Playbook Social Customer Service

Social Media Customer Service Playbook

Use When

• Build or improve a repeatable Social Customer Service workflow for a client or delivery team.
• Turn an approved objective into roles, controls, handoffs and measurable actions.

Do Not Use When

• The task is a single-channel presence plan; use the closest platform-* skill.
• The task is upstream positioning or channel choice; use the closest strategy-* skill.

Required Inputs

Capability and Permission Boundaries

Read supplied artefacts and search relevant evidence. Treat review, audit and planning as read-only. Editing the requested draft is allowed; publishing, messaging, production changes, personal-data processing, spending, destructive actions and certification claims require explicit authority. Use network access only for authorised verification.

Degraded Mode

If accounts, files, network, rendering or current evidence are unavailable, return the narrowest useful qualified Social Customer Service playbook plus an evidence-gap list. Mark each unavailable check not assessed; never convert it into a pass.

Decision Rules

Workflow

1. Confirm the consumer, objective, market, decision owner and permission boundary; stop if the objective or owner is missing.
2. Inspect supplied evidence and verify volatile claims; record missing inputs rather than filling them with assumptions.
3. Apply the decision rules, preserve useful existing material and draft the Social Customer Service playbook.
4. Test each action against platform, privacy, safeguarding, brand and approval constraints; stop and escalate a blocking risk.
5. Run the quality and anti-slop gates. If a check fails, correct the draft and rerun it before handoff.

Outputs

Evidence Produced

Quality Standards

Use British English and the specified market context. Recommendations must be executable with the stated capacity, current claims must be verified or qualified, and acceptance conditions must be observable. A worked example must use a labelled scenario, not fabricated client evidence.

Anti-Patterns

• Inventing a client fact, benchmark, budget or approval. Fix: cite the source or label the assumption and its effect.
• Copying one channel or client pattern unchanged. Fix: tie each choice to the named audience, objective and evidence.
• Stating volatile platform or legal details from memory. Fix: verify the current official source or omit the claim.
• Treating an inaccessible account, file or metric as healthy. Fix: mark it not assessed and bound the conclusion.
• Publishing, spending, messaging or changing production state from planning authority. Fix: obtain explicit action authority.
• Delivering actions without owner, timing or acceptance. Fix: assign all three or return the item as an unresolved gap.

References

• Anti-AI-slop production gate
• East African English standard
• Use the directly cited sources and companion skills in the domain guidance below; verify time-sensitive claims before use.

Required Input

Ask for the following before generating the playbook:

1. Client business name — trading name as it appears on social media
2. Industry — e.g. retail, F&B, logistics, hospitality, professional services
3. Country/city — default is Uganda; specify if different
4. Primary goal — e.g. reduce complaint escalations, reduce response time, train new staff, build a saved-replies library
5. Business size — small (1–2 staff managing socials), medium (3–10), or large (dedicated customer service team)
6. Active channels — confirm which of WhatsApp, Facebook Page, Instagram, Messenger, X/Twitter are in use
7. Current pain points — e.g. slow replies, staff not knowing what to say, public complaints going unanswered, abusive customers

Context: Customer Service on Social in Uganda/EA

In Uganda and across East Africa, customers use WhatsApp, Facebook Messenger, and public Facebook/Instagram comments as their primary channels for complaints, enquiries, and bookings. Email and formal telephone support are secondary. Volume is high, turnaround expectations are short, and poor responses are routinely screenshotted and shared publicly.

Many clients delegate social inboxes to junior staff or interns without formal training. This playbook addresses that gap with clear processes, ready-to-use scripts, and an escalation structure.

1. Response Time SLAs by Platform and Business Size

Define and publish SLAs internally. Post expected response times in your Page bio or WhatsApp Business profile.

Business hours default (Uganda): Monday–Friday 08:00–18:00 EAT; Saturday 08:00–14:00 EAT.

Rules:

• Acknowledge first, resolve second. A holding reply within SLA is better than silence while you investigate.
• Weekends and public holidays: set auto-replies and nominate one on-call staff member for urgent complaints.
• If a complaint has gone unanswered beyond double the SLA, escalate immediately to a supervisor before responding — the response now requires more care.

2. Three Core Customer Service Metrics

Source: Macarthy, A. (2023) 500 Social Media Marketing Tips.

Track these three metrics monthly for every active channel:

1. Queries by Channel (Volume Tracker) Count incoming messages and comments by platform and by query type (see triage framework below). Identifies where volume is highest and where to focus training or automation.

2. Speed of First Reply Average time from message received to first response sent. Most platforms (Facebook, Instagram) display this publicly as a response rate badge. Aim for the "Very responsive" badge on Facebook (90%+ of messages answered within 15 minutes — use auto-reply to achieve this).

3. Resolution Rate Percentage of queries fully resolved without the customer needing to follow up. Track by logging outcomes: Resolved / Escalated / Unresolved / No response needed. Low resolution rate signals script gaps or staff knowledge deficits.

Record these in a simple monthly tracker (spreadsheet). Review at the monthly customer service meeting.

3. Query Triage Framework

Classify every incoming message before responding. Apply the correct response path.

