For SaaS Industry Difficult for Stand as no. 1 in Competitive Market

For SaaS Industry Difficult for Stand as no. 1 in Competitive Market

   July 12, 2017       prabhakardalvi

---

There is no longer any doubt that Software-as-a-Service (SaaS) solutions have become the preferred method for organizations of all sizes to acquire business applications to satisfy their escalating customer and end-user demands while keeping pace with intensifying competitive pressures.

But, the SaaS industry and its growing legion of enterprise customers are falling into the same software development and implementation traps that derailed the previous generation of on-premise, perpetual license ISVs who the leading SaaS vendors successfully disrupted over the past decade.

Gartner latest forecasts estimate that SaaS revenue worldwide will increase 20.1% in 2017, and jump from $46.3 billion at the end of this year to $75.7 billion by yearend 2020. Gartner says, “…more than 50 percent of new 2017 large-enterprise North American application adoptions will be composed of SaaS or other forms of cloud-based solutions."

Corporate software acquisition preferences and policies have dramatically shifted away from traditional, on-premise legacy applications to a new generation of on-demand, Cloud-based alternatives for a variety of reasons. SaaS adoption has gained momentum as a widening array of organizations have taken advantage of the lower upfront costs and faster time-to-value of many of today’s SaaS solutions.

Some organizations won't spend more on annual SaaS subscriptions because they are stuck on a previous version of the SaaS solution and are no longer able to take full advantage of the latest features - Jeffrey Kaplan in the founder and Managing Director of THINKstrategies

Although many SaaS deployments have taken longer than anticipated and entail specialized software development and systems integration skills to connect the new applications with legacy databases, most organizations have still been pleased with the operational efficiencies and additional functional capabilities delivered by the SaaS solutions.

As a result, many organizations are expanding their SaaS subscriptions to support additional workers, and adopting additional SaaS solutions to redesign more of their business processes. However, these organizations are often finding that their SaaS implementations are getting a lot more complicated.

- Advertisement -

There are two primary reasons SaaS deployments become harder rather than easier over time.

First, most organizations are customizing the SaaS solutions so they will support their existing operations.

And second, the SaaS vendors are more than happy to let their customers do as much customization work as they like because it locks the customers into the SaaS vendors’ solutions.

In fact, many SaaS vendors are increasingly willing to let their enterprise customers sit on an old instance of their SaaS solution to accommodate all their customizations. However, this tactic is preventing these organizations from capitalizing on the latest advancements in their SaaS solutions.

SaaS wasn’t supposed to work this way.

The pioneers in the SaaS market, such as Salesforce.com, have always promoted the virtues of a single version of their applications being able to address the common needs of their customers. But they have recognized that there are industry-specific requirements and other operational issues facing many organizations that demand specialized skills and SaaS products. As a consequence, today’s SaaS product portfolios are becoming as complicated as the previous generation of perpetual license software applications.

Third-party software development and systems integration firms are prospering in this environment as they capitalize on the rapidly growing market for SaaS customization projects. It is no wonder that the biggest booths at the front of Salesforce.com’s Dreamforce conference show floor are populated with the largest professional services firms, such as Capgemini and Deloitte.

In fact, the market for SaaS/Cloud integration services has grown so rapidly that nearly all of the most prominent Cloud integrators founded over the past decade have been acquired by the biggest professional services companies in the world. Over the past two years, Accenture gobbled up Cloud Sherpas, IBM bought Bluewolf, and Appirio was acquired by Wipro.

Although everyone expects the rapidly evolving assortment of artificial intelligence (AI) and machine learning (ML) capabilities to automate various aspects of software development, deployment and support, the reality is that most organizations need a new set of experts to help them evaluate, implement and administer these new solutions in their environments. In response, Salesforce.com and a handful of venture firms are establishing dedicated investment funds to support the next generation of AI/ML oriented systems integrators.

Even with the promise of AI and ML on the horizon, I’m now hearing from a growing number of organizations that they don’t want to spend more on annual SaaS subscriptions because they are stuck on a previous version of the SaaS solution and no longer able to take full advantage of the latest features.

