“the next frontier for innovation, competition, and productivity”.
It stated; “a retailer using big data to the full could increase its operating margin by more than 60 percent”. The whole article is chocked full of mind-blowing statistics. The article presents impressive statistics about the benefits of data analytics across;
- Increased quality
- Increased operational efficiency
- Reduced expenditure
- Improved revenue collection
- Reduced fraud and errors
- Increased consumer value
Despite this research being from 2011, adoption of data analytics is relatively slow and low. It is even slower among small businesses.
For your small business, this is great news. Being an early majority right after the chasm, you could gain a valuable edge over your peers. You could even compete strongly against the big guys too. This means starting your data analytics now while the spoils are good.
The key challenge contributing to slow adoption is the complexity and cost of data analytics projects. These can be particularly significant for small businesses. To help you get started while avoiding these pitfalls, we put together a few pointers to consider.
Set SMART goals for your organisation & Project
It’s important to know what you want to achieve before you set off. All the elements of SMART goals are the building blocks of a good data analytics project. Analytics is broad in scope and applicability and without a clear set of goals, it will be a never-ending endeavour that doesn’t yield very much. Data analytics is a means, not an end, and it is important to make sure it can be reconciled with what the organisation wants to achieve as a whole. Projections of resources that will go into the project should yield your desired return on investment when compared to your projected benefits.
Determine your KPAs
Key Performance Areas (KPAs) are the sections of your business whose performance is critical to attaining your goals. These will mostly lie in two broad segments;
- Long-run KPAs that are aligned with the vision, mission and long-term goals of the firm
- Short-run KPAs that are critical to taking advantage of, and eliminating threats, that arise in the short term
Though it is desirable in the long run to implement data analytics throughout the organisation, first priority goes to KPAs. Think about a situation where you want to implement marketing and product analytics. Even if you manage to improve your product, that would yield limited value if you have no one to sell the product to. Under any limitations, applying your marketing analytics should take precedence. If the situation where high customer churn from dissatisfaction, then product analytics would take precedence. Properly drafted KPAs should be relatable to the SMART goals set above. Prioritizing the KPA that takes precedence according to these goals is crucial.
Determine your KPIs
Key Performance Indicators (KPIs) are the critical metrics that show the performance of your KPAs. KPIs should be crafted to relate movement of KPAs to SMART goals closest. This is critical to ensure you come out with as few KPIs as possible. Data analytics projects that analyze too many metrics can cause information overload and reduce implementation of insights.
Also, because moving from metrics and KPIs to insights takes a lot of critical thinking, pursuing too many KPIs could result in getting insights that no longer have value when applied. A 50% insight today is better than a 100% one a month later. Benchmarking can be used to make metrics even more useful. Comparing metrics to appropriate industry peers will quickly reveal note-worthy insights.
Craft your data analytics strategy
The key to a good data analytics strategy is driving actionable insights from data. Avoid the common misstep of focusing solely on movements in KPIs and metrics and not underlying intuitions. As the American Marketing Association puts it ‘The “what” pertains to analytics, while the “why” and “how” provide insight.’. Analytics will only give you the trends in the data; you have to apply critical thinking to get valuable insights. It is this thinking that the means to a competitive advantage will come from.
Determine the right data sources
The quality of the insights you get is only as good as the data you analyzed to get them. Avoiding bias is important because if you look for evidence to prove preconceptions you will likely find it. Organizations generate a lot of data every day and doing an audit of available data before considering additional data sources specific to the data analytics project is important.
The level of granularity to be applied is also an important consideration. When it comes to data analytics, optimization is critical. You may believe indexing your data over the smallest possible time intervals will glean better insights, but it could be detrimental to the project. Besides significantly increasing the amount of time it takes to get insights, it makes underlying trends highly susceptible to periodic outliers and anomalies.
This could result in insights that reveal threats or opportunities with no real long-term impact. Acting on them would waste valuable resources. At this stage, external data sources also need to be considered. These will be used for benchmarking purposes and understanding the industry-wide picture your data is painting. With data being such an important resource, a lot of effort has been put to building publicly accessible data resources. Datafloq provides a list of public data sets you can consider for your project.
Determine the right data for the tool
The analytics tools landscape has come very far. Many players now exist, creating tools specialized for different uses. The best development has been the simplification of analytics tools. Initially, one would have had to know every programming language to do any meaningful data analytics. Today though, tools are increasingly designed to cater for their everyday users, business professionals. There has also been growth in cloud-based tools. This, of course, eases the cost of analytics projects, by eliminating the need for massive storage and supercomputing on your part.
The human element is the most important element in turning data into insights. A tool should just be sufficient to process the necessary quantities of data and give the desired outputs. One should never get hung up on choosing the biggest tool out there, but the most functional one. For the longest time, people have been getting very good insights from just an excel spreadsheet.While applicability to the desired outcome is the first consideration, the cost is a close second.
Data analytics tools tend to be quite pricey. Quite a few open source tools exist, but they require additional programming to be usable and implementation is quite harder. But if you possess the right knowledge, open source tools are definitely worth considering. Besides being absolutely free, the resulting platform is a lot more customized to your end needs. Datafloq provides a list of open source data analytics tools you can consider for your project.
Deploy your data analytics project
Execution is everything. Even the best-laid plans will fall flat on their faces if they are not implemented properly. Executing a data project is an organisation wide exercise. Everyone who’s affected by the project should be involved in so far as possible. As a matter of fact, they need to be involved as early as the planning stage, but involvement in the implementation stage is most critical. Understanding the intimate workings of all the elements will enable them to better understand the outputs.
Bringing together the technical and end-user aspects of the data analytics project will help tweak the finer details that will determine what insights come out of the project. The formula is masterfully blending the expertise of the technical team with the business case knowledge of the end user.
Review, feedback and improve
A data analytics project must foster dialogue within the company. It should be the starting point for robust conversations about steps that can be taken towards the organization’s goals. This is also an opportunity to cast a wide net to gather input on the project. It is from this dialogue that insights will be gleaned from the underlying trends in the data. Individuals from across departments can give input on how these trends interact with their spheres of operation.
These inputs can then be collated into an organisation wide picture, insights that give guidance towards set goals. Armed with knowledge of what they can achieve with data analytics, participants can also give inputs on how to improve the analytics project. The operation should be fluid, taking into account changing circumstances within the organisation regularly. Changes can then be made, improve the project, then the process can start over.
Data analytics is no longer a luxury but an absolute necessity for your business. After being considered a fad by some, it has proven itself to have tangible benefits. All businesses, especially small ones cannot ignore its potential because they risk being left in the dark ages by their competition. Data and analytics have evolved much from their early days, and today they’ve been simplified enough to allow affordable and scale-able implementation in small businesses.
The key to successful implementation is a strategy that leverages existing data sources. The goal should always be to unearth insights, not just revealing trends in metrics. And keen insights are only possible where comprehensive sets of goals are set across the organisation.
- You can be a successful entrepreneur with limited business management skills – The Business Plan - March 15, 2018
- Important accounting ratios for small business financial analysis - March 14, 2018
- User-centric SEO strategy for user satisfaction and retention - March 9, 2018
- Why getting business funding early can do more harm than good - February 26, 2018
- Start a business with limited funding (Lean Start-Up) - January 17, 2018
- 8 trends that have made it easier to start a business - January 10, 2018
- Lessons from traditional store-fronts that can help websites convert better - January 2, 2018
- Improving agricultural yields with data analytics part 2 - December 29, 2017
- Improving agricultural yields with data analytics Part 1 - December 27, 2017
- How your business could benefit from an analytics consultant - December 23, 2017