Is your data cheating on you? Five life lessons from the Ashley Madison hack

6322992684_d252f695ed_oIf you’re not one of the 37 million people whose data was hacked in the Ashley Madison breach, you can breathe a sigh of relief.

Sort of.

The Ashley Madison story may be great for a few news cycles of schadenfreude, but it also illustrates the realities we face in the age of data ubiquity: as people, consumers, businesspeople, patients and citizens.

1. Intimate data about us is everywhere. Our purchases, location, sexuality, religion, health history, political party, whose house we went to last night, the stiletto heels or sleek watch or expensive bourbon we clicked on on a website–is out there, somewhere. In most cases this data is protected by layers of security, encryption, policy and regulation, but, as we’ve seen from Anthem to Target to Ashley Madison–it’s not always effective. Beyond data security, however, is the question of how this data is actually used by the businesses that collect it. Is it to deliver better services, products ads? Is it being sold to a third party?

2. Profiling is not just for the FBI. Marketers love profiling. Why? Because good marketers realize that it’s good business to sell you something you are likely to want, rather than wasting your attention (and their money) on trying to sell you something you don’t. So, naturally, they want to know more about you: who you are, what you covet, where you shop, where you live, how old you are and how much money you have, so they can target ads and products and services more effectively. Whoever you are, you’re profiled somewhere: thrifty boomer, young married, millenial hipster; sounds like a Hollywood casting call, doesn’t it? Like any tool, profiling can be extremely effective when properly used, dangerous if not.

3. You leave digital footsteps everywhere you go, and they just may live forever. Everywhere you go, you leave digital traces. Even if you were “just browsing” in a store, you may have left a digital trace if you used a retail app, and/or the store used beacons or shelf weights. Add to that your web, mobile and social activity, and any apps you’ve used. Now imagine a ten-year timeline of that data being used to try to predict your next purchase. Or next spouse.

4. Chances are, you haven’t the slightest idea what data is being collected about you at any given time. If you want to do a simple test, install Ghostery on your web browser for a while. It’ll tell you what data is being collected by the website you’re using. Did you know this data is collected? Do you know how it’s used? I bet not.

5. Your data may be cheating on you. When you clicked “Accept” on any one of a number of apps you used, or bought a book, or downloaded a movie, you may have digitally consented to share this data with third parties. But did you really know what you were consenting to? Sometimes this is a non-issue (some companies will never share your data with others). Sometimes it can have uncomfortable implications, as when Borders declared bankruptcy, and decided to sell one of its greatest assets–its customer purchase history. (The FTC stepped in and required Borders to provide an “opt out” option).

To be clear, I’m not saying any of this is inherently bad, or suggesting we can roll back the clock; it’s just reality these days. But as data becomes more intrinsic to our lives and our business, I believe in finding “teachable moments” anywhere we can:

  1. As individuals, there will never be a better time to educate ourselves about what tradeoffs we are making, consciously or unconsciously, with our our data.
  2. As business people, we need to decide what kind of data stewards we will be, especially as data becomes more ingrained in business strategy.
  3. As an industry, we need to start putting clear and practical norms in place to clarify these issues so that we can have a fair and productive conversation about them and, frankl,y set a good example.

I’ve outlined a lot of these issues and recommendations in The Trust Imperative: A Framework for Ethical Data Use. If you’re not lying on a beach somewhere, I’d love your thoughts and feedback.


Posted in altimeter group, behavior, data privacy, data security, digital ethics, Ethics, Internet of Things, Predictive Analytics, Privacy, social data ethics, Susan etlinger, Uncategorized | Tagged , , , , , | Leave a comment

Altimeter Group Joins Forces with Prophet

In the flurry of excitement today, I wanted to make sure to mark a momentous occasion; the company I work for, Altimeter Group, announced today that we have been acquired by Prophet. It’s a great move for both teams; we have similar clients, outlooks and cultures, which always makes for a great partnership. And we can help each other in may ways.

Here’s a video describing the new relationship, and what we hope to do together…

…and a nice story by Anthony Ha in TechCrunch. I’m thrilled for Charlene, who has taken this company to an important milestone, and for my colleagues old and new.

More to come!

Posted in Altimeter, altimeter group, Uncategorized | Leave a comment

What Brands Can Learn From Pinterest’s Privacy Updates

Screen Shot 2015-06-30 at 10.17.03 AMIn the midst of all the complexity and fear about data usage and privacy, it’s nice to see an example of disclosure done well.

