Author Archive


5 Essential Practices of the Data-Driven Organization

Written by Andrew Edwards. Posted in Digital Analytics

Analytics may be multi-channel these days – mobile, digital, social, email, and yes, even “web analytics” as once it was known. Analytics also means audience measurement via surveys and panel-based analytics as part of a site-ranking methodology. Whatever its latest form, analytics only helps when integrated into the creative and decision-making processes of the organization.

Too often analytics itself works in a small building off on the edge of the corporate campus. Occasionally a paper airplane flies into the executive window with a message from the analytics team. And the executive, on his way to a meeting, steps on the folded paper. It sticks to his shoe and by nightfall the janitor has swept it into a dustbin. So ends the influence of analytics in the non-data-driven organization.

Try to imagine an organization that is designed around intelligence gained from the data it collects about its customer interactions. This organization will have at least these five characteristics:

Define what it wants to measure. Analytics deployment experts will work closely with marketers and business owners to understand the key business drivers for the organization. These drivers will be closely aligned with actual business success. For instance, branding will want close interaction. Services will want leads. Sites that rely on advertising will want volume plus careful segmentation. The analytics experts will turn these needs into reporting structures that can answer important questions – going well beyond the baseline reporting that comes before customization. As important as deciding what to measure is deciding what not to measure. The organization will avoid overcrowded report structures; will have realistic goals about what is measurable and why; and will make sure expensive tools are not overburdened with whimsical profiles that never get viewed.

Deploy analytics tools expertly. That clamor you hear is the sound of the poorly chosen agency or miseducated IT team grabbing their pitchforks and torches as they try to kill this initiative – they may believe their jobs depend on “controlling analytics.” But in the data-driven organization, they are quietly asked to go back to doing what they do best – create great content; make sure systems function. Perhaps the most common difference between the properly aligned team and the data-flunking team is the way expertise is deployed. The data-driven organization will not blame the tool. It will not permit content creators to claim they “also do measurement.” They will audit their tags for completeness and functionality. They will have a neutral third party – expert specialists in analytics – deploy the tool with clarity of purpose, and without bias to anything but the truth about content success, campaign success, and ROI. They’ll know that any other road leads to costly waste, bad politics, and ultimately, no insight.

Analyze results and make recommendations. At the successful organization, reports won’t gather dust. They will be shared with the right people in the organization – people who can make decisions and direct action based on the findings. The findings will be clarified by people who understand how the data was gathered, can vouch for its accuracy, and put the data into context. In the same meeting, an action plan will be devised. What content will live – or be retired? Which partners seem to drive the most successful visits? What campaigns pulled not just the most visitors, but the most desired actions? Resource reallocation may be in order. Cuts in a place least expected. Additional effort toward an area that measures well – perhaps one that had been undervalued. The ability to reprioritize and react to data is one of the most important characteristics of the data-driven organization.

Create changes based on data. The developers may say they’ve got the best new wireframes this side of paradise, but somehow the numbers indicate rethinking the existence of that entire module. The content provider may insist on real estate where its click-throughs don’t pay the rent. The data-driven organization knows how to say “no” to the ineffective; and to insist on creative, content, and campaign changes based on what came out of the analytics. Was that partnership successful? Yes? Why? The data will reveal why – and the next partnership had better incorporate some of the characteristics of the winner. The organization won’t have time for slight tweaks to a tanking campaign – it will demand wholesale change if it’s needed. It will stand by the numbers, and insist its agencies and other content managers stand with them.

Measure again – and again. Done with that measurement exercise? Here’s the next. Analytics is a cycle of definition, measurement, planning, informed improvement, remeasurement, definition, measurement…forever. The organization knows that optimization is never one-and-done. It should not be enslaved solely to IT or developer “release cycles.” It should not come into the picture too late to have an impact. It should be as close to real time as possible without anyone twisting an ankle by being in too much of a hurry. Here is where you ask: how’d it go with those changes? Helpful? Better? Even a little better? OK, let’s make it better still. Incrementally, perhaps…but always on the road to Betterville. The data-driven organization lives in a cycle of measurement and improvement.

Do you recognize your organization in the above? If so, you’re in a great place. But chances are you’ll notice where your team is off in some other direction – they’re not on the proven path to improvement. And with soaring emphasis on digital enterprise, how long can you afford to be part of that team? Making the necessary changes may not be easy. Failing to take advantage of data is worse.

