Applying Big Data technology is perhaps the most important project in the minds of most executives today, all with high expectations about the monetisation of this data. However, a very high percentage of them fail. To be more specific, according to a compilation of studies carried out by consultants specialising in strategy and Big Data, including Gartner and NewVantage Partners, in recent years about the implementation of Big Data in companies, around 80% of Big Data projects fail”. And the first question we must ask ourselves is why?

Reasons for failure in data projects

The figures show us that although companies are investing capital in launching Big Data projects, many are not getting value. There are common reasons why projects tend to stagnate, and we’ll explain them below:

Mismatch between strategic objectives and technical skills: the first thing to do when implementing such a project is to understand the real business problem and accurately assess its risks from the outset. Launching the project without a clear analysis of the problem and without defined objectives in line with a data strategy will blind the strategic vision and turn the project into a failure.

Lack of infrastructure and resources: large data projects require a solid infrastructure and key resources, especially with regard to talent. Those who try to move forward without having covered these crucial functions experience a skills gap and end up finding their projects stalled due to lack of adequate education, training and experience in applying analysis.

Complexity and inflexibility: however, much you have the necessary ingredients to develop this type of project, if the architecture and scope of the project is too rigid, it will fail. There is a tendency to complicate the problem and create solutions that are too complex. This practice takes the focus off the big picture and diverts it from the right solution.

Failed data science model: poor data quality and accuracy are a major obstacle to project success. Integration costs are high and data often remains a large source of undeciphered information. All relationships in the data must be extracted or inferred and made explicit so that the machines can interpret the data properly.

Integration gaps: poor communication between data scientists and business stakeholder’s results in guaranteed failure. It is the most common reason why large data projects fail to be applied to production and fail. The flow of information is critical to integrating a project of this magnitude.

The keys to success in data projects

Reality shows that the path to data-based benefits is neither easy nor simple. Companies need to develop a range of capabilities to be able to transform large data into business, which must be global (involving the whole organisation) and based on routine and repeatable processes and workflows.
There are 4 components of success that will help you mitigate risk and ensure that the results match your expectations for the project:


Data-based growth is not possible without a robust data structure. Organizations must collect internal and external (even purchased) data, store it and make it available to the organisation.
The amounts of data are increasing and new formats are constantly emerging. Companies must be able to analyse this data through sophisticated models and report it in a way that allows all parties to draw valuable conclusions.
This capability requires skills in developing sensors, infrastructure for storing data, and methods for safeguarding data from hackers.


The notion of autonomous teams and decentralised decision-making is fundamental to the development of any strategy, allowing employees to propose their own ideas and even make their own decisions. But autonomy is challenged by increasingly complex projects, agile large scale teams and perhaps increasingly multi-disciplinary.
This is why it is essential to create the post of data manager (CDO), responsible for establishing common communication and ensuring that the group acts as a self-organised or autonomous team.
At the same time, it is important that the team is trained in what the data scientist can and should contribute, and that the data scientist ensures that the data will help employees make new decisions.


Technology is equally essential to the success of the data. It involves an important first step in the application phase, as well as an essential component for the digital spine.
No matter how much you have the best ideas and the most valuable data, if your technology is not up to the task or is not scalable, it will not ultimately create real value for the company. That’s why it’s essential to have a strategic partner who is an expert in the field to advise you.


When considering a data strategy, it is essential to consider what is allowed and what is prohibited. There are three areas of responsibility to consider; the law, formal contracts and social standards.
When data has the ability to identify an individual, certain data security and enforcement regulations apply. It is therefore important to clarify the expectations and objectives of all parties in order to ensure a sound contractual basis.
On the other hand, the use of data may be legal and lawful for the parties involved in the contract, but may be inappropriate for customers or society. The company should have someone responsible for dealing ethically with the data and setting the parameters.

Further implications

The importance of culture and the human factor play an equally important role in the development of data projects. Discover the 10 steps to achieving a good data culture, from the importance of having a committed manager to the need for a fluid communication system.

The distribution of resources is another key point when considering the investment needed for the project. If resources are distributed too thinly, we can jeopardise more priority projects. For this reason, it is necessary to take advice and find a balance in the investment, making a forecast of funds and considering implementation problems that may arise as the project progresses.

Your Big Data project

Scope, time, budget and quality are critical components of any project. Failure to comply with one or more of these measures is the reason why most data projects are challenged or fail completely.
As a business, you must first ensure that data projects are aligned and related to business priorities, and then build an entire agile, visible and non-visible infrastructure around them. It’s worth doing this little by little, with simple models and guided by a strategic partner like Nexus Integra.

The Nexus Integra integrated operations platform consists of a powerful three-layer structure that helps your company to integrate, acquire, standardise, unify, manage and display data in a simple way.
– The first layer, Nexus Connect, helps your company integrate and acquire unstructured data through sophisticated IoT devices and tools.
– The second layer, Nexus Core, is responsible for standardising and unifying this data through the Big Data.
– The third layer, Nexus Applications, is made up of native Nexus Integra applications and others developed by third parties that provide you with the necessary tools in a global operations environment to manage and display the data in a simple way.

Contact us and let’s talk about how Nexus Integra can help your company to develop a successful project with Big Data technology.