A couple of weeks ago, Pinterest announced Buyable Pins, which will enable their users to buy products directly from Pinterest on iPhones and iPads. Like any new feature, this one comes with data privacy implications: if I buy something on Pinterest, both Pinterest and the seller will have access to this transaction information–and possibly more about me.

I’m a Pinterest user myself, so last week I received this email.

Screen Shot 2015-07-06 at 9.59.26 AM

Long story short: Pinterest and the seller receive enough information to complete the transaction, facilitate future transactions and make promotions more relevant to me. If I don’t want to share information to customize my experience, I can turn it off. Short, sweet and to the point.

If I want more information, Pinterest’s privacy policy covers a range of other issues in similarly clear language. The other thing I like about it is that it prompts me to dig deeper if I want to. Clearly, this should be true of any privacy policy update, but the naturalistic and concise nature of the language makes that process a little less initimidating.

I asked the Pinterest team what they were trying to achieve with the privacy language, and here’s what they told me:

Buyable Pins has been a highly requested feature, so we wanted to make sure the language for the policy was clear right from the start. The goal was for Pinners to have an understanding of why the updates are being made, how they can customize settings, and where they can learn more. The approach was similar to past policy updates, where we aim to put Pinners first and be as helpful and concise as possible.

There are two really important issues at play here: 1. people have been asking for this feature, so there is going to be a lot of scrutiny among the pinner community; and 2. Pinterest is now dealing with people’s money. So there’s a lot at stake.

Privacy Policies in Context

Two weeks ago, we at Altimeter Group published The Trust Imperative: A Framework for Ethical Data Use. The central framework in this report combines the data life cycle with ethical data use principles developed by the Information Accountability Foundation (IAF).

Screen Shot 2015-06-25 at 9.55.06 AM 1

The Pinterest privacy policy explicitly fits into two areas of the framework:

  • Collection and Respect. Have we been transparent about the fact that we collect data?
  • Communication and Respect. Have we communicated clearly about what information we collect, and why?

This is why our use of language is an ethical choice:

While dense and legalistic language may satisfy the legal team, clear and simple language demonstrates respect for the user. 

You could further state that Pinterest, like many other ad-supported sites, is arguing that increasing the relevance of promoted pins is a benefit to pinners, which would cover Collection and Benefit as well. [That argument only holds up if users agree that the benefit is worth the exchange of data.]

This is not to say that a privacy policy is the only thing organizations need to consider when it comes to ethical data use. Many other issues have gotten organizations into hot water, whether in courts of law or public opinion. Some top-of-mind examples include Borders (for attempting to sell customer transaction data as part of its bankruptcy process) or Anthem and others (for data breaches). These examples map to Respect/Fairness and Usage, and Respect/Fairness and Storage and Security, respectively.

But now that the framework is out, I will be testing it (and suggest you do too) against real-world examples, using the IAF principles and the data lifecycle stages to examine and illustrate examples of ethical data use in theory and, most importantly, in practice.

Posted in Analytics, data privacy, digital ethics, Ethics, iPad, Pinterest, Privacy, Social Data, social data ethics, Uncategorized | Tagged , , , , , , | Leave a comment

The Trust Imperative: A Framework for Ethical Data Use

Screen Shot 2015-06-25 at 9.58.13 AM 1Consider this: Consumers don’t trust the way organizations use their data. CEOs are concerned that lack of trust will harm reputation and growth. People who don’t trust companies are less likely to buy from them. Yet the default option for businesses using consumer data is that it’s a right, not a privilege.

That tide is turning, and quickly.

Altimeter Group’s new report, “The Trust Imperative: A Framework for Ethical Data Use,” explores the dynamics driving people’s concerns about data use, recent research about their attitudes and behaviors, and proposals by industry leaders such as The Information Accountability Forum, The Governance Lab (GovLab) at New York University and The World Economic Forum.

Our objective for this research is to propose an approach for data use that reveals insight and honors the trust of consumers, citizens and communities.

And, while ethical data use is a fraught issue today, it will be even more so in the near future. As predictive analytics, virtual reality and artificial intelligence move into the mainstream, the implications (and capabilities) of data will become even more urgent and complex.

We no longer live in a world where privacy is binary; it’s as contextual and fluid as the networks, services and devices we use, and the ways in which we use them.

The next step is to make the topic of ethical data use–admittedly a broad and undefined one–pragmatic and actionable. We do this by bringing together the principles of ethical data use developed by the Information Accountability Foundation (IAF) and the specific stages of data use into a cohesive framework.

