Our Data & Analytics team explores key features, business benefits, industry applications, and technical capabilities of Microsoft Cortana Intelligence Suite.

Blog five of a series.

With big data analytics revolutionizing the way companies make business decisions, there are multiple products on the market that can help.

When selecting a product to meet your organization’s needs, you should consider several factors: data size and complexity, objectives for the data, ease of use, ease of implementation, and cost.

Because big data analytics products come in two main product implementation categories, businesses must also decide whether to choose on-premise or cloud, as discussed in a previous blog.

In this blog, we’ll assume your organization is looking to move its data and analytics to the cloud given that Cortana Intelligence Suite is a solely cloud-based product. We’ll also mainly focus on the ease of use, implementation, and cost factors.

Implementation Structure of Microsoft Cortana

Many products are structured to be sold as a single product or framework, which means you get a set product from data storage to tools at unit price points. A prime example is IBM Watson.

Having a single setup does take some of the guesswork out of what to implement, however, it can also limit users to a single framework. This works out well if you have a robust product and a project team with expertise in getting things up and running with that platform.

However, platforms tend to limit an organization’s ability to evolve and innovate, mostly because they can pose limitations to the features they provide.

Cortana has a more al-a-carte model when it comes to their implementation and pricing. Based upon the user’s level of commitment, expertise, and budget, you can mix and match various Microsoft solutions and other third-party products to create a unique ecosystem that works for your organization.

This model has the benefit of allowing customers to ease their way into an analytics framework, while letting the framework evolve as the organization evolves and innovates.

Associated Cost Factors

However, the scope of Cortana’s ecosystem makes it a challenge to evaluate the overall cost. There is no good way to quantify how much it will cost without knowing what’s being implemented and the strategy that your organization wants to take. There are a number of variables at play that need to be analyzed and evaluated during implementation to ensure that Cortana is cost-effective for the organization.

This approach, while very flexible, means that your organization will need to constantly re-evaluate strategic initiatives to ensure you’re paying for what you need. It also means you have a need for a Cloud and Data Architect.

As with any Data & Analytics project, it is always good to have the following skill sets on hand (these skills can be found in one or more people). The roles are not listed in a particular order:

  • Cloud Architect – Someone who has a firm understanding of system infrastructure: both on premise and hosted. This individual will be critical in helping determine the best and most secure way to connect various systems together to get the most out of any implementation.
  • Data Architect – Your go-to resource for making sure that you have a complete set of data from all of your critical systems, and that your semantic layer and data models accurately reflect responses to business questions.
  • Data Scientist – Your wizard. This person is key to setting up those complex neural networks that divine the answers to the questions you didn’t even think to ask.
  • Business/System Analyst – A BA/BSA is your key to repeatable results. This is someone who understands the business needs to help the team determine what systems need to be reviewed. An analyst makes sure that architects and scientists abide by the project’s strategic initiatives.
  • Developer(s) – For Cortana, developers that are well-versed with MSDN and Microsoft App development will be important to ensure that all the work that is being done can be served up to the day-to-day end users in a way that will allow them to make critical business decisions.
  • Technical PM (Project or Program) – The Technical PM id a key role to help coach the team toward achieving the objectives of the organization. This role understands the function and limitations of the different systems so that risks can be mitigated and raised before they become a hindrance to progress. This personal will also need to work closely with the architect(s) and scientist(s) to make sure the correct number of instances are active and the correct cloud infrastructure is in place. As with any cloud system, if you aren’t careful the transaction and server time, costs can quickly blow your budget.

A benefit that Cortana has over some of the other Open Source data analytics packages is Cortana’s extensive Microsoft Partner Network and user base, which posts various experiments to Azure and end user tools that you can download and install in your instance.

This can help alleviate some of the immediate pressure from the developers and data scientists by giving them a starting point and easing the learning curve of getting your Cortana instance setup.

Change Management Phase

So, you have decided to go with Cortana. Your experiments are in the queue and you are starting to see some results. Now comes the change management phase of the project.

Getting people to start using the data and make decisions based on data – and not on the “how we have always done it” mentality isn’t always the easiest.

Having spent my earlier development career building various business and engineering tools for companies, I have found that there are two key components to help organizations adopt new tools:

  • Single Point of Access: If users have to go to multiple places to get access to different reports, or have to log in and log out of different systems – they tend to forget the new tools exist. This is one thing that Lotus Notes actually got right.
  • Incremental changes: Ease the users into the new tools and reports that you want them to start using. Typically the first response is to introduce users to systems rather quickly and all at once to get it over with – much like taking off a Band-Aid. But, by phasing in analytics reports slowly, you can build trust in the system, test the reports and structure in a production environment, and work out any kinks. This makes it easier for everyone to learn how to use it.

Like many analytics packages out there, Cortana allows customers to use “third party” visualization tools like Power BI (yes, this is a Microsoft package, but here it acts like a separate application) or Tableau.

Where I think Cortana really starts to separate itself from other analytics packages, is its ability to integrate with the Windows OS. Microsoft has created an ecosystem that will allow you use the Cortana personal assistant that comes with Windows 10, so you verbally interact with your data.

Evolution and Innovation

To ensure a smooth deployment, structure your implementation as a program with multiple projects that work together. Full implementation may take more than three years – depending on the complexity of your data and ecosystem.

Once the ecosystem is in place, assign a smaller project team with a few key resources to ensure efficient operation. That team should include a project manager, system architect, BSA, and developer(s).

By requiring this team to keep up with the industry’s ever-changing products and upgrades, you can keep your ecosystem up-to-date with the latest technology and get the most out of your investment. Many organizations make the mistake of letting tools become outdated to the point that they need to be completely replaced. But that can be far more expensive than regular maintenance.

Although Cortana is still a new player in the analytics market, it shows a lot of promise. By allowing organizations to customize their ecosystem to their individual needs, they have opened up the market to organizations that are just starting to wade through their data before diving head first into data lakes.

But like all data exploration journeys, knowing your appetite for risk and being well-prepared with a team of experts is essential to success.

I, for one, am excited to try out Cortana Suite with Cortana Assistant. Imagine this: Asking your computer to pull status reports while drinking your first cup of coffee in the morning – and receiving those reports verbally without ever having to bring up a single website.

 In the next blog, we will explore how Microsoft Cortana Intelligence Suite can be applied in the healthcare industry.

Go Further

About Manish Gupta

Manish is a Management Consultant based in Cincinnati. He’s experienced working with companies to identify the best data and analytics packages. During his career, he’s implemented various cross platform systems. He also has a strong background in small and large scale hardware and IoT solutions. Follow Manish on Twitter.