People are losing it over AI. From analyst predictions, to industry events, to earnings calls, artificial intelligence and machine learning are all the rage. And dollars are flowing in response: the number of AI start-ups getting funding has spiked 4x from 2012 to 2016 (from 160 deals to 658 deals, according to CB Insights) and a recent Forrester study projects a 300% increase in corporate investments from 2016 to 2017. Yet, we’ve seen this hype before. Especially those of us who were at the front lines of AI back in the early days (or at least the mid-90’s!).
Yes, AI is super-exciting. Yes, machine learning (a sub-set of AI, but sometimes not) offers to revolutionize the way we monitor, and model and pattern match and interpret all the big and small data that is swirling around all of us. But no, it’s not all going to happen overnight. And (especially) no, many businesses and consumers still don’t have a clue about the best ways to select, apply and monetize AI for practical, everyday stuff. The following rules should help in this regard – but mainly are envisioned as a (time-tested) sanity check for surviving the latest edition of the AI hype machine.
Rule #1: Scope Matters. Creating a general-purpose thinking machine is hard. Creating an intelligent agent (or bot) that automates a single or small set of everyday, repetitive, “standard” tasks is a lot more tractable. Just as the key to early AI – and the Knowledge Management movement that followed – was finding narrow but high value applications like streamlining problem resolution in call centers or processing loan applications, the same type of “think global, act local” approach applied to today’s AI is equally important. For the same reason, starting with small-ish data vs. super-large big data sets can make sense when applying analytical techniques to many non-scientific business applications (more on this next).
Rule #2: Machine Learning is Not Magic. And it’s not easy (for most) to get right the first time either. Experimentation with a tool like RStudio is key, and there are many algorithms to choose from (Bayes, decision trees, regressions, and once-again popular neural network models aka “deep learning”). Of course training deep learning models can be both an art and a science. However, the good news is that an excellent recent article by Andrew Beam shows that you don’t need Google-scale data to use deep learning. I saw this in some of my graduate work when I was training neural nets to do simple pattern recognition, and it’s great to see this type of small data approach continuing to stand the test of time!
you don’t need Google-scale data to use deep learning”
Rule #3: Data is King. Getting close to customers, understanding their journey, tailoring their experience, and selecting just the right offer are all outcomes that are enabled by insights powered by big and small data. Generating these insights in a timeframe and cost that make them readily available to front line teams (and consumers themselves) is where advanced analytics and techniques like deep learning need to go. But as mentioned above, what you get out is very much a function of what (data) you put in. Where will your training data come from? How will you prepare it? Who will test the performance? These are questions as important as what tool or algorithm you’ll use.
Rule #4: People Matter. Even as AI systems become more skilled at complex decision-making, and take over some “back of house” functions previously performed by humans, we are a long way from creating virtual human beings, a point that Om Malik made in his excellent piece for The New Yorker last summer. Which is why some of the best, most impactful use cases will continue to augment rather than replace human workers, such as the AI voice analysis and feedback system from Boston-based Cogito that gives real-time guidance to employees as they engage customers on the phone, or the “davis” AI powered virtual assistant for IT ops from APM pioneer Dynatrace.
Rule #5: Consumers Don’t Care About Your Technology. Data nerds want to know what flavor of machine learning you are using. If you are selling to them or other techies, than skip to Rule #6. For everyone else, take note that it’s more important to focus on the “why” than the “how” when selling the value of your AI initiative to internal or external stakeholders. Why is the problem interesting? Why is it hard to solve with traditional (non-AI) approaches? Why is this repeatable/scalable vs. one-off solution? Even more so, what unique value is AI providing to your initiative or app? And how will you show ROI going forward?
Rule #6: Embedding AI Drives Adoption. Back in the day, the old joke among AI researchers was that when something in AI become successful, it wasn’t called AI any more. Today, many successful AI powered apps and services have AI “in them” but the technology is not apparent to the end-user. And that’s the point really – embedding AI drives adoption. Fortunately there are a growing number of tools to add AI or machine learning or other intelligent capabilities. These include open source development frameworks and engines like Apache Lucene (NLP) and Mahout (ML), Eclipse BIRT (Developed by Actuate – now part of OpenText) for embedded analytics and visualization, and RapidMiner for machine learning; embedded analytics specialists like Izenda and Sisense; developer platforms like IBM Watson APIs (conversation, speech, vision) and Microsoft Cognitive Services (decision rules, search, vision); and even custom hardware like Nvidia’s Jetson TX2 card.