Our thesis is simple: ethical data use must be woven into the fabric of the organization;
weakness in one area can leave the entire organization exposed.

Screen Shot 2015-06-25 at 9.55.06 AM 1

In addition to this framework, the report lays out an argument for why ethical data use is a brand issue, annotated by examples from multiple industries. It includes actionable recommendations to enable organizations to apply these principles pragmatically.

Clearly, this is just the beginning; we will continue to deepen this research and learn best practices from academics who are exploring these issues and businesses who are faced with these questions and challenges every day. We are also in the process of building a roadmap for ethical data use from this framework that will help organizations assess and remediate the risks (and uncover the opportunities) related to their use of data.

My thanks to everyone who, implicitly or explicitly, contributed to this report. Most importantly, I’d like to express my deepest gratitude to the Information Accountability Foundation, whose excellent “A Unified Ethical Frame for Big Data Analysis” underpins this work. I would also like to thank my colleague Jessica Groopman, who collaborated on the research and framework development, and whose excellent research on “Privacy in the Internet of Things” will be published shortly.

As always, we welcome your feedback, questions and suggestions as we work to add clarity and action to a complex topic.


Posted in altimeter group, Analytics, Artificial Intelligence, behavior, Big Data, data privacy, Data Science, digital ethics, Ethics, Internet of Things, Law, Privacy, social data ethics, Uncategorized | Tagged , , , , , , , , | Leave a comment

Is Twitter Obligated to Preserve Politicians’ Deleted Tweets?

Twitter_logo_bluePoliticians everywhere are probably celebrating like crazy this week.

The Sunlight Foundation, an organization dedicated to making politics accountable and transparent, was just told by Twitter that it would no longer have access to its developer API, which enabled the site’s “Politwoops” app to track politicians’ deleted tweets. The story has been covered in news outlets from Gawker to The Washington Post, but the consensus is pretty much the same: Twitter was wrong.

This little story explains a lot about why the digital medium is so confoundingly hard. As citizens, we want to know that politicians’ communications are preserved; after all, as Philip Bump argued in the Post article, “The reason Anthony Weiner is no longer a member of Congress is because he sent a photograph of himself in his underwear to someone on Twitter.” Politwoops is further credited with capturing deleted tweets from politicians on everything from Cyndi Lauper’s hotness to backtracking on Bowe Bergdhal.

It’s a fair point. Richard Nixon resigned because the release of the Watergate Tapes (and revelations about that 18-minute gap) made it impossible for him to continue to serve as President.  As Justice Louis D. Brandeis famously said, “sunlight is the best disinfectant.”

But unfortunately it’s not that simple in this case. Bear with me.

The Sunlight Foundation issue brings back the question, which has existed since the beginnings of the Internet, as to whether a site such as Twitter is more like a magazine (which created the information it contains) or a newsstand (which simply makes it accessible to the public). My bet (and I have no inside information here) is that Twitter would argue that it is more like a newsstand, where the user (in this case the politician) is the magazine. So if a politician decides to remove content, Twitter would have no obligation (and actually might incur liability) by replacing it. This example also shows the limits of interpreting 21st-century realities using 19th century concepts, but that’s another issue for another day.

So, as a legal expert explained to me, the public may have an interest in seeing deleted tweets, but in the ordinary course, they wouldn’t have a right to see them. I imagine that the phrase “in the ordinary course” is the key here; if there were a court order, that would possibly be a different story, and the Twitter lawyers would have to hash it out with the politician’s and the prosecutor’s lawyers. 

Now this isn’t to say that those deleted tweets can’t be captured at all; it just means that Twitter will not be party to it by providing access to its developer API. Doing so would mean the company is selectively enforcing its own terms of service, which could compromise the trust of all users. How would we then know who they enforce it for, and who they don’t? This issue becomes even more complex outside the United States, where privacy norms and the “Right to be Forgotten” legislation in the EU add a completely different dimension. We can’t forget that Twitter is a global company, after all.

So there you have it. As a citizen, I wish Twitter had made a different choice. As a user, I’m relieved they didn’t.

Posted in data privacy, digital ethics, Digital Media, Law, social data ethics, Social media, Twitter, Uncategorized | Tagged , , , , , , , , , | Leave a comment

Tim Cook Just Threw Down on Data Privacy, And It Was Awesome


Photo: Jessica Paterson, cc. 2.0

Apple CEO Tim Cook gave what TechCrunch called a “blistering speech” on data privacy Monday night at the Electronic Privacy Information Center (EPIC) “Champions of Freedom” event.