Triage rule: When in doubt about whether something is a Complaint or a Crisis-trigger, treat it as a Crisis-trigger until a manager confirms otherwise.

4. Public-to-Private Escalation Protocol

Move a conversation from a public comment to DM or WhatsApp when:

• The customer is sharing personal information (order number, phone number, address)
• The complaint requires investigation that cannot be resolved in one reply
• Emotion is escalating and continued public exchange risks amplifying the situation
• The resolution involves compensation, a refund, or an exception to policy

How to escalate without appearing defensive:

Post a brief, empathetic public reply first — never disappear into DM without acknowledging publicly. Use this structure:

"Thank you for reaching out, [Name]. We are sorry to hear about your experience. We want to look into this properly for you — please send us a direct message / WhatsApp us on [number] with your order details so we can resolve this as quickly as possible."

This signals to other readers that you have responded and are taking the matter seriously. It does not admit fault. It moves the detail off the public thread.

Never:

• Delete a legitimate complaint comment (screenshotting is immediate)
• Ask a customer to "email us" — they will not, and it appears dismissive in EA context
• Move to private without any public acknowledgement

5. Complaint Handling Scripts

Adapt all scripts to the client's brand voice. Use "we" throughout — never "the company" or third-person references to your own team.

Script 1 — Product Complaint (Public Comment)

"We are sorry to hear this, [Name] — this is not the experience we want for you at all. Please send us a direct message with your contact details and a brief description of what happened, and we will sort this out for you right away. Thank you for letting us know."

Script 2 — Delivery/Order Complaint (Messenger or WhatsApp)

"Hello [Name], thank you for getting in touch. We sincerely apologise for the delay/issue with your order. Could you please share your order number or the phone number used to place the order? We will investigate immediately and get back to you within [X hours] with an update. We appreciate your patience."

(After investigation:)

"Hello [Name], we have looked into your order and [explain what happened briefly and honestly]. We [state resolution: resend / refund / discount / apology]. We are truly sorry for the inconvenience. Please let us know if there is anything else we can do."

Script 3 — Unresolved Repeat Complaint

Use when a customer returns with the same issue unresolved.

"Hello [Name], we are very sorry that this issue has not yet been fully resolved — you should not have had to come back to us about this. I am personally escalating your case to our [manager/supervisor] right now. You will receive a direct call/message from us within [X hours]. Thank you for your continued patience, and we will make this right."

Script 4 — Unreasonable or Abusive Customer (Firm Close)

Do not match the customer's tone. Respond once, clearly and professionally.

"Hello [Name], we understand you are frustrated and we take all feedback seriously. However, we are not able to continue this conversation while it includes [threatening/offensive language]. We are still happy to assist you respectfully — please message us again and a member of our team will respond. Thank you."

If abuse continues after one further attempt: document the exchange, block or restrict the account, and do not respond further. Note in the monthly tracker.

Script 5 — Positive Feedback / Compliment (Acknowledge and Amplify)

"Thank you so much, [Name] — this genuinely made our day! We are so glad you [enjoyed the product / had a great experience / loved the service]. We will pass this on to the team. We look forward to seeing you again soon!"

Amplify: With the customer's permission, share the comment as a testimonial on your Stories or feed. Ask: "We would love to share your kind words — are you happy for us to repost this?"

6. Saved Replies Library

Build a saved-replies library covering the 10 most common query types for the client's business. Saved replies are stored in WhatsApp Business, Facebook Business Suite, or a shared Google Doc for staff to copy quickly.

The 10 query types to cover (adapt to the client's industry):

1. Pricing / quotation request
2. Product availability / stock check
3. Delivery timeframe and area
4. Order status update
5. Returns / refund policy
6. Booking / reservation confirmation
7. Operating hours and location
8. Complaint acknowledgement (holding reply)
9. Out-of-stock / waitlist notification
10. General "thank you for your message" acknowledgement

Three Worked Examples:

Saved Reply: Operating Hours

"Hello! Thank you for your message. We are open Monday to Friday, 8am–6pm, and Saturday 8am–2pm. We are closed on Sundays and public holidays. Feel free to place your order or ask your question here and we will respond as soon as we open. Thank you!"

Saved Reply: Delivery Timeframe

"Hello [Name]! Thank you for your order. We deliver within Kampala within 24–48 hours of order confirmation. Orders outside Kampala are dispatched within 48 hours via [courier name]. We will send you a confirmation message once your order is on its way. Thank you for shopping with us!"

Saved Reply: Complaint Acknowledgement (Holding Reply)

"Hello [Name], thank you for getting in touch. We are sorry to hear about your experience and we want to resolve this as quickly as possible. Could you please share your order number / booking reference? We are looking into this now and will update you within [2 hours]. We appreciate your patience."

7. Empathy Language Guide for the EA Context

Ugandan and East African customers respond well to warmth, directness, and respect. Corporate coldness, deflection, and over-formal language read as dismissive.

