Making Predictive Analytics a Routine Part of Patient Care


Over the last five years, electronic health records (EHRs) have been widely implemented in the United States, and health care systems now have access to vast amounts of data. While they are beginning to apply “big data” techniques to predict individual outcomes like post-operative complications and diabetes risk, big data remains largely a buzzword, not a reality, in the routine delivery of health care. Health systems are still learning how to broadly apply such analytics, outside of case examples, to improve patient outcomes while reducing spending. From a review of the literature on health systems that have successfully integrated predictive analytics in clinical practice, we have identified steps to make predictive algorithms an integrated part of routine patient care.

Determine the clinical decision. There is now a plethora of data available for nearly every potential clinical outcome. And where you have data, there is a potential predictive algorithm. But while it may be easy to develop clinical algorithms, it is equally necessary to be specific about which specific clinical decision(s) that algorithm will inform.

For example, there are many algorithms predicting a patient’s risk of hospital readmission (although the vast majority performs poorly). But simply knowing the percentage risk of readmission does not answer the questions that physicians and nurses typically ask before a patient is discharged: Should I discharge this patient now? Should I assign this patient to a readmission prevention intervention? Should this patient go to a short-term rehabilitation facility? Does she need a home care visit in the next two days?

Parkland Health and Hospital System in Dallas, Texas, has developed a validated EHR-based algorithm to predict readmission risk in patients with heart failure. Patients deemed at high risk for readmission receive evidence-based interventions, including education by a multidisciplinary team, follow-up telephone support within two days of discharge to ensure medication adherence, an outpatient follow-up appointment within seven days, and a non-urgent primary-care appointment. In a prospective study, the algorithm-based intervention reduced readmissions by 26%. Parkland’s success stems from focusing its algorithm on a specific population and tying it to discrete clinical interventions.

Leverage the data from EHRs. Algorithms are only as reliable as the data they are based on. While algorithms for acute clinical issues (e.g., heart attack, septic shock) may not require large amounts of data to predict risk, algorithms that utilize greater amounts of clinical data have greater accuracy and potential clinical applications.

The Veteran’s Health Administration (VHA), the largest health system in the United States, has collected electronic data from its patients for over three decades. Beginning in 2006, the VHA built a corporate data warehouse as a repository for patient-level data across its national sites. The sheer amount of inpatient and outpatient data has allowed the VHA to create comprehensive algorithms that reliably predict meaningful outcomes such as risk of death and hospitalization. Nurse care managers use these scores to guide intensity of outpatient services, including end-of-life and palliative care, delivered by multidisciplinary teams. The VHA’s investment in an integrated EHR and data repository — 5% of its total health spending — is substantial. However, the ability to reliably predict outcomes to improve quality of care may explain why the VHA’s net return on EHR investment is over $3 billion.

Focus on low-value decision points. Uncertainty over a clinical decision often leads physicians to overtreat or undertreat patients. Predictive analytics can allow clinicians to steer high-cost interventions to those high-risk patients who actually need them.

Consider the use of antibiotics to treat newborns. While less than 0.05% of all newborns have infection confirmed by blood culture, 11% of them receive antibiotics. Kaiser Permanente of Northern California has used predictive analytics to reduce this overuse. Its researchers have developed an algorithm to accurately predict the risk of severe neonatal infection based on a mother’s clinical data and the baby’s condition immediately after birth. Using this algorithm OB/GYNs can better determine which babies need antibiotics, sparing up to 250,000 American newborns each year from receiving unnecessary antibiotics. This could reduce medication costs and side-effects among vulnerable newborns.

Source: Predictive Analytics Times


The Digital Business Architect: Lessons from the Field

The other day I was contemplating what business capabilities were enabled by connecting a toaster to the Internet of Things. (Yes, this really exists. And no, I didn’t come up with an answer.) Clearly technology is having an impact on every aspect of our lives and many business architects now find themselves in a strange place: after years of trying to distance themselves from IT and technology and be seen as part of the business, now the business is trying to embrace technology and understand what a digital transformation could mean to their business model. So what is a business architect to do? Adapt!


Paradigm Shift
Years ago the mission of many IT shops was to “serve the business”. Corporate strategists would build grand strategies, business architects would develop a robust strategy execution plan, and then IT would bend-over-backward trying to deliver on what was asked. Then along comes companies like Amazon, Google, Uber, and Netflix, and entire industries are being disrupted and CEO’s are taking note. Digital is no longer an option: transform or disappear. Companies are pouring investments into understanding how technology can be used to enable new capabilities, differentiate themselves from their competitors, and move into new markets. And whenever a business needs to make a big change, it is prime time for business architects to open their toolboxes and get to work.

