How to ace your first data hire
When you don't know what good looks like
A challenge I’ve seen many companies struggle with is how to hire a 🌟 standout person 🌟 for their first ‘data’ role … when no one currently in the company knows what ‘good’ looks like.
How on earth do we pick who to start with? 🤷♀️ And because talent attracts talent, early hires are key.
It’s a challenge made harder by the fact that ‘data’ in all its multi-dimensional glory (from AI products, to stream processing, to management reporting and everything in between) is a young profession with none of the standard qualifications, industry associations, certifications, etc that many use as a shorthand for ‘good’ in other areas.
Much depends on the current state, industry, growth prospects, etc of your business so there’s no ‘one size fits all' answer. But let’s make a start:
Hire someone more senior than you think you need
If data is important to your business, then data needs representation at your leadership table. While good data people will have some software engineering background, software engineers are NOT by and large particularly expert in defining, collecting, structuring and storing the data needed to build compelling data products, let alone AI products. So hiring someone with 4-6 years data experience, popping them next to your eng team and expecting emergent value is asking for trouble.
Hooking up the latest OpenAI LLM via API doesn’t count BTW! AI products need soup-to-nuts thinking to bring sustainable business value. AIaaS solutions might be an important part but it will be a long time before they approach being a stand alone, drop-in piece of your customer facing product for a team that knows nothing about experimentation and measurement.
Look for someone who is very interested in the breadth, depth and connectedness of your existing operational and behavioural data
Good folks want to continue to be impactful and they know this will come hard and company patience will wear very thin if they have to start from a scrappy, incoherent base. Every company wants the icing on top, and preferably yesterday. Good folk know that they need to start with at least the ingredients for a good cake first. 🍰
Make sure your prospective hire can hold a strong conversation with engineering and product and customer experience / service
Ensuring credibility with the engineering team means you’ll avoid folks whose experience is limited to spreadsheet modeling and powerpoint presentations. While you want someone who can pull together your strategic approach to data and data products, you don’t want to hire someone who can write a plausible strategy - and then wants to hand it off to someone else to execute. Similarly, if your prospective hire isn’t eager to talk to both your product peeps and the folks who deal with customer onboarding and complaints then they probably haven’t shipped anything to production and supported it through first contact with the customer.