Use:

• "We are sorry" — not "We apologise for any inconvenience caused"
• "We will sort this out for you" — not "Your query has been escalated to the relevant department"
• "Thank you for telling us" — not "We note your feedback"
• "We understand this is frustrating" — not "We regret any perceived service failure"
• First names where known — customers respond well to being addressed by name
• "We" throughout — you are the business, not a separate entity from it

Avoid:

• Passive voice that obscures responsibility ("mistakes were made")
• Jargon or policy-speak ("as per our terms and conditions")
• Promises you cannot keep ("this will never happen again")
• Over-apologising without a resolution ("So sorry! So sorry! So sorry!")
• Emojis in complaint handling — appropriate in compliment replies, not in apologies

Tone calibration: Warm and competent. You are a trusted person helping a neighbour, not a call-centre script.

8. Out-of-Hours Management

Auto-reply setup (WhatsApp Business and Facebook):

Configure an auto-reply that:

• Confirms receipt of the message
• States business hours clearly
• Gives an expected response time
• Provides an emergency contact if genuinely urgent (e.g. perishable delivery, safety issue)

Sample auto-reply:

"Hello! Thank you for messaging [Business Name]. Our team is currently offline — we are available Monday–Friday, 8am–6pm, and Saturday, 8am–2pm EAT. We will respond to your message as soon as we open. If your matter is urgent, please call [number]. Thank you for your patience!"

Handling urgent out-of-hours complaints:

• Nominate one on-call staff member per week for urgent messages only
• Define "urgent": a complaint that, if not acknowledged, will generate public escalation before morning (e.g. a public comment gaining traction, a delivery that has not arrived for a vulnerable customer)
• On-call staff send a brief holding reply only — no investigation or resolution out of hours unless they have the authority and information to resolve it

9. Staff Training for Customer Service on Social

Briefing new team members — cover these five points:

1. Triage first — classify before you type. Use the five triage categories.
2. Never delete a legitimate complaint — hide the comment if necessary (only you can see it, not the public) while you investigate, but do not delete. Screenshot before hiding.
3. Never argue publicly — if a customer is wrong, correct gently and privately, never in a public thread.
4. Know your escalation trigger — any mention of illness, injury, legal action, or a post gaining rapid shares goes to a manager immediately. Do not attempt to resolve it independently.
5. Stay in role — personal opinions, humour, and off-script language are not appropriate in the service inbox. When in doubt, use a saved reply.

What staff must never say (in any channel, in any format):

• "That is not our fault" / "You should have…" / "Read the terms and conditions"
• "I cannot help you with that" without offering an alternative
• Anything that implies the customer is lying
• Information about internal processes, staff names, or supplier details
• Any commitment to a refund or compensation without manager authorisation

Escalate to management when:

• The complaint involves a potential health or safety issue
• The customer is threatening legal action or a media complaint
• A post is gaining shares or comments at an unusual rate
• The staff member does not know the answer and cannot find it within 10 minutes
• The same customer has complained more than twice about the same issue

10. Monthly Customer Service Review

Conduct a 30-minute review at the end of each month. Use the three core metrics (Section 2) as the baseline.

What to track:

• Total query volume by channel and by type (Enquiry / Complaint / Compliment / Abuse / Crisis-trigger)
• Average speed of first reply per channel
• Resolution rate — what percentage were fully resolved at first contact?
• Number of escalations to management
• Number of repeat complaints (same customer, same issue)
• Any complaints that went public or were shared before resolution

Identify recurring issues:

• Which product, service, or process generated the most complaints this month?
• Is there a pattern (same day, same time, same delivery route, same product)?
• Are staff consistently unsure how to handle a particular query type? (Add a saved reply.)

Feed insights back:

• Share recurring complaint themes with the product, operations, or service delivery team — not as blame, but as diagnostic data
• Update saved replies if a new common query type emerged
• Update staff training if a recurring staff error was identified
• Adjust SLAs if volume has grown beyond current capacity

Output of monthly review: A one-page summary covering: top 3 complaint themes, average response time vs. SLA, resolution rate, and one recommended process improvement for the following month.

Quality Criteria

Output meets the standard of this playbook if it:

1. Is immediately operational — a junior staff member with no prior training can pick it up, follow the triage framework, select the correct script, and handle the most common query types without additional guidance.
2. Reflects EA channel reality — WhatsApp and Facebook are treated as primary service channels, not secondary; scripts and SLAs are calibrated accordingly.
3. Uses warm, direct language — all scripts read as human, respectful, and brand-appropriate; no corporate deflection, no passive voice, no empty apology loops.
4. Has clear escalation paths — the boundary between what a junior staff member handles independently and what goes to management is unambiguous at every point.
5. Covers the full triage spectrum — Enquiry, Complaint, Compliment, Abuse, and Crisis-trigger each have a defined response path; nothing is left to the staff member's judgement without a framework.
6. Is measurable — the three core metrics are defined, tracked, and reviewed monthly; the client can demonstrate service improvement over time.
7. Protects the brand publicly — the public-to-private escalation protocol and the "never delete" rule are applied consistently; the playbook reduces the risk of a complaint going viral due to mishandling.
8. Feeds insight upstream — the monthly review process ensures recurring service issues reach the product or operations team, not just the social media manager.