Architecting Digital
The foundation of business architecture is the use of business capabilities, and understanding what is needed before defining how to deliver it. Emerging technologies and digital opportunities certainly bring new technology into the equation, but there is also a material impact on people, process and data. Business capability modeling remains an excellent tool for gaining consensus on what the priorities are and the potential impacts before diving into how they will be delivered. Framing up digital opportunities for the business is an important first step. McKinsey provides an articulate framework on how digital can reshape every aspect of the modern enterprise:

This framework demonstrates that digital opportunities extend beyond well-known areas such as big data and process automation. What digital products could you offer? How could you improve integration with partners? Help your business teams think more broadly. If this leads to a desire to change the business model and move into new areas, new capabilities may be required and traditional business architecture techniques can facilitate this.

“Doing digital” may require changes to your capability model that you may not expect. For example, the ability to deliver software iteratively and quickly is crucial in the digital age. Managing partnerships with numerous niche product and service providers in a fragmented market of suppliers can be challenging. Obtaining and retaining talent in such a hot area is also difficult. These are just some examples of areas where your business may need to adapt.

Digital Skills
Since digital programs are top priorities at most major companies, get ready to skill-up. Here are some of the most critical skills to focus on:

•Business Acumen: As a business architect, hopefully this one is not too much of a challenge for you. As business and technology continue to become even more intertwined, having a deep understanding of the business model, strategy, key capabilities, and value streams is critical.
•Digital Consumer Trends: Stay current on how consumers are reacting to marketplace shifts and emerging technologies.
•Collaboration: The ability to drive alignment and collaboration with business and IT stakeholders across the enterprise has never been as important. In the digital world, new and different stakeholders may appear, such as data scientists, customer experience designers, and digital marketing managers. Getting all of them to work together is the challenge.
•Data and Analytics: Data is at the heart of most digital transformations and part of your job now is to harness the power of your data. You need to know ways to monetize your data, develop deep insights into customer behaviors, leverage the Internet of Things, and improve decision making at all levels.
•Emerging Technologies: The pace of technology advancement is not slowing down, so be sure to devote time to staying on top of what is coming, what is already here, and how it can help your company. How do you do this? Read, listen, and experiment!
•New Ways of Working: The focus on speed and agility isn’t going away, so you need to understand new ways of working, such as bimodal, agile, and scrum.

In summary, adapt, and do it quickly. There is no better time than now to build up these skills and provide your leadership something they are looking for: someone they can trust to successfully guide them through identifying what is needed to enable a digital transformation.


Data Scientist – best job in America, again

The popular job site Glassdoor published a list of 50 Best Jobs in America, and Data Scientist is again the no. 1 job in USA, with Job score 4.8 out of 5, $110,000 Median Base Salary, and 4,000 job openings.


Half of the top 10 jobs are related to Analytics, Big Data, and Data Science!

Rank Title Job Score Job Satisfaction Median Base Salary
1 Data Scientist 4.8 4.4 $110,000
2 DevOps Engineer 4.7 4.2 $110,000
3 Data Engineer 4.7 4.3 $106,000
5 Analytics Manager 4.6 4.1 $112,000
7 Database Administrator 4.5 3.8 $93,000


Compared to Glassdoor 2016 post where Data Scientist was also no. 1 job in USA, we note that the median Salary has declined from $117 to $110, but the number of listed job openings has increased from 1,700 to 4,200.

Indeed job trends also show continuing growth in the demand for Data Scientists:

You can also reach those talented but hard-to-find Data Scientists by placing a job ad on KDnuggets – email to See Details and posting info.

According to CareerCast recent report, Data Scientist is the most demanded job in 2017. Here are their top 10 jobs for 2017.

Profession Annual
Median Salary*
Data Scientist $128,240 16%
Financial Advisor $89,160 30%
General & Operations Manager $97,730 7%
Home Health Aide $21,920 38%
Information Security Analyst $90,120 18%
Medical Services Manager $94,500 17%
Physical Therapist $84,020 34%
Registered Nurse $67,490 16%
Software Engineer $100,690 17%
Truck Driver $40,260 5%


To determine the most in-demand professions, evaluates Bureau of Labor Statistics (BLS) data on growth outlook, as well as industry and profession hiring trends over the last decade; trade statistics; university graduate employment data; and the database of listings to determine the factors driving hiring needs.

*Median Annual Salary and Projected Hiring Growth by 2024 are via the U.S. Bureau of Labor Statistics.

Source: Predictive Analytics Times


How IT can empower the enterprise with self-service analytics at scale

Business intelligence used to be a top-down affair which IT often approached in the same manner as traditional IT projects. The business makes a request of IT, IT logs a ticket, then fulfills the request following a waterfall methodology.

While this approach centralized data and promoted consistency, it sacrificed business agility. There was a significant lag between question and answer. And this delay led to lackluster adoption and low overall business impact.

Fast-forward to today and IT finds itself at a crossroad with self-service BI as the new normal that can no longer be ignored. The business demands the agility that comes with self-service to drive and improve business outcomes through data-driven decision-making.

This presents IT with an important choice. Either embrace the demand for self-service BI and enable the broader use and impact of analytics, or ignore the trend and continue producing lower-value enterprise reporting stifled by the limitations of traditional tools. IT professionals who are ready to serve as the catalyst will deliver far greater value than those who choose to ignore the real needs of their business users and analysts.