If this trend continues, the SaaS industry could face a significant speedbump in the future and independent systems integrators will be the only winners in this environment.

Note : Any blog OR content suggestion you have , please mail me on prabhakara.dalvi@gmail.com

Read More

Augmented Intelligence combining human and AI to change behavior

Augmented Intelligence combining human and AI to change behavior

   April 18, 2017       prabhakardalvi

---

Dr. Cameron Sepah

Featured in: Big Data, Big Ideas & Innovation, Entrepreneurship, Healthcare, Technology

Artificial Intelligence, or AI, is the most promising and overhyped technology of our times. AI techniques such as deep learning have allowed computers to match or even beat world experts at games like chess and Go, and even board-certified doctors at diagnosing diabetic retinopathy and skin cancer.

While AI is great for recognizing patterns in puzzles and pictures, it is much harder for AI to change the behavior of people, in all their fascinating and frustrating complexity. In an earlier post on the Keystone Habit, I introduced the concept of using goals and habit change loops for personal development. Now let’s explore how to combine these loops for product development, in order to design a system to help someone change.

Here, I argue that human intelligence is better applied to helping people form goals, while AI is better applied to helping people form habits. This is not to say that an effective fully-automated AI system could not be built. But if you already have human and software resources, they can be synergistically combined to create an “Augmented Intelligence” system for behavior change.

The Sepah Behavior Change Model:

While there are a lot of behavior change models out there, I created a new model to integrate two important concepts—the goal formation loop on the left, and the habit formation loop on the right—into a unified loop that allows for continual improvement. Let’s walk through an example of how this works:

Plan the Goal:

All behavior begins with intention. Let’s say my intention is to exercise more often. While I may want to start working out every day, I decide to work with a personal trainer who takes into account my current routine to help me set a S.M.A.R.T. goal, that is Specific, Measurable, Agreed Upon, Realistic, and Time-Bound. For example, since he knows my gym is conveniently located two blocks from my work, and I’m more motivated to exercise in the morning because I work long hours, we mutually set a goal of going to the gym next to my work at 7:30AM, to lift weights for 30 minutes on Monday, Wednesday, and Friday, and to check-in at week’s end.

Theoretically, an AI-based system could also ask me a series of questions about my routine and preferences, then creates a reasonably-tailored S.M.A.R.T. goal for me. But these systems often run into problems adjusting goals over time, as we’ll see later in the loop, which is why humans are better suited to plan and revise goals.

Act on the Goal:

Next comes an attempt to act on the goal. I’m motivated on Monday morning, and I successfully go to the gym for 30 minutes before work. Great! Action is the centerpiece of the model for good reason, since goals are merely dreams without action. It’s worth noting that inaction, such as procrastination, is counterintuitively an action itself (what the Taoists call ‘Wu Wei’, the action of inaction). So whether you do or you don’t, you’re still acting. The difference is direct action takes you towards your goals and values, while avoidance takes you away from them.

Reward the Habit:

Here we cross into the habit loop: actions only become habits when they are repeatedly rewarded over time, even if the rewards are occasional. These can be intrinsic rewards (e.g. the mood boost from the workout) or extrinsic rewards (e.g. my personal trainer telling me I did a good job). While social reinforcement is an incredibly powerful motivator—it fueled the rise of phenomena like Crossfit—AI can reward habits equally well.

Gamification, which is the application of behavioral principles to game mechanics, is the best example of AI reinforcement. It successfully gets people to play video games for hours on end using tokens such as points and badges to designate accomplishment and skill development. This would seemingly pale in comparison to the sense of satisfaction that comes from a real person giving you a heartfelt high-five after a workout. But the high-five after every single workout can become repetitive, while gamified systems use variable schedules, quantities, and types of rewards (like how slot machines randomly dole out different jackpots) to prevent habituation and continually provide the addictive dopamine hit that powerful rewards bring.

AI is also superior to human reinforcement on an economic basis. While a personal trainer is motivational, they are usually not as cost-effective as an AI system emailing me a coupon for free protein shake at my gym when I achieve 3 workouts a week, or notifying me that I am among the top 20 most active gym-goers via an automated text message. Applications like Pact, DietBet, and Stikk are examples of companies that blend social and behavioral economic rewards to get people to go to the gym more often.

