Is there a future for manual (non-machine) data collection?
Tearsheet's Challengers Conference 2020 - Sept 24, 2020
A lot of different information can be obtained using machine data collection, but it seems ghostly that this is real. Let's say, collecting information on a heap of organizations of a certain subject.
Take, say, 1000 organizations in the subject of investment funds.
There are a lot of variables that need to be collected: date of foundation, stages of investment, declared capital, country of head office , investment spheres, etc.
It is clear that we cannot find everything by a machine method. A lot of it is hidden inside the site, in the context of the offers, on the social networks of these companies. Some of the information will have to be collected using third-party sites (Wikipedia, Webarchive and others).
Is it true that in any case manual collection of information will remain relevant?
In our work we use a mixed type of information collection: more manual, but partially machine. Therefore, I wonder whether it is worth making big bets on technology?