As organizations begin the transition from a traditional approach driven by IT to a self-service approach enabled by IT and led by the business, a new framework is required. This means that past decisions supporting the core foundational components of a BI program—people, process, and platform—must be revisited.

Prioritize people and their needs
A successful transition to self-service analytics begins with people. In a traditional BI model, people were often considered last after platform and process. IT often took the “if you build it, they will come” approach.

But even after they built it, most people did not come. That’s because there was little to no collaboration between the business users and IT during the process of building the solution after an upfront requirements-gathering phase.

Collaboration between the business and IT is critical to the success of the implementation. IT knows how to manage data and the business knows how to use the insights to drive business decisions. Early collaboration will not only lead to the deployment of a platform that meets the needs of the business but also drives adoption and impact of the platform overall.

Reimagine your process
Self-service analytics does not mean end users are allowed unfettered access to all data. It means they have the freedom to explore pertinent business data that is trusted, secure, and governed.

This is where process comes into play. This is the component that requires the most significant shift in traditional IT thinking. A successful modern BI program can deliver both IT control and end-user autonomy and agility. A well-established process is required to strike this delicate balance.

A waterfall-based process limits access to only a few specialists who are expected to meet the needs and answer the questions of the many. This approach often fails to deliver on the promise of BI—to deliver tangible value through improved decision-making with minimal time, effort, and cost.

A modern analytics solution requires new processes and newly-defined organizational roles and responsibilities to truly enable a collaborative self-service-based development process. IT and users must collaborate to jointly develop the rules of the road.

IT’s success is highlighted, and its value to the organization realized, when the business can realize significant value and benefit from investments in analytics and BI.

Implement a platform that IT loves and the business trusts
Since BI has been historically viewed as an IT initiative, it is not surprising that IT drove virtually every aspect of platform evaluation, selection, purchasing, implementation, deployment, development, and administration.

But with drastic changes required to modernize the people and process components of a BI and analytics program, IT must change the criteria for choosing the technology to meet these evolving requirements. Perhaps the most obvious change is that IT must intimately involve business users and analysts from across the organization.

A modern platform must address a wide range of needs as well as the increased pace of business and the exponential growth in data volume and complexity. Organizations need a platform that can adapt to an evolving data landscape and insulate users from increased complexity and change.

The most critical aspect is the ability to meet these diverse needs in an integrated and intuitive way—without having to introduce separate products or modules to execute specific tasks along the way.

Become a strategic partner to the business
As organizations shift their approach to analytics, IT leaders should seize the opportunity to redefine their role. Adopting a collaborative approach to truly support self-service is the key to changing the perception of IT from a producer to a strategic partner and enabler for the organization.


3 Content Marketing Trends That Will Dominate 2017

Assuming by now you have read an overwhelming number of “marketing predictions” and “industry trends” and have been disappointed – as I have – by the obviousness of those, we have separated the old from the new for you.

Because “content will rule” is not the hottest prophecy of the century anymore at seoplus+, we put together only the three most noteworthy developments based on industry stats, our own insight, and a little bit of gut feeling.

1. Condensed content will be the key to your mobile audience.
With Google’s search algorithm updates keeping us on our toes, the mobile first indexing shift, and the unstoppable rise of AMP, content marketers need to minimize stuffy word counts and channel their messaging into clear and compact storytelling. In the coming year, if your website is already a responsive one, there’s no need to be concerned. However, if your website’s mobile version has significantly less content than the desktop site or is incomplete, you can expect this to negatively affect your search rankings.

“The volume and richness of content that visitors want to see when first exploring a topic on desktop is very different than what they need to see when they’re on their smartphone ready and willing to make a purchase,” says our Content Marketing Specialist Amanda Murray. “Think of condensed content like your ‘elevator pitch’ that cuts directly to your core advantages. Know what your customer wants and needs to hear and eliminate anything that confuses or distracts from that.”

People consuming content on their mobile device versus a desktop need to be offered information in bite-sizes or you’ll lose their attention. It’s a matter of accessibility and to-the-point content for your on-the-go visitors. By switching to condensed content you’ll ultimately be more concise without sacrificing any meaning or intended impact of your message.


2. Visual content dominates on all platforms.
In 2016, over 50% of marketers used visuals (images, GIFs, videos, graphics) in between 91 and 100 percent of their published content. Compared to 2015, that’s an increase of over 100 percent and we can expect that trend to continue in 2017.

Due to the increase of mobile usage, which overtook web last year for the first time ever, visuals are not just a nice-to-have anymore. They’re a must marketers should give their audience if they want to keep them. With Facebook taking steps towards live streaming and a new “mid-roll” ad format, video marketing will experience a strong push in 2017.

Zuckerberg’s latest project allows publishers to insert ads into their clips after their audience has watched at least 20 seconds. By offering video producers 55 percent of the ads profits, companies who previously have been investing substantial amounts of money to grow their brand presence, will now get actual revenue out of it.

