Prioritizing Ad-Hoc Requests
Part of the Data Analyst Roadmap
In short
Once people know you can pull data, requests arrive on chat, email and in the corridor. Without a way to prioritise, you'll be busy all day and still miss what matters.
Analysts are flooded with incoming requests — a lightweight framework (urgency, impact, effort) keeps you from either saying yes to everything or becoming a bottleneck.

Think of it like this
The emergency ward nurse
In a busy hospital, the triage nurse doesn't treat patients in order of arrival. A chest pain goes first, a broken arm next, a mild cold waits. She sorts by urgency and impact in seconds. You can sort data requests the same way.
Watch it happen
Scoring this week's requests
1/5Five requests land in one morning, in this order. Doing them first-come-first-served would be a mistake.
The big ideas
What you need to know
Impact first
Ask how much money, how many users or how big a decision depends on the answer.
Urgency is not loudness
The person who messages five times isn't always the most urgent. Ask when the decision is actually being made.
Do tiny things at once
Anything under ten minutes that unblocks someone is often worth doing immediately instead of tracking it.
Keep one visible queue
A shared sheet or board of requests lets people see where they stand and stops side-channel pressure.
See it in code
A tiny example
1A: Request B: Impact(1-5) C: Urgency(1-5) D: Effort(hrs)2E2 =B2*C2/D23F1 =SORT(A2:E10, 5, FALSE)A simple scoring sheet. Fill impact, urgency and effort for each request, then sort by the score in column E.
Save yourself the pain
Common mistakes
Oops
Working on whatever was asked most recently or most loudly.
Do this instead
Score every request on impact, urgency and effort, then work from the top.
Oops
Silently dropping low-priority requests.
Do this instead
Tell the requester where it sits in the queue and when they can expect it, or suggest a self-serve option.
Your mission
Triage your own to-do list
Quick check
A senior manager wants a "quick number" that needs a full day of cleaning. Another team needs a 5-minute fix that unblocks their launch today. What do you do?
Go deeper
- 1Design a simple triage framework for incoming requests (urgent/important quadrant works).
- 2Practice saying 'I can get to this by X, or Y sooner if we drop Z' instead of just yes.
- 3Learn to batch similar small requests instead of context-switching constantly.
More topics at this stop
Stuck on this topic? Ask an Insider
Get 1:1 guidance from people who've walked this exact path — on the InsideEdge app.