3 best practices for Data Estate Modernisation

data_estate_modernisation_best_practices

Better decisions, efficient operations, growing revenue and increased competitive advantage - these are the top 4 business intelligence and data analytics goals modern organisations strive towards today.

Investment in cloud-based business intelligence (BI) data analytics technologies is projected to reach US $2.94 billion in 2018, and their benefits -  visualisation, dashboards, data warehousing, reporting and end-user self-service - round out the top five big initiatives and technologies every year for the journey to a modern data platform.

We speak to many customers who don’t know where to start when it comes to leveraging the many benefits modern data solutions offer. This comes down to the absence of a robust data estate modernisation strategy for organising, governing and analysing information, tailored around business goals - not just technology enablement.

We breakdown what data estate modernisation is, how BI fits into your data solution, the important first steps in establishing your data strategy, and why data estate assessments are key to avoiding common data pitfalls.

 

What is BI and how does it help my data strategy?

Business intelligence (BI) refers to the best practices, applications, infrastructure and tools that help organisations access and analyse their business-critical data to make better decisions.

At its core, BI is achieved through your business analytics solution (or data analytics solution for advanced data specialists) such as Microsoft Power BI. These are designed to help you gain a historical overview of your current data estate and what’s happening in your organisation, so you can increase information accessibility through dashboards, visualisations and reporting, and produce more insights to make better decisions.

With powerful business data analytics tools like Microsoft Power BI, you can expect to leverage:

  • A single view of your business critical data

  • Live monitoring of the health of your data

  • Rich visualisations to understand how you’re running at a glance

  • Centralised storage and analysis, eliminating data silos

  • A quicker way to turn detailed insights into action

None of these benefits can be realised, however, if you have no plan and only focus on the technology.

 

Common BI and data platform problems

We encounter many companies who are excited to jump into the latest business analytics tools like Power BI solely based on their many consolidation, integration and visualisation capabilities - specifically, so they can centralise their most important business data into one place and better prepare it for analytics.

Organisations often fall into the trap of pushing for a massive integration of existing applications into solutions without an optimal data platform, or fail to get a current data estate assessment or detailed plan for modernisation.

In turn, many strive for a quick return on investment (ROI) when they’re not ready to use analytics as an organisation.

Due to the complex nature of centralising and analysing enterprise-grade big data, this rushed approach results in several mistakes and problems:

  • Inflated budgets with overly high costs

  • Confusion and misunderstanding surrounding

  • Project timelines with little to no successful outcomes

  • Reduced perception of value of BI across the organisation

The truth is getting your data strategy and the technology working in tandem is hard. Wanting dashboards and reports that source and display the data you desire without understanding the questions that need to be answered within the business is a more common roadblock than you may think.

While the planning elements behind data management are not as fun as interactive dashboards and handy visualisations, it’s the priority for decision-makers to clarify the purpose of their desired data solution with a data estate modernisation assessment and plan, to help guide the eventual data analytics strategy.

 

How do I approach data analytics more strategically?

It’s clear the businesses doing data analytics right are those who have not just taken the time to understand the capabilities of the most popular tools like Microsoft’s Power BI, but who have:

  • Assessed their current data estate

  • Identified their current operational gaps and needs for greater data initiatives

  • Can articulate and understand, as an organisation, how data analytics supports their goals

A study conducted by A.T. Kearney and Carnegie Mellon University surveyed 430 international companies and found the enterprises extracting the most value out of their data initiatives had:

  • Taken the time to assess their data platform and plan

  • Leadership that fostered an in-house data analytics culture

  • Focused on data collaboration across the organisation

  • Instilled confidence in BI’s benefits

Meanwhile, the laggards in data analytics focused solely on the tech’s capabilities.

Dresner also found businesses which define their success by aligning with analytics and BI tools are getting the best results, while the most unsuccessful are those solely focused on big data adoption.

In summary - in order for your BI initiatives to deliver the desired results, you have to first establish a plan, assess your current state, identify data stores and define long-term goals. Here are our recommended three best practices.

 

1. Establish your data plan with a current data estate assessment

Data Estate Modernisation assessment

 

A data analytics solution fully aligned with your business goals necessitates the thorough planning and organising of several elements to establish an effective plan - and to avoid the trap of adopting a wasted and expensive solution.

Take the time to sit down with your decision makers, work out if you need outside expertise, and begin your data estate modernisation initiatives by mapping out your current data estate, basic objectives, business processes, capabilities and policies and what you want to enable with tools like Power BI with these key questions:

  • Analytics and reporting: What are the key analytics and metrics you want to track? Does everyone understand why they’re most important to improve your business, or how they will improve things? Start building from the ground up and take the time to define the data most relevant and actionable for your ideal data analytics solution, that can be improved with better data storage, performance, and insights capabilities.

  • Customers: Will your data solution be used for your customers? Ensure you align your overall analytics roadmap with your clients BI needs, in addition to your own.

  • Data environment: Whether you’re on cloud, on-premises or hybrid, how are you currently handling consolidation, storage, and transformation of your historical data? This can help you get a 360 degree view of your business so you can pinpoint where you’re struggling and where you need to adjust in your strategy.

