ACAW study plan

Build five small workflows that each join two datasets using at least two different join types, and document which rows are kept or dropped in each.

Configure a union tool on mismatched schemas and manually reorder or rename fields so the output schema is deterministic.

Load flat files with inconsistent delimiters and encodings, then fix field parsing without editing the source file.

Write a one-page reference mapping each join type to a concrete business scenario you can recall cold.

Clean a deliberately messy dataset by trimming whitespace, normalizing case, and standardizing date formats using only transformation tools.

Use text-to-columns and regex-style parsing to split a combined address or name field into discrete, typed fields.

Convert between data types across string, numeric, date, and boolean without silent truncation, and verify counts before and after.

Build a reusable cleaning macro that accepts any input table and applies your standard cleanup steps.

Write conditional expressions that handle nulls, blanks, and mixed types without throwing errors.

Implement multi-condition logic using nested or chained conditionals and refactor one into a cleaner lookup-driven form.

Use string functions to extract, replace, and validate patterns such as emails or IDs.

Build date arithmetic expressions that compute durations, aging buckets, and period flags correctly across month and year boundaries.

Design a branching workflow that processes two data streams in parallel and rejoins them without row duplication.

Replace repeated tool blocks with an iterative or batch macro to eliminate copy-paste logic.

Add containers, annotations, and tool naming conventions so a stranger can read the workflow top to bottom.

Trace a single record through a multi-step workflow and confirm the path matches your documented logic.

Assign spatial objects to point data and compute distances or containment relationships in a small test dataset.

Build a trade-area or drive-time style analysis using spatial tools and summarize results by region.

Train and score a simple predictive model, then interpret the output fields and explain what each score means.

Compare model predictions against a holdout set and record accuracy or error metrics.

Build a standard macro with a control parameter and validate it against two different input schemas.

Create a batch macro that loops over a list of files or values and confirm it processes every item exactly once.

Chain two macros together and handle the interface contract between them without hardcoded field names.

Wrap a finished workflow in a macro that can be dropped into another workflow as a single tool.

Configure interface tools such as dropdowns, text boxes, and file browsers so they drive downstream tool settings.

Build an analytic app that takes user input, runs a workflow, and renders a formatted output table or chart.

Add input validation and friendly error messages for bad user entries.

Test the app as a non-technical user would and fix every confusing label or dead end.

Profile a slow workflow, identify the heaviest tools, and reduce runtime by filtering early and limiting data volume.

Add error-handling paths that catch failed records and route them to a review output instead of halting the run.

Set up logging or run metadata so failures can be diagnosed after the fact.

Complete two timed mixed-domain practice sets and review every missed item by domain.