“When you’re browsing Facebook, be honest – you don’t always take the time to click away to an article, no matter how much the title draws you in. But you’ll always stop to watch a Tasty video, right? We’re all willing to cut a little more time for video, as long as it captures us right away,” says our COO Brock Murray.

“Businesses can learn a lesson from this. It’s all about presenting content in a way that makes the most sense for the audience in that moment. That might mean reading an in-depth case study on your work desktop, listening to a podcast on your commute home, or watching an explainer video on your phone. That’s just one reason video content is going to be huge in 2017.”
3. Buying influencers will overtake partnerships.
Influencer marketing was huge in 2016. With everyone wanting to jump on that train at the moment, we’ll see a movement from making the effort of establishing true partnerships with influencers to just flat-out buying them.
“People don’t buy goods and services. They buy relations, stories and magic.”
Seth Godin

However, marketers, don’t be fooled. It’s very easy to look like an influencer online – paying for fake followers can boost any user to the top of its game. When choosing an influencer in 2017, it will get increasingly difficult to distinguish the valuable ones from the fake. Our SEO expert, Lindsay Kavanagh, says “I always check followers and engagement of users before I even consider hiring them. Moving forward I foresee better tracking and metrics on social accounts so brands can determine who the authentic, successful influencers are. Once advanced metrics are available to see if an influencers followers are legitimate, I expect tracking becoming more advanced – and better options to shop directly from posts.”

For instance – wouldn’t it be great for users to shop directly from social posts? Then influencer campaigns can be tracked better. When influencers pitch themselves, they’ll be able to show past ROI or conversion rate. You can see the success of a landing page or ad campaign, influencers will be able to display metrics. This will separate the real deal from the rest.

Even though we clearly can’t guarantee that any of our marketing predictions are spot on, the one tip that always works for digital marketers is: no matter what specific change is headed your way, the better you become at adapting quickly and effectively the better you’ll be able to leverage those changes as opportunities.


5 Necessities of an Effective Closed-Loop Customer Feedback Program


If you’ve been keeping up with this series, you’re familiar with the idea of gathering continuous customer feedback. But it’s important to note the changing environment for how that feedback is handled.

Traditionally, customer satisfaction surveys have focused on collecting aggregate data. In the world of market research, this approach makes sense. It’s statistically accurate, high-level, and shows trending data—all great things for market researchers. But as customers have become more aware and their expectations have risen, this “open-loop” system falls short. Customers expect that if they take the time to provide personal feedback, then someone should take the time to provide personal follow-up.

A closed-loop customer feedback management system gives institutions the tools they need to take this personal follow-up and “close the loop” on each piece of customer feedback. Ideally, your system should work on two different but complementary levels:
1.Support follow-up with individual customers based upon their feedback (as depicted by the smaller circle in the diagram below), and
2.Deliver insights from aggregate feedback to drive action that will benefit all or a sub-set of the bank’s customers (as depicted by the larger circle):

Five Requirements for Effective Follow-up with Individual Customers:
These qualities should be present in any effective customer feedback management solution as it relates to following up with individual customers:

1.  Non-Anonymous
Patricia Smith, President of Wealth Management at First Interstate Bank, explains the difference between our ongoing Voice of the Customer program and prior efforts the bank had undertaken to listen to customers:

“The biggest component in our relationship with PeopleMetrics is that the surveys aren’t anonymous, and it is a continuous improvement process—meaning we are able to attach the responses, then look at the customer relationships and better understand what’s working well and where our areas of opportunity lie.”

When you’re trying to close the loop, you do not have to protect the identity of your customers who give you feedback. Two-way communication and dialogue is fundamental to a solid and healthy relationship with them.

2.  Action Triggers Built into Survey Design
Don’t fall into the trap of asking your customers about all the elements of the experience that only matter to the bank’s leadership, managers and employees. Did our staff greet you by name? Did you have to wait in line longer than 2 minutes? Were you able to easily find our mobile banking app?

If you impose such questions, but they aren’t tied to issues or important to your customers, you can do little with the insights. Instead, structure an extremely short survey that elicits what’s important to the customer, what happened to them that warrants sharing, provides room to elaborate in their own words, and points bank employees to what they should do next.

The survey questions used in a closed-loop feedback program should clearly point to the best next action the bank should take.

3.  Automated Notifications
It is possible now to trigger automatic notifications (e.g., email, SMS) to bank representatives that alert the employee to a piece of customer feedback requiring action.

These automatic notifications are a core element in a closed-loop customer feedback system. It helps branch managers hear from customers, quickly determine the context of feedback, and respond accordingly.

Common automatic notifications or Action Alerts include:
•Recover Notifications or Course Correct Alerts, when an unresolved problem is dissatisfying the customer;
•Improve or Innovate Alerts, when a customer has a suggestion for how the bank or the channel can get better; and
•Recognize or High Five Alerts, when an employee has gone above and beyond and is deserving of recognition.