Rule #7: Focus on Improving Everyday Work. Much of my research and writing over the past few years has focused on turning small and big data insights into everyday value. For marketers there are established use cases for data-driven marketing (see some of them in the piece I did for DMN a few years back), and there’s also a helpful framework for considering which marketing processes are mostly likely to be disrupted by AI from the folks at TopRight Partners. And for others looking at the bigger picture, there’s a very cool study (and poster!) on the overall potential of automation in the workplace – “Where machines could replace humans” – from McKinsey that is worth checking out.
Last week I gave a talk at the excellent SPARK Boston event along with some other marketing heros of mine including CC Chapman and Kyle Lacy. Of course the topic was Small Data, with a brief history of the topic and a drill down into my thoughts about the movement providing key inputs into a new design philosophy. I also teased the new mobile app my colleagues have been building with the smart folks from Raiz Labs – which provides a great use case for small data in action.
You can check out my entire presentation and a few highlights of our new app in my SlideShare below. Enjoy!
In a thought provoking interview with CNET published this past week, Fitbit designer Gadi Amit explores the use of wearables in everyday applications – and introduces the notion of “wise” devices that provide just the right information, when and where we need it.
Beyond the fact that Amit’s firm is designing wearables for unique markets like babies (well I guess really for their parents) and pets (!), what struck me in this piece was Amit’s perspective – certainly shaped by his role as president and lead designer at design firm NewDealDesign – on the state of wearables, their future, and our relationship with them. Specifically, in response to a question about how wearables will be integrated into our daily lives, he states:
The interesting thing is when I say that, people immediately jump to the conclusion that we will be cyborgs. My goal with designing this is that we won’t be cyborgs. We actually will become more human and more free from the technology. What we have now in the design business is two camps: there is the camp that wants to create a lot of data and wants to analyse a lot of data; and there is the other camp which I belong to that tries to create devices that are not smart, they are actually wise. They are more than smart, they are wise enough to understand you, to filter and allow you to go on with your life with all their data processing in the background giving you hints of what is essential when it is essential.
Having data processing in the background and focusing on what information is essential is of course very much in line with the small data “aesthetic” we’ve been promoting here and in a number of venues over the past 2 years, so it’s cool to hear validation from another corner. As a former AI/machine learning guy, I also like the idea of “wise” devices that understand context and personal preferences, and can make a case that small data will in fact be the new “OS” for these devices (more in a future post).
But even more so, if we think of the cyborg comment as a challenge to all of us, I think we need to consider the element of “humanness” as we create new apps and digital experiences. And perhaps provide better opportunities and incentives to untether/unplug (partially?) from our digital devices, even as consumers clamor for faster, more personal, more portable, and ultimately more satisfying data-enriched experiences.
Designing Data-driven Apps
Speaking of the new data consumer, I’ve been spending more time with developers and those thinking about the future of customer facing apps, and recently created a talk on design principles that builds on some of the work you’ve read about on this very blog. As always I believe that data-driven design is an art and a science, so it’s been fun to brush up on the science/tech part for sure.
Of course our first job is still to think about the end-consumer, and how we can inform, connect, and motivate them to get involved or take action. As an aside, if you’ve paid attention to how I’ve presented this last point, I’ve always used Nike Fuelband as my example, so with news that Nike is getting out of the fitness hardware business (good analysis in this Gigaom piece), it’s been interesting to see Fitbit and even Samsung step up their efforts ahead of the likely fall iWatch debut.
On the business side, beyond understanding the value of data along the customer journey and focusing on “last mile” functionality, having a scalable foundation that can potentially support millions of users and large data sets from many sources (before it is transformed into useful small data) is essential as we look to bring powerful, yet human-scale, smart (wise) apps to the masses. So is a community to drive innovation – like the 3.5 million BIRT developers, or 600K+ Drupal users and coders.
Many of these ideas (and some examples) were covered in the talk I did with SD Times recently. There’s a link to the replay and a summary by my colleague Fred Sandsmark on the Actuate blog – which you can read here.
I also presented a longer version focused on bringing the power of advanced analytics to “everyday tasks” at the CAMP IT big data event this past week, (a well-produced event by the way) and plan to post those slides to my slideshare shortly.
Finally, I will be moderating a very cool expert panel on “building the next big app” at a special event Actuate is hosting in San Jose on the evening of July 10. Scheduled to join me on stage will be Eclipse Foundation Executive Director Mike Milinkovich, plus industry watcher and enterprise apps futurist Esteban Kolsky, along with 1-2 other special guests. We’ll explore how consumer experiences will (and are) be shaped by new devices and data, open source driven innovation, and next-generation design tools and practices.
