The task to crawl Singaporeans on Linkedin was dispatched to a programmer two weeks ago. But up till yesterday, it was not close to being done yet. The programmer was still extracting everything he can from the profile.
There lies the problem(s)
The first is obviously in my specifications. I did not scope the task explicitly, so the programmer took to himself to deliver beyond expectations --- to extract as much information as he can from a Linkedin profile.
The first mistake is my laze, but the second mistake is what I want to talk about – that no one at Nubela should try to get things to perfection as a first step. Get it to work first, then make it good. In other words, one step at a time.
Business is an iterative process
This principle applies in most business processes from product to sales pitches (with PR and marketing being the exception).
Because nothing is "correct" from the get-go because no one is smart enough to know what people want. Even people do not know what they want.
From my experience, great products start with a good product. And a good product is the result of a product that was launched fast, and polished over time with feedback from the market.
Back to basics
So the CTO and I got on a video call with the programmer (because remote offices, yo) and figured that the issue with the slow progress was that our programmer tried to overreach.
So instead of delaying any longer, we reduced the scope of the task. And the task was done that very day.
Until we launch the product, we do not try to be perfect — one step at a time.