technology that starts as an adaptation becomes maladaptive. Retroviruses, such as HIV, use the technology of our immune system against us. The BBC made a modern recreation of the Domesday Book in the 1980s, smartly storing it on high-tech (for the 1980s) laser discs, which are now less accessible than the original book from 1086, which was written on parchment.

Faced with continued technological escalation, the best strategy can simply be to step aside. Many successful organisms have split off from their species' escalatory pathways, so that the planet now has flightless birds, stingless bees, and rattle-less rattlesnakes. There are models in our past of how to work without information technology. News reporters, in the wake of the recent Justice Department blanket raid of AP phone records, are watching All the President's Men again and realizing the best way to talk with a source is not by email or text, but in a shadowy parking garage. I recall pulling out a notebook to jot some ideas during a meeting I had at the venerable Cosmos Club in Washington, DC. I was quickly and discretely chastised by my host, who informed me that one does not take notes in the Cosmos Club. No one would say this rule has hampered the many expeditions supported, deals created, and confidences shared in the Club's 135-year history, but it has preserved their integrity in a perpetually leaky city. Yahoo's decision to put a stop to employee telecommuting was made for many reasons (which vary depending on who you ask), but one of the underappreciated benefits is that it adds to the company's security by requiring fewer online conversations about new technologies and acquisitions. Not to mention petty spats between employees; now those are presumably carried out the old-fashioned way, in whispered hushes at the water cooler.

There are organisms that avoid security problems altogether. Certain deep-sea animals are so far removed from any competition that they live quite easily in their isolation. Unfortunately, they don't evolve and change, they don't transform resources or innovate — in fact, they don't do much of anything. Provided you want your organization to grow and innovate, you can't reject technology altogether and you can't wall yourself off from all threats. The best bet is to do what the most successful organisms on Earth do — accept the risk and adapt to the changes.

Data Under Siege
An HBR Insight Center





Welcome to the "Data Under Siege" Insight Center
Does Your CEO Really Get Data Security?
The Companies and Countries Losing Their Data
Hack-Proof Your Company's Social Media