Choosing a artificial intelligence provider for analytics projects, dynamic pricing, demand forecasting is, without a doubt, a process that should be on the table of every manager in the industry. Therefore, in case you are considering to speed up the process, an exit and the hiring of companies specialized in the subject.

A successful implementation of analytics is, to a large extent, a result of a well-balanced partnership between the internal teams and the teams a analytics service provider, so this is an important decision. Herein, we will cover some of key concerns.

Assessing the AI provider based on competencies and scale

First, you must evaluate your options based on the skills of the analytics provider. Below we bring some for criteria:

• Consistent working method in line with your organization’s needs and size.
• Individual skills of team members and way of working.
• Experience within your industry, as opposed to the standard market offerings.
• Experience in the segment of your business.
• Commercial maturity of solutions such as the analytics platform.
• Market reference and ability to scale teams.
• Ability to integrate external data to generate insights you can’t have internally.

Whether developing an internal analytics team or hiring externally, the fact is that you will probably spend a lot of money and time with your analytics and artificial intelligence provider(partner), so it is important that they bring the right skills to your department’s business or process.

Consider all the options in the analytics offering.

We have seen many organizations limit their options to Capgemini, EY, Deloitte, Accenture and other major consultancies or simply developing internal analytics teams. Although:

But there are many other good options on the market, including the Brazilian ones which are worth paying attention to the their rapid growth. Mainly within the main technological centers of the country, such as: in Florianópolis or Campinas.

Adjust expectations and avoid analytical frustrations

We have seen, on several occasions, the frustrated creation of fully internal analytics teams, be they for configuring data-lakes, data governance, machine learning or systems integration.

The scenario for the adoption of AI is similar, at least per hour, to the time when companies developed their own internal ERPs in data processing departments. Today of the 4000 largest technology accounts in Brazil, only 4.2% maintain the development of internal ERP, of which the predominant are banks and governments, which makes total sense from the point of view of strategy and core business.

We investigated these cases a little more and noticed that there are at least four factors behind the results:

• Non-data-driven culture and vertical segmentation prevent the necessary flow (speed and quantity) of ideas and data that make analytics valuable.
• Projects waterfall management style performed in the same manner as if the teams where creating a physical artifacts or ERP systems, this style is not suitable for analytics.
• Difficulty in hiring professionals with knowledge of analytics in the company’s business area together with the lack of on-boarding programs suited to the challenges.
• Technical and unforeseen challenges happen very often, so it is necessary to have resilient professionals used to these cognitive capoeira (as we call here). Real life datasets are never ready and are as calibrated as those of the examples of machine learning of the passengers of the titanic dataset. They usually have outliers (What are outliers?), They are tied to complex business processes and full of rules as in the example of the dynamic pricing of London subway tickets (Article in Portuguese).

While there is no single answer to how to deploy robust analytics and governance and artificial intelligence processes, remember that you are responsible for the relationship with these teams, and for the relationship between the production and analytics systems.

Understand the strengths of analytics provider, but also recognize their weaknesses

It is difficult to find resources with depth and functional and technical qualities in the market, especially if the profile of your business is industrial, involving knowledge of rare processes, for instance, the physical chemical process for creating brake pads or other specific materials.

But, like any organization, these analytics provider can also have weaknesses, such as:

• Lack of international readiness in the implementation of analytics (methodology, platform), to ensure that you have a solution implemented fast.
• Lack of migration strategy, data mapping and ontologies
• No guarantee of transfer of knowledge and documentation.
• Lack of practical experience in the industry.
• Difficulty absorbing the client’s business context

Therefore, knowing the provider’s methods and processes well is essential.
The pillars of a good Analytics and AI project are the Methodology and its Technological Stack (What is a technological stack?). Therefore, seek to understand about the background of the new provider, ask about their experiences with other customers of similar size to yours.

Also, try to understand how this provider solved complex challenges in other businesses, even if these are not directly linked to your challenge.

Data Ethics

Ethics in the treatment of data is a must have, therefore we cannot fail to highlight this topic of compliance. It is not just now that data is becoming the center of management’s attention, however new laws are being created as example of GDPR in Europe and LGPD in Brazil.

Be aware to see how your data will be treated, transferred and saved by the provider, and if his/her name is cleared on google searches of even public organizations.

Good providers are those who, in addition to knowing the technology well, have guidelines for dealing with the information of your business, such as:

• It has very clear and defined security processes
• Use end-to-end encryption
• Track your software updates
• Respect NDAs (Non-disclosure Agreements) – NDAs should not be simply standard when it comes to data.
• All communication channels are aligned and segmented by security levels.
• They are well regarded by the data analysis community.

Conclusions and recommendations

Choosing your Analytics provider is one of the biggest decisions you will make for your organization’s digital transformation.

Regardless of which provider you choose for your company, it is important that you assemble an external analytics consulting team that makes sense for your organization, that has a technological successful and proven business track that supports your industry’s demand.

What is Aquarela Advanced Analytics?

Aquarela Analytics is the winner of the CNI Innovation Award in Brazil and a national reference in the application of corporate Artificial Intelligence in the industry and large companies. Through the Vorteris platform and the DCM methodology, it serves important clients such as Embraer (aerospace), Scania, Mercedes-Benz, Randon Group (automotive), SolarBR Coca-Cola (food retail), Hospital das Clínicas (healthcare), NTS-Brasil (oil and gas), Auren,SPIC Brasil (energy), Telefônica Vivo (telecommunications), among others.

Stay tuned following Aquarela’s Linkedin!

Author

Joni Hoppen

Founder – Commercial Director, Msc. Business Information Technology at University of Twente – The Netherlands. Lecturer in the area of ​​Data Science, Data governance and business development for industry 4.0. Responsible for large projects in key industry players in Brazil in the areas of Energy, Telecom, Logistics and Food.