Customer Service

Rewrite my draft response to an angry customer. My draft is too defensive — I need it to be empathetic, solution-focused, and professional without being robotic.

Customer complaint (via email):
"I have been waiting 3 weeks for my refund and nobody is giving me a straight answer. I called twice, got put on hold for 20 minutes each time, and the last agent told me to just wait. This is unacceptable. I want my money back TODAY or I am going to Bank Negara and posting this everywhere."

My draft response:
"Dear customer, we apologise for the inconvenience. Your refund is being processed. Please allow 7-14 business days for the refund to appear in your account. If you have further questions, please contact our support team."

Rewrite rules:
1. Acknowledge their specific frustration (3 weeks, hold times, runaround) — do not genericize it
2. Take ownership without blaming "the system" or "the process"
3. Give a specific timeline and a specific person who will follow up (use "I" not "we")
4. Address the threat without escalating — do not mention Bank Negara in the response
5. Under 150 words
6. Tone: a real person who cares, not a corporate template
7. Include one concrete action they can take if the timeline is not met (direct phone number or escalation path)

Analyse these 15 recent support tickets and create an FAQ page that would have prevented at least 60% of them.

Recent tickets:
1. "How do I change my password?" (asked 12 times this month)
2. "My invoice shows the wrong company name"
3. "Can I add more users to my plan?"
4. "The app is not loading on my phone"
5. "What happens to my data if I cancel?"
6. "How do I export my reports to Excel?"
7. "Why was I charged twice this month?" (SST added separately)
8. "Can I integrate with SQL Accounting?"
9. "My team member cannot see the dashboard"
10. "How do I upgrade from Basic to Pro?"
11. "The system is showing dates in American format (MM/DD)"
12. "I forgot which email I used to sign up"
13. "Can I use this on iPad?"
14. "How do I add my company logo to invoices?"
15. "Your chatbot could not help me — how do I speak to a real person?"

Create:
1. Group tickets into 5-6 FAQ categories
2. For each FAQ:
   - Question (as the customer would phrase it, not how we would)
   - Answer (under 100 words, step-by-step where needed)
   - Screenshot placeholder note (what to capture)
   - Preventive measure: how to reduce this question at the source (UX fix, onboarding change, or clearer labelling)
3. Priority ranking: which 5 FAQs would prevent the most tickets?
4. The one product improvement that would eliminate an entire FAQ category

Build an escalation decision tree for my customer service team. We handle 200 tickets/day and currently every hard question gets escalated to the team lead, who is overwhelmed.

Our product: SaaS platform for managing employee claims and reimbursements
Team structure: 6 agents (Tier 1), 1 team lead (Tier 2), 1 CS manager (Tier 3)
Average ticket complexity split: 70% simple, 20% moderate, 10% complex

Build the decision tree covering these scenarios:
1. Billing disputes (wrong charge, double charge, refund request)
2. Technical issues (app not loading, data not syncing, login problems)
3. Feature requests ("can your system do X?")
4. Compliance questions ("does this meet LHDN requirements?")
5. Account cancellation requests
6. Angry/threatening customers
7. Data privacy requests (PDPA — "delete my data")
8. Partnership or sales enquiries that come through support

For each scenario:
- Who handles it (Tier 1, 2, or 3)
- What Tier 1 can do without escalating (specific authority limits)
- Trigger for escalation (exact condition, not "if it seems complicated")
- Maximum response time SLA
- Template response or first message script

Also include:
- Dollar thresholds: Tier 1 can approve refunds up to RM200, Tier 2 up to RM1,000, Tier 3 for anything above
- A "de-escalation script" for angry customers that Tier 1 should try BEFORE escalating

Write a 3-email win-back sequence for customers who cancelled their subscription 30 days ago. These are not angry customers — they just quietly churned.

Product: Project management software for Malaysian SMEs (RM199/month)
Average customer tenure before churn: 5 months
Top 3 churn reasons from exit surveys:
1. "We did not use it enough to justify the cost" (40%)
2. "Switched to a competitor" (25%)
3. "Budget cuts" (20%)

Email cadence: Day 30 (after cancel), Day 45, Day 60

For each email:
- Subject line (under 50 characters)
- Body (under 120 words)
- The specific approach:
  - Email 1: Acknowledge and ask what went wrong (pure learning, no selling)
  - Email 2: Share what has changed since they left (new features, improvements)
  - Email 3: Make a specific come-back offer (2 months at 50% off) with a deadline
- One psychological principle behind each email

Constraints:
- Do not sound desperate
- Do not promise things will be "completely different" — be honest about what changed
- Include a genuine "we hope you found a great solution" sentiment — not passive-aggressive
- The Day 60 email should create a clean break if they do not respond (close the loop gracefully)

Redesign our customer satisfaction survey. Our current survey gets a 4% response rate, which makes the data useless.