Remind the Habit:

Habits not only need to be rewarded, but they need a reminder (also known as a cue or trigger) to initiate the behavior regularly. Since I am out of town for work on Tuesday, I call the hotel front desk and set a wake-up call on Wednesday morning to remind me to get up for my scheduled workout. While that is an effective reminder, it is not a good use of human resources, which is why most hotels have automated wake-up call systems (or most people just use their smart phone’s alarm).

Though it seems obvious, Omada ran a randomized controlled trial to validate this hypothesis. We found an automated email performs just as well as a human coach in reminding a participant to weigh-in on our connected scale. Given that reminders are quick and easy, outsourcing this to AI is an effective move.

Act on Your Goal (Again):

For an automatic habit to form, the action must be repeated. My alarm goes off on Friday, but I am so sleep-deprived and tired that I decide to hit the snooze button. I text my trainer to cancel our session last minute, and fail to go for the third scheduled workout that week. Thus, I unfortunately fall short of my initial goal.

Reflect on the Goal:

Here we cross back into the goal loop by reflecting on the week and how well I’ve executed against my goal. This step is missing or glossed over in most versions of habit loops that you’ll see in books, but is critical if you want to continually iterate and improve upon your goals. In a previous article, I argued that reflection is the Keystone Habit when it comes to personal development, and articulated a concrete system to achieve your goals. Without reflection, we are doomed to repeat mistakes, and as I always tell my patients:

“a mistake repeated more than once is a decision.”

My trainer calls me over the weekend to reflect on our first week together. I share that I used my initial motivation to successfully go to the gym on Monday. I then used an alarm to remind me to go again on Wednesday, and was motivated by remembering the sense of accomplishment I felt after the first workout.

However, by the time Friday rolled around, we identified I was sore from starting a new exercise routine and the accumulated sleep deprivation from a long work week hindered my physical recovery and sapped my motivation. Here we see the power of human intelligence. While there were many factors that contributed to my skipping the gym on Friday, a quick conversation with my trainer identified the two major ones: muscle soreness and lowered motivation from sleep deprivation. An AI system has a much harder time identifying and isolating variables, especially psychological ones.

Plan the Goal (Again):

Based on the information gathered, my trainer and I work together to plan a revised goal for next week. He recommends I go to the gym next Monday and Wednesday at 7:30AM, but knowing I’m likely to be sleep-deprived and less motivated on Friday, recommends I go at 6PM just to warm up and stretch (thus reinforcing the habit of going regularly, while making the workout easy enough for my level of motivation). If that doesn’t work, we will switch the third weekly workout to Saturdays when I’m feeling more recovered and motivated.

Human intelligence shines at iterating on goals. Good trainers and therapists accurately identify individual strengths and barriers and leverage them to creatively optimize goals towards success. People’s physical and psychological states also vary tremendously on a day-to-day basis. Humans are better at picking up on these through our intuitive ability to read body language and facial expressions. Once a new goal is planned, the behavior change loop is executed repeatedly until the action becomes an automatic habit.

The Future of Augmented Intelligence

While AI has tremendous promise (and I advise AI startups in Silicon Valley because I believe they will be transformative), it’s currently better suited for specific intelligence, rather than general intelligence. When it comes to behavior change, AI is a clinically effective and cost-effective tool for habit formation by automating reminders for behaviors and providing variable rewards. But when it comes to goal formation, human intelligence (particularly that of an effective manager, trainer, or psychologist) is currently better leveraged to help people with goal planning and reflection to continually improve.

As a result, I believe the immediate future holds tremendous promise for hybrid systems, what I call “Augmented Intelligence,” which best combines human and artificial intelligence to change human behavior. Rather than replacing human coaching altogether, AI can support coaches by automating the easier tasks of reminding and rewarding habits. Thus freeing time to focus on what’s difficult and meaningful: helping others find and achieve their dreams.

source :- https://goo.gl/ZFxhI3

Note : Any blog OR content suggestion you have , please mail me on prabhakara.dalvi@gmail.com

Read More