  • KPIs: What are the industry KPIs you’re aiming to hit - ROI? Productivity? Sales? Profit? What are you achieving now? Take the time to outline the internal targets you want to meet specific to your business, so you know exactly what you’re aiming for with your data analytics aspirations from the beginning.

  • The Big Question: Why does having a data solution align with your objectives best? If you can’t articulate this one off-the-bat, it’s time to re-evaluate before going ahead.

The purpose of this step is to aid gap analysis of your existing state and desired future state. Whether you seek help from professional consultancies like Xello for your data estate modernisation or spearhead your initiatives in-house, it’s essential to have a solid foundation to build your eventual data strategy upon.


2. Identify and organise your data sources

It’s safe to say the majority of organisations have multiple islands of data (data stores) across their cloud and on-premises environments. Disparate sources makes it harder to access, share and analyse, so before you even can think of adopting self-service BI or other data solutions, you have to do some heavy lifting and identify and organise where everything is.

  • Cloud, on-prem or hybrid?: Where is your business-critical historical data currently stored? Are they spread out in the cloud, on-premises or both? Ensure you know exactly where your most important information is spread out so you can bring it all together with the right solution based on your setup. For instance, if most of your data is across Azure and on-premises, it’s best to rule out using data services from Amazon and Google.

  • What type of business data?: Determine the types of business data (machine data, master data, qualitative data, metadata, market research, customer data, transactional data, etc) you collect, store in databases and can be processed for insights. Whether it’s generated by your website, purchased products or services like CRMs, gather it all.

  • Data quality: Assess the consistency and quality of your data - is it raw, structured summary, metadata, unstructured? The quality of your data will influence what type of warehousing solution you need to kickstart your analysis initiatives.

This step helps you establish what sources and data you have, the quality and environments, they’re currently stored in, and what information is most relevant from each store.

 

3. Assess your data warehouse needs

Modern Data Platform lifecycle

 

Once you have identified what data stores you have and what your most relevant information is, the next step is to examine where you need to gather your data for transformation and analysis.

A Modern Data Warehouse (MDW) is the most popular solution to bring together your critical data and organise it in a way that gives you a proper historical overview of your performance and success, so you can better pinpoint where you’re doing well in the business and what you need to change for better strategic decision-making and analysis.

  • Data schema: How will your DW be structured - De-normalised design? Star schema? Snowflake schema? Much like databases, a warehouse needs to maintain a logical structure of its contents to be accessible, structured and able to support queries.

  • Data Warehouse model: Does a data mart warehouse, virtual data warehouse or enterprise data warehouse model suit your data assets better? Whether you need to aggregate data that spans the entire organisation or only from a specific area, the structure of your data model sets the foundation for all your dashboards, data analysis and reporting initiatives.

  • ETL vs ELT: Which method of loading your critical data into a DW best suits your current business environment - Extract, Transform, Load (ETL) or Extract, Load, Transform (ELT)? Knowing how data is extracted, held and then converted into a suitable form for your target warehouse system, or loaded directly into the warehouse without a staging database is key for successful analysis once it’s all loaded in.

  • Recovery and backup: Does your preferred DW solution have options to backup, recover and monitor your data? Whether you opt to go for a cloud-based or on-premises warehousing solution, being able to protect your data and view its current health is a no-brainer.

  • Scalability: Is the DW built for both current and future storage and insight extraction needs? Ensure it has the scalability to accommodate change and growth in your analysis efforts, and adhere to regulatory and compliance requirements.

Being able to articulate what you need to consolidate and structure your key data for analysis is a key step forward in creating a comprehensive data strategy for your organisation, and a roadmap towards a modern data estate.

 

Data Estate Modernisation - Key takeaways

  • Planning is the first but most important phase to successfully adopting your ideal data analytics solution.
  • Understand the business questions you’re seeking to answer before you jump into the tech.
  • Take the time to define your own unique data collection and analysis requirements.
  • Significant preparation and forward-planning on your part is required to properly leverage data analytics tools like Power BI to its full potential, without facing resource wastage or skill gaps.
  • It’s always better to start small and focus on mapping your business needs and pain-points to establish a data modernisation plan and align your future data solution with your desired future business outcomes.

 

Data Estate Modernisation - Next steps

If you’re able to define and articulate exactly how a data analytics solution helps your business, the next step is establishing a comprehensive data blueprint/strategy to modernise your data estate. This guides your Proof of Concept (PoC), which you can then better present to your CEO and better illustrate the potential ROI and benefits.

If you need help with your data estate modernisation, data platform consultancies like Xello provide comprehensive data solutions to help you align your estate with Azure Cloud and the capabilities of its latest tools with your goals.

With our Data Estate Modernisation blueprint, we help businesses by performing a discovery of your current data estate, provide expert advice and a detailed blueprint to help you modernise your data architecture. We also offer guidance on other key parts of your data strategy, including governance, consolidation, migration and management. 

 

Xello
Author: Xello

We believe there’s a better way to provide business solutions. Our team takes a forward-thinking approach where customers are partners, and digital transformation is all about the ‘why’ for today – and tomorrow. We’re on a continuous mission to deliver you the best Azure Cloud, Data Platform and Modern Workplace solutions that keep you competitive and ahead of the game.