Be sure to let me know if you’ll be in the area and want to join us, since I have a limited number of VIP passes to share.
The small data movement continues to gain followers. In fact, just in the past couple weeks we’ve seen articles in a wide range of publication including Ad Age, CITEWorld (smart data!), InformationWeek India, RetailWeek, and the SmartData Collective, as well as new opinion pieces on the problems with big data in the New York Times and FT magazine. Meanwhile one of the more popular sessions at Adobe’s marketing summit tackled the topic and featured my former colleague Scott Liewehr who reviewed findings from the study we did at DCG last summer (see nice recap of the session by Trueffect here).
As I’ve discussed over the past 2 years, many of the benefits of thinking small focus on making personalized insights more pervasive via new, smarter devices, tools, and apps. Bringing big data down to size helps both decision makers and “everyday” business users or consumers find (just the right) information/content they need, in a format that makes it super clear and easy to take action. This to me is the essence of what it means to be data-driven, and in many ways is powering the excitement around wearable technology like Pebble, Fitbit, and Google Glass.
So, for consumers, small data continues to be about connecting (with data and peers), convenience, and portability. Yet for businesses looking to tap the power of big data for everyday decision making, it’s also about speed. In fact, one could argue it’s mostly about speed and being faster to market and more responsive to customer needs. How so?
As brands strive to be more data driven in their product development, marketing campaigns, and experience design, thinking fast means getting actionable insights in minutes and hours vs. days and weeks as we’ve become used to in “traditional” BI and Big Data (thanks to having to wait for IT resources or our neighborhood data scientist to pay us a visit!).
In today’s business environment, it’s increasingly tough to justify these bottlenecks. Especially when many business users are already part of the way there with tools like Excel or Tableau, and others have seen the potential of a small data approach via their experiences with consumer apps and Websites.
The bottom line? When social and mobile are the rule, and data increasingly drives competitive advantage, speed, and being more responsive, is the new “big.” For business users, this starts with being ruthlessly focused.
Staying Focused to Move Fast
Moving fast, failing fast, even being an agile fast follower, are all proven business strategies. This is the case for organizations in growth industries, and especially for leaders who need to deal with external competitors as well as internal adversaries like budget cycles or organizational inertia. But to go fast, you need to have focus. And think small (and fast), when it comes to data access, discovery, and delivery.
In business, being focused is critical to bringing benefits of big and small data to the broadest set of users. And often requires that we reduce scope to just the essential questions/algorithms/visuals needed to help business analysts and their peers understand the essence of their data and optimize campaigns and experiences in near real-time.
Building on our small data design principles, this means striving for simplicity in our tools, being smart “enough” to handle role-specific tasks, being responsive in terms of portability AND delivering immediate value, and making it easy to compare notes and socialize findings.
Customer Analytics for the Rest of Us
Digging deeper, if we review our definition for small data, the last part (be accessible, understandable, and actionable, for everyday tasks) effectively becomes our checklist for designing and evaluating prospective customer analytics tools that are geared to the non-data scientist and everyday sales, marketing, and service tasks (like Actuate’s BIRT Analytics), as follows:
- Is it fast and simple to access all the data we need – from CRM, ERP, Web, campaign tools, social channels, and other sources – to create a true picture of customer and market needs (and easy to clean the data if needed)?
- Can non-technical users explore, enrich, and understand this multisource data in one place – creating an analytics “sand box” as Boris Evelson from Forrester has called it?
- It is easy for business analysts to create visualizations, share findings, and apply insights in new campaigns and experiences (via workflow and connectors) without needing expert assistance?
- Are common “everyday” tasks/use cases like segmentation, cross-sell campaigns, or social/Web targeting supported – and are there pre-build algorithms tied to these use cases?
Moving fast shortens the time to value, and allows more time for socializing insights and field-testing new campaigns and offers (as a type of “rapid prototyping” perhaps). Doing so requires simpler and smarter tools, and removing IT bottlenecks. But most importantly it means putting the power of advanced analytics in the hands of everyday business users.
Big data has never been bigger. On the heals of a new study (reported here) that shows that investors have pumped $3.6 billion into big data startups this year, and Gartner’s latest Hype Cycle which shows big data quickly approaching the top of the hype “cliff” (along with Consumer 3D Printing and Gamification), one is tempted to see more of the same in 2014. Yet all the steam coming out of the big data hype machine seems to be obscuring our view of the big picture: in many cases big data is overkill. And in most cases big data is useful only if we can do something with it in our everyday jobs.