“I’m speaking to you from Silicon Valley, where some of the most prominent and successful companies have built their businesses by lulling their customers into complacency about their personal information,” Cook said. “They’re gobbling up everything they can learn about you and trying to monetize it. We think that’s wrong. And it’s not the kind of company that Apple wants to be.”

Cook’s speech has sparked thousands of news articles and questions about his motives, plans, wisdom in poking the bear(s), whether Cook–and Apple–is really prepared to address privacy head-on, and what it all means for the ecosystem of digital products and services.

After all, as Vala Afshar has said, “if the service is free, then you are the product.”

We all know that, right?!? Can we please just move on?

Tempting as it may be for those distracted by all that juicy data, there is a second, critical question.

What do we do about it? As technology developers? Organizations? Consumers?

This is a conversation that has to happen. Seriously. Now. With action. And clear outcomes.

Privacy is not about some vague, rose-colored future, it’s about trust, and what happens when consumers distrust the organizations with which they interact.

  • PwC has said that, as of last year, fifty percent of CEOs surveyed identify trust “as a real threat to their growth prospects.”
  • The World Economic Forum is examining what trust means, and how to decode it.
  • The Edelman Trust Barometer reveals how lack of trust actually affects consumer behavior (hint: it’s not good).
  • We at Altimeter Group are working on research on consumer attitudes about data privacy, and a framework for ethical data use, to come.

It doesn’t have to be a zero-sum game. Just because we honor privacy on one hand, doesn’t mean we have to limit innovation on the other. It just demands a new calibration of what innovation means, and what other models we can imagine to support both insight AND trust.

Tim Cook just threw down a pretty big gauntlet for the industry (and for Apple). Facebook, who he called out in his speech, is doing promising work with DataSift (disclosure: client) to deliver privacy-safe “topic data.” I don’t know what plans, if any, Google has in this direction, but the momentum of this issue can’t be lost on them.

There are no easy answers, but there are informed choices to be made. And better do it now, before the wave of predictive analytics, Internet of Things, augmented and virtual reality and other technologies yet to be invented really get going.


Want to talk to us about data privacy? Have something valuable and unique to add to the conversation? Please let us know.


Posted in Altimeter, Artificial Intelligence, data privacy, data security, digital ethics, Ethics, Facebook, Privacy, social data ethics | Tagged , , , , , , , , , , , | Leave a comment

Seeing through the customer’s eyes: the emerging science of photo analytics

Photo: Lu Lacerda, CC 2.0

Photo: Lu Lacerda, CC 2.0

I recently had the opportunity to moderate a panel at Oracle Data Cloud Summit 2015, and to deliver a speech at SDL Innovate. The themes, respectively: “Listening to the Customer’s Voice,”and “Unlocking the Value of Social Data.” But it’s not just the customer’s voice we need to care about. We also need to care about, and better understand, the customer’s vision.

This past weekend, the Washington Post ran a story about painter and photographer Richard Prince, whose slightly-reconfigured blowups of Instagram users’ photos were recently shown (and sold) at The Frieze Art Fair in New York–for a cool $90,000 each.

The focus of the Post article was on ownership of the Instagram photos themselves, and the flexibility of copyright laws related to images. But it’s not just ownership we need to think about with images; it’s the challenge of interpreting what they mean, so we can determine what action to take, if any.  Some use cases include:

  • Image-based UGC/Native advertising
  • Content marketing
  • eCommerce (increasing lift)
  • Risk management (copyright)
  • Risk management (brand)

The Emerging Science of Photo Analytics

When we think about understanding “the customer’s voice” through social media, we generally think about text. A tweet such as this one presents a few interesting challenges for a human (not to mention a machine) to interpret; the sentiment is mixed (love the watch, hate myself), the person may or may not be an owner (did he buy it?), and there isn’t really much other behavioral data (he tried one on and wrote about it) to interpret.

Screen Shot 2015-05-26 at 10.22.17 AM 1

Beyond that, we can look at the metadata to try to determine whether and how much it was shared, the profile of the poster, and so on. But these 14 words don’t tell us a whole lot; a listening tool will tell us that it is a brand mention, in English, with mixed (not neutral!) sentiment. That’s not a bad thing; it simply illustrates the challenges that we–not to mention machines–have with the 140-character format, and with human language as a whole.

