Get started Bring yourself up to speed with our introductory content.

Sort out your analytics game before graduating to an AI solution

Data analytics has come up as a strong business differentiator and is helping companies beat competition. Across the verticals, companies have graduated from early adoption of diagnostic analytics to prescriptive analytics to have deeper insights. Another term that is bandied around unabashedly as the next frontier is artificial intelligence (AI). There is so much buzz about AI, it’s difficult to find someone unimpressed by it. Irrespective of their exact understanding of AI and how it can help them, people view AI as a panacea. Something that can help them pull the rabbit out of the hat and be one up on the competition.

It’s easy to be impressed by AI capabilities and get carried away. It is, however, very important to understand that both analytics and AI are important. It is simply not possible to straightaway leap to an AI solution without going through the rigors of establishing a robust data analytics practice. Analytics is the logical, formative, and imperative step in reaping the real benefits of an AI solution. A hasty, unplanned jump to the AI bandwagon minus the analytics background is a certain recipe for disaster. Businesses might end up burning a huge financial hole, big loss of confidence in AI and loss of credibility.

Why pursue analytics before AI

But why is good analytics a precondition to an AI solution? Let’s take a look.

To grow, every business depends on making the right decisions at the right time. And to do this, companies need data — lots of it. It’s not just the quantity of data, but the quality and speed of data as well. That’s where analytics figure in the scheme of things. Even a company not using analytics does use data. It may be in the form of the spreadsheets, emails, detailed notes, manuals and so on. Using such data is cumbersome, time-consuming and fraught with risk. Using analytics for crunching this data helps companies understand the various dimensions of their business better.

The analytics journey

The first step in the analytics journey is descriptive analytics, which is merely determining and stating what happened unambiguously. The next milestone is diagnostics analytics, which deals with determining the reasons behind what happened. Although pretty basic, these two types of analytics capabilities are still considered absolutely essential. They play an important role in structuring and standardizing the way data is used. In addition, they prepare the groundwork for further stages of analytics.

It’s only when a company reaches the third milestone, predictive analytics, that things start getting really exciting. And the AI angle starts to play as well. Predictive analytics, based on the first two types of analytics, tries to predict what’s going to happen in the future. Further into the journey, prescriptive analytics focuses on what the company should do to benefit from the future happenings. Analytics tries to predict and benefit based on the conclusions derived from crunching the data. It is, however, the AI that is going to provide deeper insights and propose better actionable recommendations.

There is a further milestone in the analytics journey called cognitive analysis. It’s where the businesses cause something to happen. So the business no longer just hopes for a correct predictive analysis, but making something happen is a certainty. And this final milestone is massively aided by an AI solution. With the built-in intelligence and ability to look at data, derivations, exceptions and conclusions objectively, AI can greatly aid cognitive analytics.

Now imagine a company skipping analytics altogether and straightaway trying to adopt an AI solution. To say the least, the move is fraught with danger. It is like starting to learn trigonometry without knowing what are angles and how are they measured.

Other potential pitfalls

Both analytics and AI are data driven. And it is of vital importance that the data is correct, timely and consistent. Having a good analytics practice already in place ensures that this is already taken care of. It alleviates any worries about bringing in an AI solution from a data perspective. This is huge benefit that can save companies a lot of money and time.

Implementing both an analytics solution and an AI solution increases transparency at several levels in an organization. AI only increases it. This might threaten some of the traditional workforce and they might work to undermine AI implementation. So, it’s necessary to make everyone comfortable with increased data transparency in a systematic manner. A data-driven culture has to be fully embraced for it to succeed.

The IoT impact

Introduction of new technologies such as the internet of things is adding another dimension to analytics and AI play. IoT is generating massive amount of data and helping everything communicate with everything else in the world. Crunching this data is what analytics does best, while making intelligent decisions out of it is the job of AI. The more data IoT generates, the better analytics and AI get at what they do.

Examples to understand the assertion

Let’s take a look at some of the examples to drive home the point.

E-commerce
E-commerce has burgeoned into huge business, with online sales now threatening traditional brick-and-mortar shops. E-commerce companies heavily depend on analytics to price the items in real time and make a sale. Due to this, these companies continuously feed their analytics system with huge amount of data. This includes latest competitor prices, their own inventory levels, changes in shipping costs and a plethora of other factors. Based on that, the analytics system predicts an optimum price to maximize the sales. Contrast this against a company still saddled with processing all this information the old way. It’s easy to conclude that without analytics, an e-commerce company doesn’t stand a chance.

An AI solution can further deep dive into the recommendations provided by the analytics solution. For example, while analytics will provide you with optimum price points for customers in different regions, an AI solution can tell you what will be the impact of selling the customer the good at that price point, how the customer might react, how could this impact the future customer behavior and decisions, and ultimately how much would it impact the long-term credibility and sustainability of the company. One might wonder that this type of analysis definitely requires a massive amount of experience and deep insight. And that’s precisely what AI systems intend to do. Replace — and even beat someday — the human intelligence.

Online conglomerates such as Amazon employ AI to even predict what a customer is going to buy into the future. So, in a way, they are reading your mind. And going even a step further, maybe planting thoughts in your mind about buying certain products. It’s mind boggling, but it’s happening right now with you and me. And it’s going to get even more amazing and astounding into the future.

Insurance
Similarly, insurance companies are extremely dynamic in offering insurance premiums to close a deal. And they leverage competitors’ data to price their premiums.

Using analytics, companies can update and advertise their revised quotes every 10 minutes with analytics in place. Armed with an AI solution on top of the insurance solution, insurance companies can make a marked improvement to their claims management and fraud detection processes. AI systems can account for new types of cases, improve assessment of severity of damage, match the veracity of information from different sources, pinpoint the areas to plug into and expand, predict with far better accuracy a fraudulent claims case and so on.

Fund management
Consider fund management. Traditionally, fund managers have relied on studying reams and reams of data to make sense out of them and taking actions based on that. Essentially, the fund manager uses his experience and acumen in deriving conclusions and taking appropriate actions. Now, analytics augments the fund manager’s capability exponentially to crunch the data and make conclusions. Now the same fund manager can summon thousand times more data from disparate and unstructured sources as well. And, at the click of a button, the fund manager knows answers to his queries, anytime anywhere.

Even without an AI solution, the manager can still make useful and compelling decisions. However, armed with an AI solution, the fund manager literally has a huge team of other brilliant minds who can provide even smarter insights.

Conclusion

Even without an AI solution in place, you still need analytics to understand the data. But what would you do with an AI solution without having any analytics to put your data through? Pray, maybe, but without having much faith in your solution. And that is a road that goes nowhere!

All IoT Agenda network contributors are responsible for the content and accuracy of their posts. Opinions are of the writers and do not necessarily convey the thoughts of IoT Agenda.

Start the conversation

Send me notifications when other members comment.

By submitting you agree to receive email from TechTarget and its partners. If you reside outside of the United States, you consent to having your personal data transferred to and processed in the United States. Privacy

Please create a username to comment.

-ADS BY GOOGLE

SearchCIO

SearchSecurity

  • Passive Python Network Mapping

    In this excerpt from chapter two of Passive Python Network Mapping, author Chet Hosmer discusses securing your devices against ...

  • Protecting Patient Information

    In this excerpt from chapter two of Protecting Patient Information, author Paul Cerrato discusses the consequences of data ...

  • Mobile Security and Privacy

    In this excerpt from chapter 11 of Mobile Security and Privacy, authors Raymond Choo and Man Ho Au discuss privacy and anonymity ...

SearchNetworking

SearchDataCenter

SearchDataManagement

Close