Home / Our thinking / Insights / How to develop an analytics strategy
How to develop an analytics strategy
Table of contents
What is an Analytics Strategy?
An analytics strategy is a roadmap that businesses develop in order to outline how data is collected and used to make important business decisions. This strategy aims to provide as much clarity as possible on reporting metrics to key decision makers.
Is There a Difference Between Analytics Strategy and Data Strategy?
The terms analytics strategy and data strategy are often confused with one another. But what is the difference between the two?
Data Strategy
When talking about data strategy, we are only interested in the actual data that is being collected by a business. Developing a strong data strategy involves looking at the following components:- Content of data
- Quality of data
- Who has ownership over company data
- Data governance
- Data security
Data strategy is the overarching process that data engineering and data analytics feed into.
Analytics Strategy
Analytics strategy translates the data strategy into an implementable action plan. Linking data to business objectives is key in an analytics strategy. A strong strategy will also educate stakeholders clearly, meaning that the plan is easy to understand. Our access to data is greater than ever, so it's vital that businesses have a plan in place to make the most of this data when making important business decisions.
Creating an Analytics Strategy
Identify Key Stakeholders
The first stage of creating a game-changing analytics strategy is identifying key stakeholders. Failing to address this stage of the process could prove to be a huge mistake in time, so make sure it is high up on your priority list. Ensuring that a mix of key stakeholders are involved in an analytics strategy can make the process more streamlined going forward. Ideally, the person who is responsible for outlining the strategy should choose and liaise with the key stakeholders. Some stakeholders you should consider involving in your strategy are:- Analytics teams (both centralised and decentralised) - these teams are made of data leads who help to oversee analytics strategy with different departments within the business including IT, Finance, Supply Chain and Marketing.
- Project managers - these team members will be responsible for ensuring that deadlines are met.
- Leadership team members - management can help align the business' corporate strategy with the new analytics strategy moving forward.
In order to achieve success, business leaders should consult data analysts before influencing data strategy. It can be very helpful to combine the analyst's knowledge and management's experience to make valuable business decisions.
Initial Research
Initial research should help you understand how each area of the business currently collects data. Holding meetings with different stakeholders is a great way to obtain this information. It is also important to consider what types of data are being collected. Getting this right will enable you to discover the best data insights for your organisation. Learn what data insights are. Some questions to ask each department might look like this:- What does data collection look like right now?
- Which tools do you use to collect data?
- Do you use data to help make any decisions within the department?
- Which metrics are important for your department?
- Do you have the access to the right data for these metrics?
Following this initial research, you'll most likely have some ideas on how each department can improve their data collection process. Once each department starts collecting data correctly, you can build this valuable information into the wider analytics strategy for the business.
Choose an Analytics Model
The third stage of developing a strong analytics strategy is choosing the right analytics model for your business. These models help inform many of the strategic decisions made from data. There are three main models to choose from:
Decentralised
In a decentralised analytics model, each business unit works with autonomy with an aim to meet wider business objectives.
Federated
A federated analytics model involves the central team providing each business department with structures to work with. The central team is the main point of control in this system.
Centralised
Using a centralised system means that no single business unit has control. Instead, the central team has full control and direct the second level of management with decision-making. Choosing one of these models is really important when building your analytics strategy. It's essential to decide the best way to make decisions around data for your business. Is your business more suited to having each business department look after their data, or is centralising this control more effective?
Select Different Analytics Strategy Tools
The next major consideration in our journey to building an impressive analytics strategy is choosing which tools to use. This investment should reflect where your analytics strategy is currently at and where you strive to take it in the future. Some considerations to have at this stage are:- Cost - When looking at the cost of analytics tools, as well as considering the monthly cost, you should also consider the cost of upskilling staff tso they can use the tools effectively. Furthermore, consider the maintenance costs associated with the tools you choose.
- Security and Privacy - Before you decide on analytics tools, security and privacy needs to be considered. Does the security and privacy of the tools fit with the wider business security standards?
- Collaboration - The tools you choose should allow you to share and analyse data in a number of locations and in different formats. They should be mobile-friendly, desktop-ready and easy to use and understand.
- Technology Assessment - An assessment of each tool will need to be made. Comparisons of features between different options, the cost of each tool and return on investment (ROI) are all important factors.
Establish a Data Culture
Sometimes considered the toughest part of producing an analytics strategy, establishing a data culture is arguably the most important aspect to consider. The aim is to ensure that everybody across the business understands the basics when it comes to using and managing data. It's at this stage that many departments realise there are opportunities to automate some of their manual processes. These are opportunities that they were unaware of before the roll-out of the new analytics strategy. A shift in any culture is a large obstacle to overcome, but if done successfully, it can have a significant positive impact on a business. This is the same when changing the data culture. We recommend focusing on the following areas to help you achieve a shift in data culture:- Accessibility of Data - Ensure individuals have access to the data they require to avoid pitfalls in your strategy.
- Trustworthiness of Data - Ensure users have trust in the data they have access to. This is otherwise known as data veracity.
- Training Strategy - Apply the right training strategy around your data can to help with stakeholder buy-in. The key is to not bombard individuals with too much information. Try assigning specific �Data trainers' to upskill each department at a comfortable pace.
- Leading by Example - Achieving a successful shift in data culture requires strong leadership that starts at the very top of an organisation. It is often the case that once data-based decisions have been made by business leaders, the rest of the organisation will start to trust the process.
Availability and Management of Data
The final stage of building an analytics strategy is dealing with the availability and management of data. Shortly after developing a data strategy, you may start to receive a high volume of data requests. This is a good sign that your newly built analytics strategy is working and different departments are now on the same page with data. Managing data requests may seem difficult to start with, but in time they'll feel like standard business practice. The key is to maintain a healthy balance where departments who have non-negotiable data governance requirements get first priority and the other department requests follow.
Considerations Moving Forward
Like with any kind of strategy, your data analytics strategy will evolve over time. Therefore, it's important that you do a number of things as you move forward.
Analytics Strategy Ownership
To maintain a strong analytics strategy, it's important to assign an owner. They'll be responsible for reviewing and promoting the strategy, as well as keeping track of how well it is followed. It's highly likely that if someone is made responsible for the analytics strategy, it will stand the test of time. It makes sense for this person or group of people to be working closely with the strategy on a daily basis.
Documentation
When developing a strategy, it is likely that you will have a lot of documents to store, therefore documenting everything is vital. This will help you to track the strategy progress, review any issues faced and outline procedures and protocols. Documents should be stored in a shared drive that is accessible. This means that if somebody else takes over the responsibility of owning the analytics strategy, they can easily see what has been previously documented.
Analytics Progress
Now that your business is on its analytics journey, you'll want to start using different analytics tools to help you report this progress to the wider business. Choosing the right tools for your business is crucial, as some reporting tools are better for analytical reporting, meanwhile others are better for generic reporting.
Analytics Strategy Shelf Life
The final point to consider is giving your analytics strategy a shelf life. Like any kind of business strategy, you need to ensure that you run with this analytics strategy for a period of time before reassessing it. Ask yourself the following questions when determining your strategy's shelf life:- Has something changed within your business that requires an analytical change?
- Are the maintenance costs of your analytics strategy in line with projections?
- Do you have any new technologies that may be better suited to your analytical needs?
