Mistakes with your data you don’t want to make

Over the years, the Stimulus blog has provided insight into innovations, enhancements, and improvements you should be making with your data. We are taking the opposite tact this time. This blog contains recommendations for what you should not be doing with your data, and how to avoid the pitfalls of common data mistakes.

 

What are the biggest mistakes companies make with data?

Many modern businesses run into trouble with data, which causes them to miss opportunities, duplicate effort, or put them at risk of reputational damage.

By failing to set up the proper data governance systems and policies, or by overlooking the importance of data quality, businesses have wasted a lot of time and effort in having to redo or rethink technology projects. Neglecting to educate and involve users in data practices and standards is a problem, so too is neglecting data security and privacy.

Data mismanagement can have dire consequences for any business. Here is our top list of data don’ts to help keep you aware of what not to do.

 

 

Don't ignore data security

Disregarding data security concerns and threats can cause major problems; data breaches are not a risk you can afford to take lightly. Unfortunately, many people who don’t understand the risks and the wider international data security context, and some it seems that organisations operate with an illusion of invincibility against data breaches or leaks. When news of data threats, hacks, or vulnerabilities go public, you can suffer serious reputational damage.

Risky situations can arise both internally and externally, including when:

• Human error causes system access issues
• You don't have multifactor authentication in place
• Staff cause a breach when they are disgruntled or as an act of theft
• Passwords become compromised due to poor internal practices

To start strengthening data security in your enterprise business, you need first to be familiar with the data you hold. If you haven’t already, begin this assessment with the development of a data inventory. You will need to get a clear look at:

• Where data resides
• Which systems hold what data
• What data security regulations or laws relate to the data you discover

A data inventory is not just a list of your data; it's a powerful tool that can help you develop a robust data protection plan or strategy. By logging your data, you can also identify and address any high-risk areas, giving you a sense of control over your data security.

Big organisations don't need blanket security, with the same level of security across all records and data sets. They need to understand the risk associated with each data set or system, which determines where to invest time and money to protect it. Your data inventory will help you understand:

• What needs to be protected?
• How will each dataset be protected in relation to specific requirements
• What plans, policies and procedures will help you protect the data?

Don’t forget to consider security related to your hardware, either. Understanding how data flows through your networks, servers, and firewalls is also critical.

 

Don't operate unethically with data

Ignoring your ethical and responsible use of data and records is another big don’t.

It can be hard to imagine how data—nothing more than coded records in your systems—can embroil you in an ethical dilemma. However, it is essential to understand the moral implications of data generation and storage in today's world. Data privacy is a top concern and needs to underpin your organisation's culture and practices. The consequences of unethical data use can be severe, damaging your reputation, and breaching your customers' trust.

Operating ethically with data is not just a choice; it's a responsibility. It means educating employees about their responsibilities through regular training and education. This comprehensive approach, driven by senior staff but with everyone empowered to act, makes data protection a collective responsibility.

Compliance and integrity can be supported with policy and regular auditing. Ethically maintaining user and customer data needs to be led from the top and made something that everyone is aware of and can participate in, fostering a sense of accountability and responsibility.

 

Don't generate so much data you don't know what to do with it

With the current obsession with data, it can be tempting to think that you need all the data and that you need it all now. Big data is typically characterised by its volume, that is how much of it you have, as well as its variety and velocity. It pays to be realistic about your data—work out what you need to know to get the best operational insights. A good data strategy will help you determine what data you need for effective operation, and which is simply nice to have.

 

Don't ignore inconsistencies and errors

In some businesses, the adage “close enough is good enough” is used to describe data acquisition and entry processes. Unfortunately, inconsistent data records and data sets full of errors can cause a big problem with the smooth operation of your data processing tools, including your CRM and ERP.

Sometimes, these solutions can do some of the work of standardising data, but you want to have a high-quality baseline before you begin. Some tools, like Pimcore, will even flag any report records with errors or corruptions.

Furthermore, if you are investing in data analytics, it's disappointing to think that the data you see through reporting might be inaccurate. While occasional anomalies in data are bound to be expected, if you want to get reliable insights, you must prioritise data integrity.

There are specific industries where data accuracy is a big concern. Businesses in particular industries cannot risk working with accurate information. Data related to medical and health matters, clinical trials, insurance, and even financial management; errors can have far-reaching ethical concerns.

