Data. Everyone is at it. And you should probably do it too. Before that happens you need to understand where the data is coming from, what it can tell you, how it can make a real impact on your business and how you are going to start the process.
Things can grind to a halt when the AI you employ tries to work with incomplete data, low quality data or badly structured data. If you start the more advanced part of the process before you start collecting and cleaning data, working out what data and why you need it and what you will need it for, then you are just throwing resources away and your timelines will start to stretch further and further into the future.
There is a lot of pressure for companies to start running to the finish line, spooked by their competitors, but in this story you will want to be the tortoise rather than the hare, taking it steady at first, to make sure you have the pieces you need in place before moving onto the latter stages.
Hiring a team before you have an infrastructure plan in place is madness. Fresh data scientists being thrown in at the deep end will not only fail at a thankless task, but they will become alienated, the project will fall behind, the delays will drain more budget and you will be where you started: a company unable to mine value from its data.
It also runs the risk of souring the C-Suite on the idea of data as a worthwhile pursuit. They threw money at the problem, data people were hired, but there is now nothing to show for it. What gives?
To avoid this frustration, a clear understanding must be sought about who is needed and at what stages. Specialist recruiters can be brought in to build a team from the ground up, with realistic time frames, and a knowledge of who is needed when. They will understand that instead of cramming a job spec with every skill you can think of, the description should be targeted for the specific role. There are a diverse set of skills out there in the workforce to be tapped into. People who could be perfect for the task you require should not be dismissed out of hand. If you are requiring someone have a number of years’ experience in something which, until 5 years ago, was virtually unknown, then you are damaging your chances of filling that role.
Before you have an answer, you must have a question, right? That is how it works in the world and it is also how it works within Data. Sometimes it is a ‘chicken and the egg’ situation: figuring out what you need done, then working out if you have, or can obtain, the data to get it done, or figuring out the data and working out whether it has value for your future. Either way, this groundwork must be completed before the hiring of the full team. You may hire in a consultant to create the plan, but it cannot be fully left to those with no, or very little, data experience. This is a time when core business problems are discussed and evaluated, while data possibilities are floated.
A good starting point is to educate everyone within the company about the importance of data. This can help, as the suggestions of the staff themselves, when it comes to problems that can conceivably be solved by data, can really drive change.
Understanding that data can be cross-departmental is another step forward. There is a habit to view this as just an IT issue, but data should be coming from multiple departments within your company, so there needs to be cooperation in order to make sure it is correctly generated and stored and can be used in a timely fashion.
Knowing where your data is actually stored and what kind of state it is in, is a great start. So many companies have it stored hither and thither, or just plain don’t know where it is because they have not had to think about it since it was created.
At all stages there should be an understanding that this will positively impact the business in real terms. There must be business value in the data if this is to be a worthwhile endeavour.
Curious about how Zenshin Talent can help your organisation? Contact us today for a no-strings conversation about your needs and our experience.