None of us fully understands what we work on. The illusion that we can control the complexity of systems is a deadly trap — and when I’m hiring, I make sure the candidate isn’t carrying that bias into my team.
So I always run a role play: I pull up a sample ecommerce dataset and ask: why has it been selling less than six months ago?
The first answers always come fast and confident:
- less traffic
- ad fatigue
- algorithm change
Fair enough. Then I start digging:
- Which traffic source dropped?
- What path did that traffic take through the site?
- Where in the funnel are you losing visitors? Product page, checkout, pricing, shipping?
Then we go deeper: could it be a site speed issue? How do you measure it? How do you optimise code for performance? How do you optimise the server? What happens if you get a 10x traffic spike?
At some point, the answer becomes “I don’t know.”
And that’s where the real interview begins.
I’m not looking for someone who knows everything. I’m looking for someone who knows where their knowledge ends — and isn’t afraid to say so.