 

Don't operate without data standards

Data standards help ensure you avoid ending up with incompatible data sets. Data standards set out clear guidelines around:

• How data is created, entered, tagged, labelled, and stored
• How data is processed in multiple separate solutions and tools

Attempting to integrate incompatible data sets poses a risk to the overall data quality and integrity. Incompatibility issues can also result in operational inefficiencies and increased costs due to data inconsistency and the necessity for manual data manipulation.

 

Don't focus on the tools at the expense of people

Your employees are not just users of your data management tools; they are a vital part of your data management strategy. They will play an even more critical part in generating and using data effectively than any tool you implement. Invest time, effort, and communication into your people. Listen to their experiences, learn the pressure points, and ask for ideas for improvement. Addressing culture and attitudes to data can prevent technology and data projects from failing.

When it comes to data security, your employees are not just users of your data management tools; they are a vital part of your data management strategy. They can play an even more significant part in protecting your assets than even the most advanced data security tools. By knowing how to handle suspicious emails, they can save your business thousands of dollars. Educating and supporting your staff about data security risks and best practices is as important as investing in new software-based security solutions.

 

Don't give up when you hit a data hurdle

You have probably heard the sobering stats about the number of failed data projects. Knowing when to give up is essential, but the time to call it quits isn't usually when you face your first project hurdle. There are many reasons you may run into problems. It could be the technology you chose or a software engineering problem. It's just as likely that non-technological factors have played a part.

Human factors, the people component, organisational challenges, or structural shifts can all threaten to derail a project. When you hit a hurdle, ask the who, what, when, where, and how the hurdle has shown up. It might be the realisation that your plans, timelines, and budget weren't realistic or failed to address critical milestones. Ask these questions when you hit that first data hurdle:

• What is the issue, and what can be done to resolve it- bring more clarity to what has led you to where you are
• Who was involved in the issue, and who can help you resolve it?
• When does the issue originate, and when must it be resolved for a positive outcome?
• Where in the technological or business network did the issue arise?
• How can the issue be resolved, and how much investment is required?

Budget concerns, challenges and changes can also turn a project for the worse. If budget issues become problematic, first review your spend to date. If there is a reduction in what you have left to spend, then assess how you can alter your existing financial plan.

Another common issue is where one technology project has overtaken another in urgency or priority. The best way to handle this is to demonstrate how reliant technological projects can be on each other. Something that has been planned out and has clear objectives and deliverables is unlikely to become important just because something else is trending right now.

 

Don't make data IT's problem

Data is everyone's business in an enterprise business. While the IT team may be tasked with implementing new tools and solutions and even setting up strategies for finding and fixing bad data, everyone needs to play a part in data handling. Unless there is a cohesive organisation-wide commitment to data quality and process, the technical team can only do so much.

There is always a need for two way communication about best practice data handling and organising data standards. The IT team need to be able to communicate the processes, needs and expectations around systems and data, in a way that non-technical workers can understand. On the other hand, everyone needs to be open and willing to communicate about how they are using, and any problems they are experiencing with data.

 

Don't disregard Pimcore as your MDM tool

Pimcore is our preferred Master Data Management tool. This open-source solution can help you turn around the don'ts on this list and take charge of your data. Pimcore can help you:

• Handle data security effectively
• Enhance your data governance
• Deal with errors
• Integrate multiple data sources to produce a golden record
• Establish and use data standards and rules
• Overcome challenges you have encountered in the past
• Deploy an easy-to-use solution that will work for everyone around the business

Pimcore is an open-source solution that can be adapted and altered to suit your specific needs. By combining different types of software tools (CRM, CMS, DAM, and PIM) into one cohesive solution.

Pimcore lets you handle data and user records with confidence and consistency. You can easily access, process, and manage thousands of customer and product records through an easy-to-implement data structuring system that organises records into categories and subcategories.

 

 

---

Related Questions

How to set a firm foundation for data governance?

A data governance framework enables you to establish initiatives, policies, guiding principles, and standards for data handling. A robust data framework is not just a good idea, it's a necessity that gives you a clear position on how data is used and how decisions are made about this use. A due diligence approach can help you ensure that you have a process for data validation and appropriate use.

 

How do MDM tools us AI to help data governance?

Many MDM tools have inbuilt data governance management capabilities, often powered by machine learning or AI.

By deploying AI in MDM, you can achieve higher accuracy and reduce the effort required for record creation and handling. When supported by a strong governance framework, AI can be used to migrate and manage data sets. This includes through:

• Recommended data alterations
• Analysis of workflow processes
• Comparing records and matching criteria
• Extracting records in response to rules
• Data optimisation