“At our company, we derive 100% value from all our Advanced Analytical Efforts.”
If you were to respond to this statement in just a ‘Yes’ or a ‘No”, can you give us a resounding ‘YES!’, with absolute certainty?
Chances are, you cannot.
And you can be assured, you’re not the only one.
One of the biggest challenges that plagues the analytics industry is ensuring that you derive ROI from your analytical efforts. Agreed, Business analytics has redefined the traditional strategic decision-making process- But there are still avoidable mistakes that most enterprises are culpable of making that end up costing them millions of dollars and hundreds of man-hours in the long-run.
And this is exactly what we are here to tackle.
Here are five of the most common mistakes that companies make, that cause their analytical efforts to fail.
Not having a clear goal for analytics
This is one of the biggest challenges across industries, across functions, across the globe. Oftentimes, companies invest heavily in building their own analytics centers without really defining the objectives or even setting a goal. This can prove rather detrimental for enterprises for multiple reasons. Even just in terms of investment, it is a huge expense for companies.
According to this study by PwC and Iron Mountain, 75% of companies feel that they are making the most of their data, while only 43% believe they derive any tangible benefit from it and 23% feel like they receive no benefit whatsoever.
This is startling, considering that large enterprises spend several millions on analytics. This discrepancy is testament to the fact that companies fail to set clear, achievable objectives for their analytics units to deliver.
In today’s hyper competitive world, businesses need to constantly up their game in order to stay relevant and analytics technology is a gold mine that more and more companies will adopt. However, it all starts with having a clear business objective that is not only achievable but also measurable.
You can start off by answering simple but extremely crucial questions like – How can we use the data we have to identify and explore new opportunities? How can we scale our business using AI and analytics?
This also includes not thinking ‘long-term’ or having a sustainable plan for your analytics initiative.
A goal without a plan is just a wish, they say. So, unless you have a fully laid-out plan in place for the sustenance, continuity of your analytical efforts, chances are you aren’t going to taste success anytime soon.
Companies often make the mistake of looking at solving problems in the short-term, even if they are investment-heavy. While this may resolve the problem temporarily, there is every possibility for the problem to recur or for it to persist in other departments/areas. Strategic planning will help enterprises not only save on such expenses but also have a system in place that guarantees sustainable solutions.
Conventional wisdom will only offer so many use cases for analytical solutions. Typically, it makes sense to identify the most recurring problems and seek the help of advanced analytics to resolve them. While this may look like an opportunity to save tonnes of money and bring in efficiency to a great extent, there is always the possibility that there are other areas or analytical opportunities that will drive a higher ROI.
This is where executives need to always make room for mining and identifying other potential areas of improvement, even if they aren’t the most obvious ones. This way you not only keep the current analytical engine running, but also take the time to build a stronger and long-term approach towards resolving business pain-points.
Measuring, tracking and observing the impact of analytics also pays in the long-term because you will know what areas need more attention and which ones have achieved self-sufficiency.
We’ve discussed the impact measurement of analytics in detail below.
Not getting your fundamentals right
Not investing in the right team – read: interdisciplinary team
One of the biggest mistakes that companies make is to expect data analytics professionals to solve their business problems for them. No matter how qualified or powerful your team of data experts, solving business problems requires cross-functional expertise. Just like every cog plays a role for a wheel to successfully, complex business problems also require several experts to play their parts, individually.
It is also common for large enterprises to invest in an analytics capabilities and mass hire data scientists, without an actual assessment of how many data scientists are needed vs. other roles in analytics. While data scientists are most definitely crucial to any analytics team, it isn’t very wise to undermine the importance of all other roles.
The right team is the coming together of multiple core competencies and it varies case to case. For a successful analytics team, you will need multiple roles like AI & ML experts, data engineers, data analysts, business analysts, product experts, where each of them brings an irrefutable unique competence to the table which you don’t want to compromise on.
Underestimating the Problem at hand – Expecting quick solutions & results
This is, perhaps, the most common and least justified of all the mistakes enterprises make in their analytical journey. No matter how common or simple a business problem appears, it is wise to not assume that the solution will be simple and fast. Analytical solutions are as subjective as they are unique considering the umpteen factors that dictate them.
Recommendation engines are everywhere now, but that is not to say that it is a one-size-fits-all approach. Exhibit A – the classic Netflix example where they spent $1 million with over 2000 hours of effort by working on 107 different algorithms which resulted in their recommendation engine improving accuracy by 10%. But it never went live! Why, you ask? Because it simply wasn’t worth it.
So, while you may deem the problem to be straight-forward or plain, a lot of decisions go in to building a solution that is at once cost-effective, time-sensitive, effort-optimized, resource-optimized and so on.
Not investing in the right tools
There is undoubtedly a lot of dependency that analytics professionals have on tools. Even the best data analysts cannot do without the right tools. And understandably so. But a lot of times, we see that companies go about investing in tools that are either not the need of the hour for them or are not in line with their analytical maturity level. There is a plethora of tools available to enterprises today when it comes to aiding their analytical efforts, but a lot of their success depends on how wisely they choose their tools. Mistaking a popular tool to be the right one for you too or judging the efficacy of a tool by how expensive it is are some mistakes you want to avoid. Instead, take demos of as many variants of a tool as there are and juxtapose the offerings and your exact requirements to make a more practical decision. Remember – the right tool for you need not be the most popular or the most expensive. It just needs to be the best fit for your requirement.
Not tracking the impact of your analytical efforts
Alright you have a state-of-the-art analytics facility with the best minds solving your business challenges. But how often do you check on the bottom-line impact of your analytical efforts? You may have spent millions of dollars getting yourself an analytics center that is envy of every other enterprise, but without a regular impact assessment, your victory is short-lived.
As Harold S. Geneen rightly noted, “It is an immutable law in business that words are words, explanations are explanations, promises are promises – but only performance is reality.”
There’ll definitely be times when it may take a while for your analytical efforts to show impact, and this is ok as long as there is a business impact. When you have the right goal in place followed with the right team working closely to resolve the issue or achieve this goal, it is but a given that you track your efforts to know for yourself how efficient you have been, where to improve, etc.
Let’s not forget the golden rule – If you don’t measure, you cannot improve it.
This blog by IBM bigdata hub lists out 3 quick ways to measure your analytical success. Your analytical investment may be considered a success, by looking for these three factors.
- Quality – it allows you to make better decisions by optimizing KPIs and what-if situations
- Speed – it allows you to make faster decisions with optimization models and automation
- Robustness – it lets you make smarter decisions by combining decision optimization & ML
What’s important to note here is that there are just as many ways to measure the success of your analytical capabilities as there are ways to solve any given problem! Whether it is revenue impact your analytical solution or stakeholder satisfaction, value-driven impact or the performance of your models, there really is no one way to go about it.