A collection project is decided before the first line of code. Four questions, asked at the right moment, save weeks of back-and-forth.
Which sources, exactly
“Property sites” is not a scope. “SeLoger, leboncoin and Bien'ici, sale listings only, across three areas” is one.
A single stable site does not raise the same problems as ten mismatched ones. Budget follows the number of sources far more closely than the number of rows retrieved.
Which data, and for whom
List the fields you genuinely need. The ones added “just in case” cost design and maintenance time, and rarely get used.
Format comes next, and depends on who will open the result. We gave that question a whole article.
What volume, how often
A few thousand records once, or several million refreshed nightly. Different infrastructure, different budget, and changing your mind halfway through is expensive.
What quality on delivery
Decide in advance what has to be true. No duplicates? Normalised addresses? Prices always inclusive of tax? Those rules are set beforehand, never afterwards.
With those four points settled, the rest follows: building the extractor, hosting, delivery. Our pricing starts from this scope, and our method sets out the stages.