Data Analysts

Clean Output Can Hide Expensive Input

Two clients can ask for the same thing: “Build a monthly performance dashboard.”

Client A has a reliable warehouse, agreed definitions, and one decision-maker. Client B sends CSV exports from three tools, has never reconciled revenue, and introduces a new stakeholder after the first review. The finished dashboards may contain the same six charts. They are not the same project.

Take an illustrative $3,000 fixed-fee engagement. You remember 20 hours spent writing queries, checking the model, and building the dashboard, which looks like $150 an hour. Then you count the less visible work:

• 4 hours inspecting exports and tracing field definitions
• 6 hours cleaning and joining inconsistent data
• 3 hours in discovery and stakeholder calls
• 3 hours documenting assumptions and edge cases
• 4 hours revising metrics after feedback
• 2 hours packaging files, handoff notes, and follow-up

The real total is 42 hours. The illustrative effective rate is about $71 an hour.

This does not mean fixed-fee analytics work is a mistake. It means “analysis” is too broad to price without knowing where your time goes.

Scope creep often enters through a question that sounds harmless: “Can we also segment this by region?” The segmentation may take ten minutes. Or it may expose inconsistent location data, require a new mapping table, and change every chart. The request is not the problem. Agreeing before checking its cost is.

A running project total gives you a factual pause. You can say the new cut fits, trade it for something else, or add it to scope. Without the record, the clean dashboard makes the extra work disappear.