Which is why against the backdrop of more hype (and more confusion) about the real role for big data, the case for thinking small looks more and more attractive. In fact, I recently penned an opinion piece for ZDNet in Paul Greenberg’s excellent column on Social CRM on 10 Reasons 2014 will be the Year of Small Data. You can read the full piece here, and here’s the Cliffs Notes version of the first 5 reasons:
- Big data is hard. Doing it at scale takes time, especially when the tech guys own it (I used to be a data scientist, so I can say this!).
- Small data is all around us. Social and mobile signals can help us understand customer needs – today.
- Small data powers the new CRM. Small data is central to the rich profiles essential to targeting offers and experiences.
- ROI is the thing. To realize the full value of data-driven apps they must be accessible, understandable, and actionable for everyday work (remember our definition).
- Data-driven marketing is the next wave. As more consumers want to access, and consume and even wear useful data, there’s an unprecedented opportunity for savvy marketers and platform providers like Adobe, HubSpot, and Salesforce, along with data specialists and tool providers like Actuate, Localytics, TrueLens, and Visible to power the next wave of apps, experiences and devices.
What do you think? Are you finalizing your big and small data strategy for 2014? If you’d like DCG’s perspectives as you formulate your plans I’ll be hosting a special Webinar on Dec 12 at 1:00pm ET to share our outlook and 2014 research agenda. Hope to see you there!
UPDATE: missed my Webinar? You can view the replay by clicking here.
With a flurry of new articles on small data in various publications recently – including Wired, Technorati, and AllThingsD – there’s a growing number of voices contributing to the small data movement (a great thing!). But with these new perspectives, I feel like it’s a good time to re-ground the conversation in an actual definition of what we are talking about when we refer to “small data” – or at least what I’m talking about!
So in this post I wanted to share my first cut at a proper definition. Yes, I’ve framed the pillars for small data in many other places, going back to my first guest piece in Forbes on the topic a year ago (!), and more recently I’ve embraced the idea of describing small data as “the last mile of big data.” But these were descriptions or principles vs. definitions for the most part.
So, after spending the last couple month working on Digital Clarity Group‘s new multi-client study on “Bringing the Power of Big Data to the Masses” (sponsored by my friends at Adobe, Actuate, HubSpot, and Visible), seeing how marketers are looking to make analytics more accessible and actionable – and creating some new starter use cases, an interesting thing happened: a definition emerged!
In fact, while our final report is still a week or so away from being available from DCG and our sponsors, I shared the definition with our audience at the Digital Pulse Summit this past week during my panel on small data, and given the response, I wanted to provide it here as well.
A New Definition for Small Data
Small data connects people with timely, meaningful insights (derived from big data and/or “local” sources), organized and packaged – often visually – to be accessible, understandable, and actionable for everyday tasks.
Note, as we describe in our report, this definition applies to the data we have, as well as the end-user apps and analyst workbenches for turning big data set into actionable small data. The key “action” words here are connect, organize, and package, and the “value” (the 4th V of big data) is rooted in making insights available to all (accessible), easy to apply (understandable), and focused on the task at hand (actionable).
In fact, I hope it’s as much a mission statement, as it is a definition. What do you think? Did we nail it?
Last week I gave the closing keynote at IMS San Francisco on the topic of Rich Media, Personalization and Small Data. The event – which brings together top marketers and thought leaders in content marketing, social media and advertising – was a great forum for testing some of my latest thinking on content and scale. And specifically the growing shift from big budget campaigns, long-form video, and big data, to word of mouth viral campaigns, short-form content like vine and snapchat, and small data.
As I shared, this shift impacts (and enables) how marketers can create more effective way to reach and engage customers AND employees, and how they can bring their stories to life. To do so, I argued we need to align around 3 goals:
- Inform – Be HELPFUL! People pay attention if you’re saying something useful or unique – and you reach them when they are thinking about what you are saying,
- Connect – Help customers connect with you and each other…and
- Motivate – The essence of influence is motivating a behavior; in the digital world, content + context drives participation. The ultimate goal of content marketing (and small data) is creating this action or behavior. No participation or action = NO ROI!
With these goals in mind we can map out specific tactics like making video content and your YouTube channel part of your product strategy, and bringing data to where your users are, via mobile apps, social coupons or QR codes.
You can check out my entire presentation and see other examples and takeaways in my SlideShare below. Enjoy!