Contrast the above with something like this image I found on Flickr:

Screen Shot 2015-05-26 at 10.44.26 AM 1

A human can detect that the focus of the photo is a butterfly-shaped object that started its life as a can of soda. A photo analytics tool such as Ditto could likely detect that it contains a partial Coca-Cola logo. If it were shared in Instagram or Pinterest, and contained enough useful metadata, Piqora might return it in a list of search results.

But the real question, if I’m a brand marketer at Coke, is this: what do I do now?

  • Is it a brand mention?
  • Is it positive or negative?
  • Can I tell the identity of the poster?
  • Is that person the same as the creator of the image? Of the butterfly?
  • How else do I categorize it?
  • Can I use it in native advertising, commerce or other types of campaigns?
  • Is it a brand risk?
  • Is it actionable in other some way?

This is particularly critical as brands incorporate user-generated content into campaigns, and as automated campaigns become more popular. In February Coke launched, then pulled, their online #makeithappy campaign when a user added the hashtag to a tweet containing white supremacist content. Gawker then created a Twitter bot to test whether it could make the account inadvertently tweet lines from Hitler’s Mein Kampf. Short answer: it could, and it did, and Coke shut down the campaign immediately.

This is where brands must balance optimism with sometimes harsh reality: optimism that involving the community can be beneficial, and the reality that someone, somewhere will appropriate that content for their own uses: artistic, competitive, financial, political, comic, or just plain nasty. Clearly, malicious intent is more of an issue on some platforms than others. Instagram and Pinterest tend to be, as Sharad Verma, CEO of Piquora says, “happy platforms.” Twitter and others tend to have more problems with abusive content, as has been widely reported.

As you can imagine, it would be fairly easy to get lost in analysis paralysis with digital images, especially since we are still so new at the science of understanding and using them. So let’s start with a few basic principles:

  • Images are brand mentions
  • The science involved in interpreting them is still very new. The main approaches focus on interpreting the image itself, the metadata associated with it, or a combination (ideally) of both.
  • A few things are possible today with image recognition: logo identification and some (very basic) sentiment analysis. Given metadata, search becomes more feasible and useful.
  • Popular social platforms such as Instagram, Tumblr, Snapchat (and others yet to be invented) are highly image-centric.
  • Images are more universal than language, though there are always exceptions 
  • Automated campaigns will always hold some element of risk if they use or even simply suggest the use of UGC.
  • You’re going to need to figure this out sooner rather than later, because GIFs, video, spherical video (coming from Facebook), augmented reality and virtual reality are here, and they’re going to be a lot more complex from an analytical point of view.

I’d love to hear your thoughts on use of images and image analytics; what you’re using, looking at, thinking about, terrified of and excited about.





Posted in Analytics, Artificial Intelligence, Big Data, content marketing, content measurement, Crisis, digital ethics, Photo Analytics, Predictive Analytics, Real-Time Enterprise, Sentiment Analysis, Social Analytics, Social Data, Social media measurement, Uncategorized | Tagged , , , , , , , , , , | 1 Comment

Data Experience: You’ve Gotta Have Trust

Photo: Nic Taylor, CC 2.0

Photo: Nic Taylor, CC 2.0

A few years ago, I started thinking about how data informs the customer experience. The catalyst was simple; I was frustrated with my fitness tracker, and felt deluged by a stream of numbers that weren’t particularly helpful (the dashboards sure were pretty though).

Part of the issue was that all the design energy had gone into the physical device. For the time, it was cool and sleek. But syncing it was clunky, and it featured a proprietary metric that may have been useful for branding purposes but did nothing for me personally. So if insight was supposed to be core to the product, the product had failed, at least for me. But wearable devices aren’t the only products in which data experience is critical. Insight has become an expectation, an essential part of the customer experience overall.

And it’s not just the production of data that forms the experience; the way the product consumes data shapes our experience too. We need to trust that our data is being used thoughtfully and ethically, or any insight that it provides is meaningless.

When that relationship is out of balance, people find workarounds. Designers and developers have a love/hate relationship with workarounds. They hate them because they expose flaws, and love them  (one hopes) for the same reason.

If you doubt that trust is part of the customer experience of data, consider these fascinating workarounds:

  • A recent story in Alternet introduced a group of fashion designers and developers committed to helping consumers block facial recognition technology and confuse drones, among other things.
  • The makeup tutorial of your dreams nightmares: artist and film maker Jillian Mayer presents this YouTube video on how to apply makeup to hide from facial recognition apps and surveillance cameras.

If you dismiss these as edge cases, you’re well within your rights. But maybe that’s just today.

