A career in data tends to attract people with a certain kind of mindset – the one that sees thousands of rows of data and thinks, “Hell yes, I want some of that!”
But, no matter how much you love data, there is an argument to be had as to how much of your time should be spent on extraction rather than insight. You’d be hard pressed to find any data enthusiast who disagrees with the benefits of automation, but the real challenge is convincing brands that the results justify the set-up costs.
Efficiencies in disguise
Many analysts spend the bulk of their time pulling and structuring data, which will come as a surprise to no one, I’m sure. This is a tale as old as time, except the real value from our work isn’t in the pull; it’s in the insights – identifying what the data tells us and translating that into business strategy.
Fortunately, there are many automated tools and processes that can enable us to spend more time finding strategic insights.
But now we’ve found ourselves in a quandary. After years of perfecting data delivery for clients, these new automated solutions change the game entirely. However, automation can come with steep set-up costs and lots of processes; and let’s not forget, while automation makes the process more efficient, the ROI is not immediately evident.
It’s a Catch-22. You can’t prove the efficiency and results of your work until you’ve done it, but a client is unlikely to sign off without proof of the results – which you can’t get because you haven’t done the work… you get the idea.
It can be a difficult sell. There, I said it!
So, how do you justify to your clients that they should spend more money by investing in automation when they already receive the reports that they want?
Flip it and reverse it
To me, there are two obvious reasons for a lack of commitment: firstly, stakeholders are having difficulty seeing the value of automation due to the up-front investment required and the length of time for them to see a return. Secondly, it’s the fear of making the wrong long-term decision and ending up in some kind of ‘blame game’.
Getting around the fear issue can be tricky. Regardless of how effective your sales skills are, overcoming a brand’s indecisiveness about investing in automation is going to be the greatest challenge.
So, you address that by working to demonstrate value from day one.
And how do you show value to stakeholders? In theory, the answer is simple: prove that automation will affect their ROI. And the proof is often found in the proverbial data pudding. A lot of people see automation of data and analysis as a back-end process that isn’t really related to ROI. However, a lot of people are wrong.
When you automate the data crunching and extraction, you can find patterns that were previously hidden from human analysis. This is thanks, in part, to the fact that our ability to recognise patterns is limited when compared to programs that use machine learning – and the fact that people also need to sleep.
Take these insights to your client and show them the benefits. Explain automation and machine learning to them and why it would be impossible for a human being to discover these valuable nuggets in the time and budget available.
Next, show how investing in automation will impact ROI by breaking down their spend. Present your case thus:
If we spend £X amount per year on data pulls and £Y amount per year on insight generation it adds up to the budget. Now, if we add in automation (A) it’ll be £X+£Y+£A = £Z, likely a higher budget initially.
But with the addition of automation, we can effectively flip the budget so that over time X is reduced, A is eliminated post-set-up and the bulk of the spend is now focused on Y.
I’m not a mathematician, so excuse my crude formula. However, the point is that time spent on data pulls will decrease and more time can be spent on improved insights – and this will drive better results for the brand. Be sure to stress that the insights and therefore the strategy will improve thanks to automation.
If you show results from the data and a rational change in budget split, you can then highlight to the client how this whole process will help them reach their targets.
Taking the longer view
This still doesn’t mean that selling automation will necessarily become easy. The fact is most businesses are interested in targets, not platforms – and, to reiterate, especially when there’s no immediate ROI. However, you are selling the promise of greater efficiencies that will generate returns in the longer-term. In the meantime, the importance of driving efficiencies should not be underestimated, especially when you consider the forecasts for data growth.
Around three-quarters of the world’s population will be connected in the next five years and the amount of data we consume grows every year. To place this in context, IDC calculated that total global data stood at 33 zettabytes last year, but predicted it will grow to 175 zettabytes by 2025.
Machine learning and automation will become ever more important in terms of helping us make sense of an ever more complex world. As data sets grow ever larger, automation is – realistically – the only way analysts will be able to dedicate sufficient time to the consultancy that will drive clients’ growth. At least without having to radically increase the size of teams.
To break it down into something quantifiable, typically, the ratio of time allocated to delivery as compared to consultancy is around an 80/20 split. Automation can reverse those figures – and let’s not forget that time is money. Automation may not be cheap, but analysts aren’t either.
