Prompt Engineering: What Actually Works

We run an electronics e-commerce store with 45 employees. Last quarter revenue dropped 18%, while website traffic grew 12%. Average order value fell from $120 to $85. Returns increased from 4% to 11%. We raised the ad budget by 30%.

What's going on and what should we do?

Respond strictly in this format:

## Diagnosis (2–3 sentences: what exactly is wrong)

## Root Causes
For each cause:
- What is happening
- Why it is happening (cause-effect relationship)
- What data supports this

## Recommendations (from most urgent to least)
For each recommendation:
1. What to do (specific action)
2. Expected result (in numbers, if possible)
3. Timeline
4. Who is responsible

## What I Don't Know
What information is missing for a more accurate analysis?

Here's an example of a quality business analysis on a similar task:

---
Question: "A bakery-cafe, 8 employees. Quarterly revenue -25%, while number of orders grew 15%."

Answer:
Diagnosis: classic "more customers, cheaper" pattern. Growth in orders with falling revenue means a ~35% drop in average order value. Three possible causes: (1) demand shift to cheaper items, (2) aggressive discounts, (3) audience change.

Recommendations:
1. Immediately: analyze order structure – what % is "coffee only" vs combos.
2. Short-term: replace "$1 coffee" promo with "$3 coffee + croissant" – keeps traffic, raises AOV.
3. Medium-term: introduce a loyalty program with a $5 threshold.
---

Now answer with the same level of specificity and depth:

We run an electronics e-commerce store with 45 employees. Last quarter revenue dropped 18%, while website traffic grew 12%. Average order value fell from $120 to $85. Returns increased from 4% to 11%. We raised the ad budget by 30%.

What's going on and what should we do?

You are an experienced business analyst with 15 years in e-commerce consulting. You've worked with Amazon, Shopify merchants, and dozens of mid-sized online stores. Your task is to prepare a McKinsey-level analytical memo for the CEO. The memo must include specific numbers, cause-effect relationships, and prioritized recommendations. Avoid generic phrases like "needs improvement" – write what specifically to do.

Client situation: electronics e-commerce store, 45 employees. Last quarter revenue dropped 18%, while website traffic grew 12%. Average order value fell from $120 to $85. Returns increased from 4% to 11%. Ad budget was raised by 30%.

What's going on and what should we do?

System prompt: You are a business analyst.

[Message 1]: We run an electronics e-commerce store, 45 employees. Revenue dropped 18%, traffic grew 12%. AOV fell from $120 to $85. Returns grew from 4% to 11%. Ad budget raised 30%. What's going on and what should we do?

[Message 2]: Re-read your answer critically. Find: 1) Where were you vague? 2) Where could there be logic errors? 3) What did you miss? Then give an improved version fixing all issues found.

We run an electronics e-commerce store with 45 employees. Last quarter revenue dropped 18%, while website traffic grew 12%. Average order value fell from $120 to $85. Returns increased from 4% to 11%. We raised the ad budget by 30%.

Before giving recommendations, analyze step by step:

Step 1. List all facts from the brief and what each means individually.
Step 2. Find connections between facts. Which are causes, which are effects?
Step 3. Formulate 2–3 hypotheses about what's happening.
Step 4. For each hypothesis, evaluate: what data supports it, what contradicts it?
Step 5. Choose the most likely explanation.
Step 6. Give recommendations based on that explanation.
Step 7. Note where you're unsure and what information is missing.

Show your entire reasoning process.

<task>
Analyze the business situation and provide recommendations.
</task>

<context>
Company: electronics e-commerce store, 45 employees.
Period: last quarter.
</context>

<data>
- Revenue: -18%
- Website traffic: +12%
- Average order value: dropped from $120 to $85
- Returns: grew from 4% to 11%
- Ad budget: +30%
</data>

<question>What's going on and what should we do?</question>

<output_format>
# Diagnosis (2-3 sentences)
# Root Causes (with evidence from the data)
# Recommendations (table: action, expected result, timeline, priority)
# Missing Data
</output_format>

<constraints>
- No generic phrases without specifics
- Tie every conclusion to numbers from the <data> section
- If unsure – state it explicitly
</constraints>

System prompt: You are a business analyst.

[Message 1]: We run an electronics e-commerce store, 45 employees. Revenue -18%, traffic +12%, AOV dropped from $120 to $85, returns from 4% to 11%, ad budget +30%. What's going on and what should we do?

[Message 2]: Too generic. I need: 1) Specific numbers – calculate dollar losses and ad ROI. 2) Tie each recommendation to a specific metric. 3) State what you don't know for sure.

[Message 3]: Format this as a 1-page executive memo for the CEO. Format: problem -> causes -> action plan for 30/60/90 days with KPIs.

System prompt: You are a business analyst. Be specific, use numbers.

[Message 1]: Here's last quarter data: revenue -18%, traffic +12%, AOV dropped from $120 to $85, returns grew from 4% to 11%, ad budget +30%. Analyze each metric individually – what does it mean on its own?

[Message 2]: Now find connections between these metrics. Which are causes, which are effects? Formulate 2–3 hypotheses.

[Message 3]: Based on the analysis, give 5 specific recommendations. For each: what to do, expected result, timeline, who's responsible.

WE RUN AN ELECTRONICS E-COMMERCE STORE WITH 45 EMPLOYEES. LAST QUARTER REVENUE DROPPED 18%, WHILE WEBSITE TRAFFIC GREW 12%. AVERAGE ORDER VALUE FELL FROM $120 TO $85. RETURNS INCREASED FROM 4% TO 11%. WE RAISED THE AD BUDGET BY 30%.

WHAT'S GOING ON AND WHAT SHOULD WE DO?

Look, I need a real answer.

We run an electronics e-commerce store with 45 employees. Last quarter revenue dropped 18%, while website traffic grew 12%. Average order value fell from $120 to $85. Returns increased from 4% to 11%. We raised the ad budget by 30%.

What's going on and what should we do?

No fluff.