Current survey (sent after every support ticket closure):
1. "How satisfied were you with your support experience?" (1-5 scale)
2. "Would you recommend us to a friend?" (NPS 0-10)
3. "Any additional comments?" (open text)
4. "How could we improve?" (open text)

Problems: Too many questions, sent every time (survey fatigue), generic, open text fields are mostly empty.

Redesign with:
1. A new survey structure (maximum 2 questions)
2. Smart triggering rules (when to send, when NOT to send)
3. Dynamic follow-up: different second question based on the first answer
4. Micro-survey option: 1-click rating embedded in the ticket closure email (no separate survey page)
5. Quarterly deep-dive survey (longer, but for a random 10% of customers)
6. How to get to a 20%+ response rate
7. What to measure instead of just CSAT — leading indicators of churn and delight
8. Reporting template: how to present CSAT data to leadership monthly (not just "our score is 4.1")

Constraint: We use Freshdesk — the solution must work within its survey capabilities or be easily built alongside it.

Write a knowledge base article for this common customer issue. It must be clear enough that a non-technical user can follow it without calling support.

Issue: "I accidentally deleted an employee record and need to restore it"

Product context: HR management platform, the delete action moves records to a "trash" folder for 30 days before permanent deletion. Most customers do not know the trash folder exists.

Article structure:
1. Title (as a customer would search for it — not technical jargon)
2. One-sentence summary
3. Step-by-step instructions with:
   - Numbered steps (maximum 8 steps)
   - What the user will see at each step (describe the screen, no screenshots needed yet)
   - A callout box for the most common mistake at each step
4. "What if the record is permanently deleted?" section (after 30 days — explain the manual recovery process: contact support with employee ID)
5. Prevention tips: how to avoid accidental deletion (require confirmation, restrict delete permissions)
6. Related articles (suggest 3 topics that someone reading this would also need)

Constraints:
- Write at a Year 10 reading level
- No "simply click" or "just navigate to" — these words minimize difficulty and frustrate users who are stuck
- No assumed knowledge — spell out where every button is
- Test: Could someone with zero training follow these steps and succeed?

Create a library of 10 response macros for our customer support team. These are pre-written responses that agents can personalise and send quickly.

Product: Cloud accounting software for Malaysian SMEs
Support channels: Email, live chat, WhatsApp

Create macros for these 10 scenarios:
1. First response to a new ticket (acknowledgement + set expectations)
2. Requesting more information to troubleshoot
3. Issue resolved — closing ticket
4. Known bug — timeline for fix
5. Feature request — logging it
6. Billing enquiry — SST explanation
7. Refund approval
8. Refund denial (with alternative offered)
9. Transferring to a specialist
10. Customer threatening to cancel

For each macro:
- The template text (under 80 words for chat, under 120 words for email)
- Personalisation points marked with [BRACKETS] — agent fills these in
- Tone notes: how to adjust for angry vs neutral vs happy customers
- What NOT to say in this scenario
- When to deviate from the macro entirely and write a custom response

Constraints:
- Never say "I understand your frustration" (overused, sounds insincere)
- Every macro must have a clear next step for the customer
- Chat macros must be shorter than email macros
- Include one sentence that sounds human, not corporate, in every macro

Map the friction points in our customer journey based on support ticket data. I want to find where our product is failing customers, not where support is failing.

Support ticket breakdown by topic (last 3 months, 2,400 tickets total):
- Onboarding/setup: 480 tickets (20%)
- Billing/payments: 360 tickets (15%)
- Feature confusion (how do I do X?): 600 tickets (25%)
- Bugs/errors: 312 tickets (13%)
- Integration issues: 192 tickets (8%)
- Account/access problems: 264 tickets (11%)
- Feature requests: 120 tickets (5%)
- Cancellation: 72 tickets (3%)

For each category:
1. What this ticket volume tells us about the product (not about support quality)
2. The likely customer emotion when they submit this ticket
3. Top 3 specific product or UX changes that would reduce this category by 50%
4. Quick win: one change implementable in under 1 week
5. Cost of NOT fixing this (estimate in churn risk and support cost)

Then create:
- A prioritised roadmap: which category to fix first based on effort vs impact
- The one insight that leadership probably does not know because they never read support tickets
- A monthly "Voice of Customer" report template I can send to the product team