Definition
Eyeballing is an informal term used widely in many areas involving Data Science . Its meaning is to obtain a rough sense or estimate of something using no quantitative (numerical or mathematical tools) but only a visual inspection of tabulated or graphical data.
The advantage of eyeballing is the speed with which an assessment can be obtained along with its intrinsic focus on "ball park" figures and assessments (zero-th order approximations). Besides inaccuracy, eyeballing is also subject to omission errors and potential biases.
A more formal expression of eyeballing is found in Exploratory Data Analysis which uses a combination of visual and numerical tools to provide more structured insights
 - Shopify)