Many of our banking clients are also adopting Grow or Opportunity Alerts, which help employees identify opportunities for cross-selling and up-selling to their clients. This soft lead generation tool is helping to shift the culture at one of our client’s banks from one that is purely about service to one where sales, in the name of helping the customer, is an essential and acceptable part of the experience.

4.  Case Management Tools
Asking managers to take action on real-time customer feedback requires work on their end. Strong closed-loop feedback programs provide tools to help managers delegate follow-up actions and track the progress of open cases. Case management tools make it easy to create transparency around the customer experience and ensure that no piece of feedback gets lost or overlooked.

5. Root Cause Tracking
As your managers or other employees close the loop with customers, they should document the root cause of the issue within the case management system. Perhaps customers are complaining about a lack of knowledge of staff at a given branch, and you know that branch has had high turnover and many new hires in recent months. Or perhaps numerous comments related to unexpected branch closings reveal a poor communications plan. By tracking root causes, you will be able to identify areas to improve your operations and eliminate the issues moving forward.


Google’s Most Interesting: Apps, Projects, and Services You Must Know About


Google has come a long way since its humble beginnings as a web 1.0 search engine. Not only has Google dominated the internet search market and its name become a verb, but Google also has a vast array of other products, services, and projects worth using.

Unfortunately, they’re pretty bad at communicating what those works are.

Following the creation of Alphabet in 2015, Google separated from some of their more outlandish projects and have turned their focus towards more consumer projects, many of which are exciting and starting to bear fruit — the kinds of projects that you may want to start using yourself!

Google’s Software Projects

Android, Chrome OS, and Fuchsia
Google currently maintains two operating systems: the smartphone platform Android and the laptop platform Chrome OS, which runs on Google’s affordable Chromebook laptops.

Since 86 percent of smartphones worldwide run Google’s mobile OS, there’s a good chance that you already own an Android device. Although its main base is open source, Google puts a lot of development into Android to make it the polished end product it is. This is similar to their development of the open source web browser Chromium, released by Google as Chrome.

Chrome OS is similar to the Chrome web browser but with additional offline features. By running such a minimal OS, Chromebooks get great battery life and are usually priced far under most Windows laptops. Chromebooks have been well received largely because most users now syncs multiple on-the-go devices with cloud-based services. When everything runs on the internet, it seems unnecessary to run a full-blown operating system anymore.

There have been rumors of Google wanting to merge their two OSes, bringing a “best of both worlds” approach to mobile computing. To mark the start of this transition, in 2016 they revealed a project that would allow Android apps to run natively on Chrome OS. The beta build was rolled out to select Chromebooks and is currently scheduled for full release in late 2017. Google even recently confirmed that every Chromebook released in 2017 will support Android apps.

Parallel to this whole idea of bringing Chrome OS and Android together is an entirely new project called Fuchsia. The project was “announced” when the development team made the code available on Github. Currently there is very little information on what Fuchsia will actually be for.

What we do know is that it is not based on Linux — as Android and Chrome OS are — which makes it a radical departure for the company. It also appears to be designed to run on lightweight, minimal hardware specifications, suggesting that it could be used for IoT products. However, as Android Things is also in development (more on that below), it isn’t clear what the overlap would be here.

Android Pay

At I/O 2015 Google announced the launch of their mobile contactless payment service Android Pay. The seeds of the service were planted in 2011 when the company launched its online payment predecessor Google Wallet. With Android Pay, they have brought contactless payments into the mobile age.

Contactless is meant to simplify payments, making it quicker, easier, and more secure for you to pay at the checkout. Android Pay takes this idea and applies it to your smartphone. Android phones have supported Near Field Communication (NFC) for some time, and the contactless payment systems found in your favorite shopping destinations also use NFC to accept contactless payments.

Setting up Android Pay is as simple as downloading the app on an NFC-enabled Android phone and adding a credit or debit card from a supported bank. Using the Android Pay app, you can also keep track of all your spending and manage your loyalty cards. As an incentive to use the service, Google has been known to add chances to win prizes just by making payments with Android Pay, as they did in the U.K. across Christmas 2016.

Android Things

The Internet of Things is a movement to add internet connectivity to almost anything in your home. It is a growing market with devices ranging from fitness trackers all the way to internet-connected Lego cities. But all of these products need an operating system in order to interact on the cloud.

Google’s attempt to create an IoT OS dates back to Project Brillo, unveiled at I/O 2015. Brillo had a name change and relaunched in December 2016 as Android Things. Along with this relaunch came the first developer preview of the stripped down Android OS.

Modifying the Android OS to fit other devices is nothing new for Google, as they have played this game before with Android Auto and Android TV. However, with millions of IoT devices expected to enter homes over the coming years, the battle for IoT OS is only just beginning. Although Android Things is still in early-stage development, Android has a large developer community so expect to see some early demos and prototypes in the near future.


Machine learning has become quite a buzzword, being found in almost every type of software you can imagine these days. However, Google has been using machine learning since before it was cool to categorize photos by faces and features.