Isn’t it worth considering whether and how customers might be signaling distrust of your brand’s data experience? If I were Samsung’s Smart TV division, I’d look at how many people are disabling voice commands. If I were Facebook, I might look at photo upload rates over time, tagging behavior and changes to privacy controls.

What would you look at? As always, I appreciate your thoughts and comments.

Posted in Big Data, Data Experience, data privacy, Data Science, digital ethics, Ethics, Innovation, Internet of Things, Quantified Self | Leave a comment

What the end of the Twitter-DataSift partnership means for customers


Photo: Bob Kelly, cc 2.0


Let’s get this out of the way right up front. DataSift is a client of mine at Altimeter Group. I am closely connected to Twitter via my role as a board member of the Big Boulder Initiative, of which Chris Moody, VP Data Strategy at Twitter, is chair. 

On Friday, Zach Hofer-Shall, head of Twitter’s ecosystem, published a post on the Gnip/Twitter blog entitled “Working Directly With the Twitter Data Ecosystem.” Shortly thereafter, Nick Halstead, CEO of DataSift, published a post (since updated) entitled “Twitter Ends its Partnership with DataSift – Firehose Access Expires on August 13, 2015“. The next day, Tim Barker, DataSift’s Chief Product Officer, added another take: “Data Licensing vs. Data Processing.”

In a nutshell, this means that Twitter has ended its reseller agreements with third parties, and, going forward, will be the sole distributor of Twitter data.

To be clear, this does not mean that Twitter is discontinuing firehose access; in the future, Twitter will control its own data relationships rather than licensing data through resellers. 

These posts have sparked a flurry of commentary across a wide spectrum, from support to vilification to philosophizing on the meaning of platforms, analytics and ecosystems. I’ve included links to a few at the end of this post.

It wasn’t a huge surprise for anyone watching Twitter go public, and subsequently disclose the revenues from the direct data business, to anticipate that Twitter might realize that this was an area ripe for a significant strategy shift. And it was a short hop from there to conclude that Datasift’s (and possibly other’s) days of reselling data received via the Twitter firehose might be numbered.

It also hasn’t been a surprise to see Twitter enhance its analytics, re-evaluate the short-and-long-term value of its data and announce strategic partnerships such as the ones with IBM and Dataminr as it seeks to build its partner strategy and revenue potential.

Meanwhile, DataSift has continued to execute on its own strategy, which includes broadening its data sources far beyond social data, announcing VEDO, its data categorization platform, and via its developing its privacy-first PYLON technology (see its announcement with Facebook on how they are providing privacy-safe topic data).

Long story short: no one was shocked at the news. But the reaction to it has been polarizing in the extreme. What seems to have fanned the flames are a few different dynamics:

  • The context of the announcement, including Twitter’s recent moves against Meerkat and a historically fraught relationship with its ecosystem
  • The fact that Gnip and Datasift were competitors before the acquisition, complicating DataSift’s relationship with Twitter
  • Datasift’s strong public reaction to the news; and to some extent, the timing of the announcement, which came late on a Friday; a classic—though ineffective—news burial tactic.

It also doesn’t help that Twitter has been (except for the original post) all but silent this past week. But that shouldn’t come as a surprise to anyone either. It’s a public company, and, as such, required to comply with regulatory requirements governing communications with the public. According to Nasdaq, Twitter is expected to report earnings on April 28. So they’re in quiet period, and, as will surprise no one, won’t talk about confidential negotiations between parties.

As a privately-held company, however, DataSift has more leeway to comment publicly. I’m not going to repeat their position here; it is clearly stated in several posts on the DataSift blog.

But none of this gets at the most important issue: the impact of this decision on customers and users of Twitter data. Here are a few constituencies to consider:

1. Companies that sell social technology

To gauge the impact, you need to consider how these companies gained access to Twitter data before now:

How they get Twitter data Impact Implications
Directly from Twitter, via firehose or public API No change None
From Gnip (and therefore now from Twitter) No change None
From third-party resellers such as DataSift Ends August 13, 2015 Must re-assess how they migrate from DataSift to Twitter

But this requires a bit of context.

Before the acquisition, there was a reason companies—social technology or enterprise—would select Gnip or DataSift (or, before its acquisition by Apple, Topsy) if they wanted direct access to Twitter data: they had different value propositions.

DataSift positioned itself as a platform in which the access to social and other types of data came with value-adds such as advanced filtering, enrichments, taxonomies, machine-learning and other sophisticated data processing capabilities to enable users to derive insights from the data.