If your company is going to thrive in the digital marketplace, it will be as a data-driven organization.



Digital Analytics Is Not Surveillance

Written by Andrew Edwards. Posted in Privacy

Sometimes you have to take credit for your own foresight.

I am going to do that today.

Referencing the column I published two weeks ago – just days before the phone record and NSA PRISM scandals broke – I said “while privacy may be dead, nobody likes surveillance”; and that “we are one or two scandals away from this becoming a mainstream concern.”

Two days later, it became a mainstream concern with no less fanfare than a NASCAR winner guzzling milk from a big silver trophy.

Just say I had a feeling.

And now marketers have to deal with lots of new questions about “digital tracking.”

First, let me vent.

What the NSA has done is a disgrace to the U.S. Constitution and a subversion of the global Internet. If the terrorists were looking for a damaging result of their outrages, they now have it: the U.S. today spies on all of its citizens just like any number of shabby little despotic regimes. Just as any freedom-hating terrorist might have wished for, we have lowered our expectations about liberty to a position no higher off the ground than a limbo pole at the end of a long night at Sandals.

I am not one of those given to parading across the village green in a tricorn hat brandishing a hateful placard. However, I am a rational observer of history and this is not any kind of change I can believe in.

Our president says it’s all cool. I say: trust the government never to misuse data they have collected about me? Absolutely ridiculous.

And it would be funny if it were not giving me that weightless feeling you get when you first notice you are on a slippery slope.

Digital Marketing’s New Challenge

We all know about the need for actionable analytics (or we are probably not reading this column). Multi -channel, or convergence analytics depends on gathering data about what people do on their digital devices.

For the most part, Americans hadn’t really cared about how much they get tracked while shopping for baby clothes or wine using a digitally connected device. They didn’t care much because, being Americans, they trusted in their sovereign rights as U.S. citizens. Corporations might know all they could about your shopping habits, but the worst they could do was send you the wrong offer. OK, maybe if you were not paying your bills, some marketers or credit companies might put the brakes on your race toward the credit cliff. But that is all.

They could not, except under pain of prosecution for breaking the law, sell your data to a thief, or give your data to any enforcement arm of any branch of government (without the transparent, due process artifact of having been served a warrant generated via publicly accessible channels for cause).

It made for a very easy case against what some would have called “privacy zealots.” It would have gone something like this:

“What do you care if Williams-Sonoma knows that your computer connected with its server to load a page about a cappuccino machine? Do you put on an invisibility cloak when you go to its brick-and-mortar store?”

That was then.

Now we have to regroup, in light of the privacy zealot being able to crow “I told you so” about digital tracking. Because now we know that every search term you entered and every email you ever sent is available to any one of hundreds of thousands of cubicle-dwellers at Booz Allen (and many others) with the right security clearance. They can’t all be paragons of constitutional scholarship. Moreover, they are the outsource destination for a government that now seems to care little for the core document upon which it was founded.

Therefore, as “digital trackers,” we have some explaining to do.

Digital Analytics Is not Surveillance

And here’s why.

Unless and until the brand or retailer or services company whose site you visited is forced to turn over their analytics reports to a government-contractor spy agency, you cannot come to any harm via the data that has been collected about your (usually) anonymous interaction with their content.

Moreover, you will at some point enjoy a better experience while interacting with them because, by inference of your behavior, they will have worked hard to figure out what might work better for you next time and what might not.

In brief, your favorite digital retailer not only cannot put you in jail; instead, they are actually looking to improve your experience with their content.

The difference between digital analytics and government surveillance comes down to that. Does it not?

The private company cannot jail you. But the government, misusing information it obtained about you could, if it chose to, dispatch any of its hurts your way: the tax collectors, the investigators, the police, perhaps a drone if you really have rubbed someone the wrong way.

Fear of much of the preceding approaches the realm of paranoia – today. But admit it – in the back of your mind, this is exactly why it feels creepy that the government is collecting data. Because of what they might do with it. They’re not collecting it to see where to put a traffic light. They’re collecting it with the intention of identifying people they might want to put in prison. Of course, it isn’t you. But still.

The government needs to quit spying on Americans. It is making it harder to sell cappuccino machines!