Machine learning neural networks try to replicate the way the brain works, and as a result are quite complicated. Just like the mind, they need information in order to learn. To train a neural network, you need to feed it training data — for example, pictures of cats — so it can learn to correctly identify cats.

However, the network can also run in reverse, adding features it has been trained to look for into images. This ends up creating some very psychedelic  images that often resemble modern art more than photography. Google named this project DeepDream and open sourced the code, placing it on Github. From there, a number of developers have created websites and apps that allow you to create your own slightly terrifying artwork.

Gmail and Inbox

Google’s web mail service, Gmail, is one of the largest email providers in the world. Given its prominence, it’s a surprise to learn that it originally started life as a side project of a Google employee as part of the company’s 20% Innovation Time Off initiative.

Gmail became one of the leading email providers due to its speed, ease of use, and useful features. It now has over one billion users and continues to grow as a result of the success of the Gmail Android app.

Majorly changing or updating the service could be a disaster for the company as they try to prevent breaking (or significantly altering) the email client and upsetting their users. Google’s solution to this dilemma was to create the new email service Inbox. While Inbox uses the same back-end data as Gmail — meaning that you can switch easily between the two services — it aims to change the way you interact with email.

Most of us receive emails and either leave them in our inboxes until we deal with them or forget about them entirely. Inbox wants to help you turn your inbox until the ultimate to-do list. This point is emphasized when you are finished with an email you can now mark as “Done” rather than “Archive”.

Inbox also brings a host of personal assistant style features, including reminders, the ability to snooze emails until another date or location, and even save links for later. Inbox also bundles together emails of similar types like Promotions or Updates for you to easily move through your email list in a quest for the elusive inbox-zero.

Currently there are no signs that Google plans to fold Gmail in with Inbox or even merge the features. There is always the possibility that they will run both services in parallel since they are both built on the same back-end. The alternative would be to test features in Inbox and potentially move them over to Gmail when they are deemed ready for prime time.

Google Assistant

In the age of Siri and Alexa, it looks as though voice-controlled personal assistants will play a large part in our lives. Of course, Google already has an assistant known as Google Now — one that can access all of the information you store with Google to give you travel updates, personalized news stories, and other information.

Google’s voice search has also been widely regarded as powerful and accurate. However, up until very recently, there was no way to connect the personalization of Google Now with a voice control that could understand conversations.

Google Assistant brings together those two products, creating a smart personal assistant filled with information pulled from your Google data. The biggest step forward for Assistant is the use of natural language processing, which means you can have a normal conversation with Assistant rather than having to talk to it like a computer and issuing specific commands.

Google has also announced Actions on Google which allows developers to add commands to Google Assistant. This can either be a direct command (like booking a specific flight) or ways to start up a contextual conversation. Google’s aim is to let you talk to Assistant and do everything you want without having to install apps or authorize new services.

Assistant is currently limited to Google Home, the Google Pixel phones, as well as their messaging app Allo. However, you should expect to see it coming to more products in the near future, especially as Actions on Google allow developers to embed Assistant into other hardware.


One of the most compelling services by Apple for iOS users is iMessage — a service that supplements your regular SMS experience by adding extra features like video calling and group chats. If you ask an iOS user why they wouldn’t move to Android, it will likely have something to do with iMessage.

Of course, as Apple’s major competitor in the smartphone market, Google is keen to create an iMessage alternative for Android. However, Apple controls all aspects of the iOS experience, from the hardware down to which carriers can support iPhones. Android is far more fragmented with many varieties of phones running on multiple different networks, so how would an iMessage competitor work?

The answer lies in Rich Communication Services (RCS), a replacement to the rather tired SMS. RCS brings with it the ability to send files between devices, make video calls, send large photos, and even support group messaging.

Following their purchase of Jibe Mobile in 2015, Google launched the Jibe Platform as a way to implement RCS for Android. Google Jibe could cause a dramatic shift in the way we message on our smartphones, potentially rendering rivals (like WhatsApp) redundant.

Messaging (Hangouts/Allo/Duo)

Since Google launched their cross platform messaging app Hangouts, they’ve put a substantial amount of effort into its development adding voice and video calls and SMS support on Android. Until recently it was seen as Google’s alternative to Skype, WhatsApp, and potentially iMessage. Its user base grew, helped in large part by it being pre-installed on the majority of Android phones and integrated with Google accounts.

That all changed when Google launched standalone messaging apps Allo and Duo for Android following I/O 2016. Hangouts has pivoted to a professional messaging service for businesses while the messaging app Allo and its video-calling companion Duo have become the company’s consumer-facing messaging platforms.

Some might argue that spinning off Android’s messaging to two separate apps, which need to be downloaded by all of your friends and family to be useful, was a poor decision. However, the performance of both apps is much improved over the often buggy performance of Hangouts.

Duo is a one-to-one video-calling app that allows you to chat with another who also has Duo. While the performance and stability of the app exceeds that of Hangouts, it is missing the ability to start a group call.

