liveBook · Manning

Data collection

As explained in Data for All

Data collection is the systematic gathering of information from actions, observations, systems, or other sources.

What can someone learn from where you go, what you browse, and what you buy? Each activity creates data. Collecting those records gives an organization information it can store, combine, analyze, and use—sometimes long after the activity itself. section 2.7

A location record can reveal movements and suggest habits. Add browsing and transaction records, and a company can draw a fuller picture of a person’s behavior. Collection is the starting point: analytics depends on having enough data, from enough sources, to model situations and predict what someone might do next. section 2.4.6 section 2.7

An offer at the right moment—for whom?

A company may combine information you provide with records about local conditions, then predict when an offer will appeal to you. The offer may arrive at a vulnerable moment. Good timing does not mean it serves your interests: the company may want you to spend when reducing debt would help you more. section 2.5

This scale of collection is relatively new. Earlier generations lived with data that was more local, static, and difficult to share. Today, online actions are recorded routinely, and organizations can integrate records from many places. They collect data for purposes including improving services, research, pricing, and advertising. Those purposes are not all the same, and neither are their effects on you. section 1.0 section 3.4

Commercial incentives matter. In 2020, advertising supplied 80 percent of Google’s reported revenue and 98 percent of Facebook’s. Narrowly targeted advertising gives companies a strong reason to keep collecting data. Collection also matters outside advertising: Australian regulators examined how combining Fitbit health information with Google’s data could strengthen Google’s market position. section 4.6 section 4.7.2

You do not have to treat every request for data as inevitable. Ask what is collected, what other records it may be joined with, and whose goals the resulting predictions serve. Individual choices matter, but changing the wider system also calls for governments, watchdogs, and new business models. section 2.5.1 section 4.6

How Data for All covers it

How deep it goes
Explains sources, combinations, incentives, and consequences without specifying collection procedures or technical implementations.
Examples
Uses location, browsing, and transaction records, a targeted offer, and the proposed Google–Fitbit combination.
What it assumes
Starts with familiar online and offline actions rather than assuming knowledge of data systems.
What you can do after
Evaluate/understand it by questioning what organizations collect, how they combine it, and whose interests the results serve.
How it connects
Connects data collection to data integration, analytics, behavioral prediction, targeted advertising, privacy, and competition regulation.

Also explained in

Compare how 10 books explain data collection →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Data for All. The section links point to the parts of the book it draws on.