4 Ways Digital Analytics Is Changing Forever

Written by Andrew Edwards. Posted in Digital Analytics

If you plan on getting your digital analytics done the way you’ve done it in the past, you may want to scrap those plans and start fresh.

Because in 2013, an entirely new approach to digital marketing and analytics is taking hold. Some have called it multi-channel analytics; some have called it business intelligence for marketers; some have called it (somewhat one-dimensionally) big data. I have called it convergence analytics.

Whatever you or your advisors want to call it, the facts about these important changes remain the same, and here are the four most important:

1. Everybody is measuring everything. And putting the reports in a dashboard. By everything, I mean everything. Begin with the desktop (this now means ” web”); add “mobile” (which means several things not very well-defined); add what is euphemistically referred to as “unstructured data” or social media; then add census or other data; CRM (like Salesforce); CDNs; predictive models; ad network data; geographic data; add in revenue data; add any data from any tool that has an API or SDK (and that includes about everyone). And you begin to get a sense of what “everything” is.

From the vendor side, the level of activity is nothing short of spectacular. To say that every company that ever measured a digital property now says they can connect to any data and visualize it for you might be an overstatement, but not by much.

From the marketer side, it’s going to suggest a need for re-architecting your entire approach to data. And possibly an entire new round of tool selections and skill enhancements.

2. Digital is getting married to television. We thought we were done with TV, right? Wrong.

All television is now digital (pretty much by government mandate). Which means TV really is just another IP address-driven content container.

Here’s why it’s important: people still love television. It’s engaging and compelling in ways that websites and apps never will be. It’s the movies. Except on a big flat screen (or a small flat screen) in your own home or office. And we haven’t yet figured out how much it will distort or even destroy the rest of the digital content universe. No matter what, we will have to figure out how to market there (again). Because it will be more like YouTube or perhaps Netflix than like “primetime.” And we will have to measure usership in more sophisticated ways than ever before.

3. Privacy may be dead, but nobody likes surveillance. I’m a digital analyst. So I’m not going to climb on a holy soapbox and complain about how corporations are tracking your every move online. They are. But then again, they’re not charging you any money for most of the stuff you use online. That’s the price users pay for “free stuff.” They pay with information about the way they interact with the content.

While Europe has gone buggy about privacy, in the U.S., only a few people really care about it. And even fewer ever do anything about it (like, for instance, reject or delete cookies at intervals).

But now that we’re starting to see so-called “surveillance scandals” at the government level, especially as regards the news media and beyond, the subject of privacy may well come into focus for the American consumer. The government can get records of almost anything it wants (and the recent advances in digital tracking make this a more important factor than ever). Most people try not to think about this. And they shouldn’t, as long as the government respects the individual’s right to due process. But we are one or two scandals away from folks deciding they don’t want to be tracked anymore because they’ve become paranoid of the Feds.

We don’t want that. So we should try to make sure the government understands that there is a huge difference between the shopkeeper knowing what frock you looked at while you were in the store, versus Big Brother knowing where you were the night of September 17.

4. Marketers are losing control of the data. To be fair, the concepts of “marketing” and “data,” while not quite an oxymoronic pairing, have never been a comfortable fit. Marketers aren’t typically wired for data. Measurement has landed on them and they have embraced it; and some of the most agile and forward-thinking have made big wins with data-driven decision-making.

But that was when we were talking about web analytics and a dash of campaign attribution. Now we are talking about an enormous new push by vendors and senior management to squeeze the marketer on measurement of ROI. Vendors have created powerful new technologies that need selling. Management has gotten the sense they can somehow know successful marketing through measurement; and thereby save tons of money. Both vendors and management are more or less correct. And marketers, caught in between, will again be forced to adapt to rapidly advancing technology.

It’s an open question whether they will, as a group, be able to do it. Or whether the entire task of digital measurement, having grown far more complex than it was even a couple of years ago, gets yanked from marketing and put back in the hands of data people who don’t report to marketers.

Marketers will need to get out in front and lead on this – or they will find themselves roasting in a rather hot sweatbox, penned in by data scientists using sophisticated tools that until recently were far too difficult to build and too expensive to buy except for the very largest and most data-intensive organizations.

Digital analytics is changing forever. Some might say it’s about time.


About Efectyv Digital

Efectyv Digital is focused on strategy for two distinct markets: digital analytics end-users; and marketing strategy for technology companies.

Click here to learn how how we can help your business grow >