Allo is much like other popular messaging apps (e.g. WhatsApp and Telegram) where you can send and receive messages, videos, and audio from your contacts who also use the app. However, its biggest selling point is the integration of Google Assistant. You can message Assistant either in a standalone chat just for your own use or integrate it into your Allo chats by tagging @google.

Google Photos

If DeepDream is the trippy, experimental side of machine learning, then Google Photos is where it finds its purpose. Google Photos started off as a replacement for the stock Android Photos app but quickly morphed into full-blown photo management software. While the option is still there to just view the images that are on your phone, its main selling point is the storage and categorization of images.

You can upload your entire photo collection from your phone or computer to Google’s servers for free (as long as you don’t mind the resolution being limited to 16 MP). The whole collection is then viewable through the Photos app or the website.

Machine learning and Google’s search prowess combine to categorize your photos and make them searchable. The machine learning allows Photos to group together images of specific people, as well as by feature like beach or sunset.

Photos also houses an assistant, but this is an entirely different Assistant to the one on Google Home, Pixel, and Allo. The Photos assistant looks at your upload photos, then groups them together by location. It can also create collages and stylized photos which you can save to your library. It even has a Timehop-style feature where you can be notified of what you were doing in years past.

Google’s Hardware Projects

Android Wear

While the now-defunct Pebble was among the first to bring wearables to the tech market, Android Wear was the first attempt from one of the industry’s big players. Android Wear is a companion smartwatch to your Android or iOS phone and runs on a modified version of Android.

The watches, made by partners like LG, Motorola, and Samsung, all look stylistically different, with some featuring additional sensors, like heart rate monitors, but Google is very much in control of the OS.

The watches integrate with Google Now to provide personalized updates alongside your notifications and call alerts. As Android Wear devices pair via Bluetooth, they can also be used to store and playback music without the need to be connected to the controlling phone.

Just like the full-blown OS, Android Wear can run apps from the Play Store to add features to the watch. While smartwatches in general, including the Apple Watch, have largely failed to capture the imagination of the general public thus far, they have been embraced by one group: fitness enthusiasts. One of the places Android Wear excels is in fitness tracking, either using Google Fit or by downloading third-party apps.

Google is currently in the process of preparing the delayed release of Android Wear 2.0. While there have been some small updates since the original release, 2.0 will bring a revamped Material Design UI, on-device Google Play Store for Android Wear apps, a built-in keyboard, and handwriting recognition. One of the most anticipated additions is cellular support to break the bond between phone and watch.

Google Home

Amazon’s Echo smart home device practically created a new market for internet-connected devices. The ability to talk to your virtual personal assistant and have it respond to you and use the device’s in-built speaker to play music was game changing. It was only natural that Google would want to build a competing device, which they did with Google Home.

Launched at I/O 2016 alongside Google Wifi, it marked Google’s first major foray into hardware manufacturing. Google Home’s design aesthetic is similar to that of Google Wifi and Google OnHub. It hopes to become a celebrated centerpiece of the home rather than an ugly piece of technology, and to highlight this, Google created customizable bases which can be changed at will.

Google Home has a built-in speaker which can be used for music playback from services including Google Play Music, YouTube Music, Spotify, TuneIn, Pandora, and iHeartRadio. The main attraction of Google Home though is its integration of Assistant. Because of the implementation of natural language processing, you can have an actual conversation with Google Home, which gives it a major advantage over Amazon Echo. Taking a leaf out of Sonos’ book, by placing multiple Google Homes around the house, you can even group them together to enjoy the same music in every room.


Google has been producing phones since 2010 when they launched the first device in the Nexus line. These phones were aimed at developers as a showcase for what a stock Android experience could be like. In order to keep the cost of the devices down, they typically had mid-range hardware — nothing that would be classed as flagship material. In actuality, Google never made these phones, instead choosing to work closely with the design teams at Samsung, Motorola, LG, and others.

Since Google turned its focus more towards the consumer market, it makes sense that they would try their hand at producing the first Google phone. The Pixel, alongside its larger sibling the Pixel XL, was launched in October 2016 as a direct iPhone competitor.

At first glance it’s easy to see how Google has taken aim at the design of the iPhone with the Pixel’s rounded edges, metallic body, and high-resolution screen. Google also decided to borrow the iPhone’s premium pricing, retailing at $649 for the base model.

While it’s expected for a modern flagship phone to have impressive hardware specs, which the Pixel does, it’s the software that really sets it apart. The Pixel comes with an exclusive launcher that has Google Assistant built in at its core.

The Pixel and Pixel XL have only been out for a few months as of this writing, but reviews have been very positive overall (we gave it an 8/10 rating). While the heavy price tag is hard to overcome, the Pixel is a shining example of what a fantastic OS Android can be.

Cardboard and Daydream (Virtual Reality)

Google Cardboard is a virtual reality (VR) platform for smartphones that can be experienced using a foldable cardboard headset that you place your phone into. Google published the designs for the headsets, allowing other companies to customize and manufacture their own versions, for a price typically below $30.

