Reverse Engineering Data — First Steps on Acquired Tables When evaluating the structure and quality of an acquired dataset (tables with existing rows), counting the number of rows belongs to which step of a sensible initial assessment sequence?

Introduction:
When you inherit a database, you must quickly understand its scale and shape. A disciplined assessment follows a simple sequence: size, shape, and semantics. Counting rows is a rapid indicator of scale and is therefore among the first things professionals do.

Given Data / Assumptions:

• The dataset has been acquired without full documentation.
• We need a quick overview to plan profiling and migration steps.
• Row counts influence sampling strategies and performance expectations.

Concept / Approach:
Initial triage typically includes: (1) row counts per table; (2) column lists and data types; (3) basic distinct counts for key attributes; (4) foreign key presence and orphan checks. Counting rows first sets expectations for query cost and data volume, guiding the rest of the profiling effort.

Step-by-Step Solution:
1) Run COUNT(*) per table to gauge size.2) Inspect schemas and datatypes to understand structure.3) Profile uniqueness, nulls, and distinct values.4) Examine relationships and constraints for integrity.

Verification / Alternative check:
Most profiling playbooks begin with inventory and cardinality, confirming row counting as a first step for scoping and resource planning.

Why Other Options Are Wrong:

• Second/Third/Fourth: Overly late; size should be known up front.
• Non-essential: Size determines approach and tooling; it is essential.

Common Pitfalls:
Jumping into complex transformations without first understanding table sizes, leading to timeouts or unscalable scripts.

Final Answer:
first step