Gnip, on the other hand, was a simpler, less expensive option: they offered enrichments, but the value proposition was historically more about simplicity and reliability than sophisticated data processing. This tended to be an easy calculation for a lot of social tech companies who wanted to add their own capabilities to the data.

So, speaking broadly, analytics or social technology companies (even brands) who could or wanted to handle raw data would have been better suited to Gnip. Those who wanted a more plug-and-play product that processed data consistently across multiple sources would more likely favor DataSift. Once Twitter acquired Gnip, it didn’t take a team of data scientists to conclude that Twitter had bigger plans for its data business, and that a lot of that development would happen under the waterline, as these things tend to do.

But that doesn’t eliminate the very real issue that migration is going to be highly disruptive for customers of DataSift.

But there is another data point that’s important to consider.

Unlike most industry shifts, it’s been very difficult to get any social analytics companies to talk on the record about this news. On background, some stated that they were never quite sure whether DataSift intended to be a partner or a competitor because they weren’t a pure reseller; the platform—with its ability to perform sentiment analysis, enrichments, and provide classifiers and taxonomies—pushed it uncomfortably for some into analytics territory.

Some said they’re concerned about Twitter’s plans as well. Now that Twitter has discontinued data licensing, what will it now monetize? Will they take more control of  or develop their own analytics? If not, what then?

This is unsettling for some in the social analytics community, who are also being buffeted by business intelligence and marketing/enterprise cloud companies (think Oracle, Salesforce, Adobe) eager to wrap social data into a broader insight offering. It’s a time of shifting strategies and shifting alliances.

2. End users of technology (brands and enterprise)

For the most part, end users of Twitter data don’t have much to worry about, unless they are current or potential DataSift customers and can’t (or don’t want to) ingest the firehose in its raw form. If they are, they’ll need to migrate to Twitter, and assess the extent to which Twitter is currently (or via roadmap) able and willing to provide the type of processing they need.

If enterprises get their Twitter and other social data from social analytics providers, they are more insulated from this news. The question I would ask is whether and how Twitter intends to normalize data from other social media platforms. Will users have a clear sightline across multiple social (and other) data sources? Will they analyze more than Twitter data? Will they handle that through partnerships (and if so, with whom?) The ability to normalize data across sources has been a clear value proposition for DataSift; less so from Gnip, especially since the acquisition by Twitter. And of course we can’t discount the fact that Twitter likely has more up its sleeve than its able to disclose right now.

3. Agencies, consultancies and other professional services firms

Agencies can be affected in any number of ways, based upon their social (and other) analytics strategy and business model. Those who offer their own bespoke analytics would do well to learn more about Twitter’s data roadmap; how much will be product and how much service. Those who use a mix of analytics tools would likely be less affected.

As for professional services firms, there is a tremendous amount of opportunity in custom analytics for enterprise. The challenge is that 1) data processing isn’t core to that business 2) developing custom analytics doesn’t scale well and 3) Twitter data, especially in the context of other social, enterprise and external data, is extremely complex. As a result, professional services firms will need to approach Twitter to better understand what the company will and won’t offer in the future and where the synergies lie. Either way, it’s going to be a delicate dance.

For all DataSift customers, Tim Barker’s post is a shot across the bow for Twitter; even if Twitter disagrees with his assessment of the impact of terminating the reseller agreement, it’s a starting point for customers to begin their conversations with Twitter and suss out what exactly a shift to a direct relationship might entail.

One other option is for customers to bring Twitter data into DataSift via a Gnip connector. DataSift is working on that now.

A few last thoughts

In the end, a lot is still unknown, and Twitter’s silence is enabling people to fill the void with speculation, which creates uncertainty and doubt among those whose businesses depend on to some extent on Twitter. But in my opinion, all of this is an inevitable if painful blip for both companies: DataSift will move on, and Twitter will continue to build out its data business, which will likely create even more uncertainty in the social data ecosystem for some time to come.

But, as one person who spoke off the record rather philosophically commented, “In a world where you’re dealing with third-party data, you can never be completely comfortable.”

 Other points of view




Posted in Analytics, Big Boulder Initiative, DataSift, Facebook, Sentiment Analysis, Social Analytics, Social Data, Social media, Social media measurement, Twitter, Uncategorized | 2 Comments

Who will win the cloud wars? (Hint: wrong question)

Photo: Nick Kenrick, cc 2.0

Photo: Nick Kenrick, cc 2.0

Like many analysts, I’ve spent the last month or so crisscrossing the country looking at clouds*: marketing clouds, sales clouds, service clouds. I’ve been on busses, at dinners, in keynote speeches and presentations and one-on-one meetings. I’ve collected more badges than I did my first (and only) year in girl scouts.