Alongside the headset, Google also developed a software development kit (SDK) for creating VR apps compatible with Cardboard. Initially Cardboard only supported Android, but it has since expanded to include iOS. The low cost coupled with wide-ranging accessibility means that Cardboard, which was originally a “20% project”, is now a huge hit.

Building on Cardboard’s success, Google launched the Daydream VR platform at I/O 2016. Daydream is built directly into Android 7.1 allowing for a more immersive and seamless experience. Instead of having to open a VR app from your phone as with Cardboard, the new Daydream view is a virtual reality world which houses all the new VR experiences built for Daydream.

The software isn’t the only upgrade for Daydream — the headset got a makeover too. A new premium headset was developed, created from a lightweight cloth to make it more comfortable to wear. New capacitive buttons replaced the manual push button of the Cardboard headset, and an NFC chip simplified the connection process.

Google also opted to add controllers to Daydream, meaning that you can interact with the virtual world by either clicking or waiving the controller bringing Google’s VR platform in line with competitors like Samsung’s Gear VR.

Google Wifi and OnHub

It’s common knowledge that many home routers are unreliable or just flat-out don’t work. Poor connection and Wi-Fi dead zones have plagued us all for many years, so in a lot of ways routers have stayed firmly planted circa 2007. The hardware has gotten better, but for most routers the software and management have stagnated.

Google decided that this was a problem worth tackling and launched the OnHub, a modern take on the humble router. Its elegant design immediately made it stand out, edging itself closer to decorative homeware than routers of the past.

Wireless coverage was improved too by rethinking the positioning of the Wi-Fi antennas. There was even an app to manage the OnHub, replacing the often clunky web interfaces of most routers.

And at I/O 2016, they proudly announced Google Wifi. This move initially stung owners of the OnHub who were still waiting on promised feature-adding upgrades. However, instead of replacing it, Google Wifi lives within the same ecosystem as the OnHub.

Where the OnHub is designed as a single router for your home, Google Wifi is a smaller unit designed to be placed across the home to create a mesh network with other Google Wifi units. The mesh network improves coverage, even in those hard-to-reach spots.

Google’s Service Projects

Project Fi

Since Google provides the OS that powers the majority of smartphones around the world, it was only a matter of time before they made the jump to mobile carrier to complete the package.

Project Fi, launched in 2015, is a virtual mobile network operated by Google. The service currently only operates within the United States, although there are partners that enable roaming in up to 135 different countries at the same domestic prices.

Project Fi runs on the infrastructure of four different operators: Sprint, T-Mobile, US Cellular, and Three. This lends itself to Project Fi’s main advantage over a traditional carrier, which is the ability to switch between 4G networks depending on signal strength and speed.

Consider what happens when you’re on a Wi-Fi call but get disconnected: Fi seamlessly merges your call onto its cellular network. Currently only Google’s Pixel, Nexus 5X, and Nexus 6P phones are set up to work with Project Fi as they were specifically designed with a mobile radio to support network switching.

Despite being a virtual network, you can move your current cell number over to your new Project Fi plan. Another benefit of Fi is that there is no annual commitment. Gone are the 24-month contracts and in its place comes a rolling monthly contract.

For just $20/month on Fi Basic, you receive unlimited domestic talk and text, unlimited international texts, Wi-Fi tethering to use your phone as a hotspot, and access cellular coverage in 135+ countries and destinations. Read more about our own experience using Project Fi.


The United States is geographically huge and serviced by only a few telecom companies capable of high-speed fiber broadband internet. And since Google’s entire business is online, it’s in their best interest to get as many people connected to the internet as possible.

Google decided to overcome this issue by becoming a broadband provider. In 2010 Google began what was seen at the time as an experiment to bring fiber broadband and cable television to Kansas. After it proved to be a resounding success, Google turned this experimental venture into a legitimate business.

Google Fiber is now available in various areas around the U.S. with more currently in development. If you live in a Fiber area, then you can get up to 1,000 Mb/s broadband speeds. The Fiber service also carries your cable television service, which includes more than 220 channels.

Included in your cable package is a home DVR capable of recording up to eight HD shows simultaneously. You can even access your recordings on the go with Google’s Fiber app. As this is Google, they also equip your Fiber cable with Smart Search. This lets you search your TV listings, and even filter by actor.

If all those feature weren’t enough to tip you in Fiber’s favor, then perhaps the complementary 1 TB of Google Drive storage will. Or what about the unlimited local and nationwide calling from your Fiber home phone? It’s hard to deny Google Fiber as a compelling service.

What’s on the Horizon for Google?

This list only scratched the surface of what Google is working on. Alongside some other consumer projects, Google is putting a lot of effort into development tools and cloud services.

And that’s not to mention the projects of Google’s now-parent company Alphabet! All things considered, 2017 could be a very exciting year for the Mountain View tech giant — and for you, depending on how many of these projects you start using yourself.