Chapter 9 · Data Analyst
From analysis to influence
~6 min read
This is the chapter most self-taught analysts skip, and it is the one that most determines whether you get promoted. A correct analysis that does not change a decision was, in business terms, a waste of time. Your job ends when someone acts on what you found.
9.1 Lead with the answer#
Executives do not want your journey; they want the conclusion, then the support if they ask. Structure findings as the takeaway, then the evidence, then the recommendation. Never make a busy reader hunt for the point through twelve slides of methodology.
9.2 Tailor depth to the audience#
| Audience | Wants | Give them |
|---|---|---|
| Executives | The decision | One line: finding plus recommendation plus impact |
| Product and marketing peers | The why and next steps | Key charts plus the driving segment |
| Data and engineering colleagues | Method and rigor | Query logic, caveats, assumptions |
9.3 Always state your caveats#
Trust compounds. Naming the limitations of your analysis, such as a small sample, missing data, or an assumption you made, makes people believe the parts you are confident about. Hiding caveats and getting caught destroys credibility permanently. Senior analysts volunteer their uncertainty, because it is a strength signal, not a weakness.
9.4 Managing stakeholders and requests#
- Clarify the decision first. Ask what will change based on this number. A request for a raw export often hides a decision that a focused analysis would serve far better.
- Scope and prioritize openly. When a stakeholder wants everything on one dashboard, split distinct needs into separate views and use an effort-versus-impact lens to agree what ships first. Saying yes to everything is not helpful; it signals no prioritization.
- Set expectations on time and confidence. A quick directional answer and a rigorous audited one are different deliverables. Say which you are giving.
9.5 A worked example, end to end#
A marketing lead asks for a spreadsheet of every signup last quarter. Instead of exporting 40,000 rows, you ask what they are trying to decide. The real question is which channels to fund next quarter. You reframe it, pull signups joined to their acquisition channel and 30-day retention, and find that one paid channel drives high volume but the worst retention. You present a single chart: channels ranked by retained users, not raw signups. The recommendation writes itself, and you have saved the team from spending into a leaky channel. That reframing, from a literal export to a decision, is the entire value of the role in miniature.
Get the next chapter and weekly interview tips by email
One short email per week. Skim in a minute. Unsubscribe anytime.