One of the things that stands out throughout all of these presentations and conversations is how tactically similar yet philosophically different each approach turns out to be. Salesforce, as always, is most forward in its messaging. Adobe is doing a lot of work under the waterline to deepen the integration (and therefore the value proposition) of its marketing cloud. Oracle is using its expertise in data to paint the picture of a customer-centric enterprise. And yes, each approach has its standout strengths and clear weaknesses.

The fundamentals—whether it’s the presentation, the messaging, the suite-versus-point-solution positioning, the discussions of acquisitions and integrations and features—are consistent. And so it tempts us to rate one over the other. Who will win the cloud war? Oracle, Salesforce, Adobe? Will Sprinklr be the dark horse? Someone else?

Except while we argue over the finer points of Adobe’s cloud strategy versus Salesforce’s versus Oracle’s, we overlook one basic fact. The war is over. And not in the “congratulations to the winner; please pick up your statuette” sense. It’s just that looking at it as a “war” doesn’t really cut it anymore.

Paul Greenberg incisively argues in a recent post that we’ve moved past the suite wars and into an age of ecosystems for CRM. Consider the collapse of what we used to call the marketing funnel: awareness, consideration, conversion. Leaving aside that human beings have never been that linear to begin with, the Internet and social web have conspired in the past 20-odd years to upend the “funnel” model.

Instead of a single workflow that culminates in victory conversion, we at Altimeter think about an “Influence Loop,” which lays out a more detailed version of the customer’s path, and which more importantly includes the experience after purchase—the point at which he or she is the most literally invested in the product or service. But another fundamental difference between the funnel and the loop is this: customers can and do communicate with each other in public at every stage. Sellers no longer have the information advantage.


As Daniel Pink put it in his keynote at Oracle’s Customer Experience World last week: (my paraphrase):

Information asymmetry is about ‘buyer beware’. Buyers have to beware because they can be ripped off. We no longer live in a world of information asymmetry; we live in a world where buyers have lots of information about sellers.

But this isn’t the only playing field that’s being leveled. Sellers now have more information about each other. They can integrate tools more easily, build products more quickly, gain access to open data. They can build ecosystems more effectively. Granted, none of this is easy, but it’s easier. And becoming easier still.

And so the dated, us-versus-them ethos of the 90s and 00s no longer applies. It’s not which cloud approach is better or will win. In fact, that old model actually damages each player’s position. Why?

If we’re fighting to beat our competitors, we’re not fighting for our customers. 

Instead, we should be asking:

  1. What is our strategy for serving customers in the cloud economy?
  2. How robust is our ecosystem?
  3. How strong are our offerings in sales, marketing, customer service, commerce? Where they’re not strong, is it in our roadmap, and do we make it easy to integrate other tools?
  4. What is our data and analytics strategy? Does it promote real insight?
  5. Is our offering siloed, or does it facilitate the customer’s sightline across the customer journey, and the business?
  6. Are we spending more time on our competition than on our customers?
  7. Are we thinking past the horizon?

And the most important one:

Do we make it easier for customers to understand and serve their customers?

Let’s be honest; nobody gets a perfect score on all of these questions. Each vendor has strong marks in some categories and weaker ones in others, and it’s always changing. Some have standout infrastructure and weak tools. Others have the opposite. And, yes, we still need to make decisions in a world in which 1,000-plus marketing vendors (never mind web, commerce, service, mobile, sales) vie for our attention.

This isn’t going to get any easier, so we need to start to think about technology selection a bit differently than we did in the past. If we don’t, given the speed of development and the pace of change in general, we risk entering into an infinitely spiraling arms race.

We also need to do some soul-searching. If we really are committed to (in that overused phrase) trying to understand the customer journey, are we willing to give up some short-term budget/power/control/ease-of-use to do so? What does that mean for performance management, bonuses, culture, leadership?

My colleagues at Altimeter and I will be exploring these issues in more depth in upcoming reports. In the meantime, I’d be grateful for your thoughts; what questions should we be asking as we assess cloud vendors? What are the great examples and untold stories? What’s keeping you up at night?

* Thanks and apologies to Joni Mitchell, who I hope is recovering quickly.

Posted in Adobe, cloud computing, Oracle, Salesforce, Sprinklr, Uncategorized | Tagged , , , , , , , , , , , , | 1 Comment