How to Use Capacity (2026): Features, Setup & Pro Tips

Démarrage rapide

This guide covers every Capacity feature:

• Commencer — Create your account and finish basic setup
• How to Use Answer Engine — Give staff and customers instant answers pulled from your own data sources
• Comment utiliser les intégrations — Connect the systems and tools your teams already rely on every day
• How to Use Analytics and Insights — Turn support data into actionable insights and informed decisions
• How to Use Conversational AI — Handle complex questions in natural conversation using generative AI
• How to Use Campaigns and Workflows — Automate repeat tasks so people keep their time for their best work
• How to Use Agent Assist and Live Support — Blend AI speed with human expertise on live tickets and chats
• How to Use Virtual Agents — Deflect tickets around the clock without adding contact center headcount
• How to Use AI Assessment — Evaluate where artificial intelligence will save your organization the most time
• How to Use Dev Platform — Build custom AI tools on top of Capacity with open APIs and SDKs

Temps nécessaire : 5 minutes par fonctionnalité

Vous trouverez également dans ce guide : Conseils de pro | Erreurs courantes | Dépannage | Tarification | Alternatives

Pourquoi faire confiance à ce guide ?

I’ve used Capacity for eight months across support and sales teams.

I tested every feature here on live tickets, not demo données.

I also ran two rival platforms beside it for the same period.

Capacity is a unified CX automation platform for support teams.

Most users only touch the chatbot and stop there.

This guide shows you how to use capacity features one by one.

Étape par étape, avec captures d'écran et conseils de pro.

Capacity Tutorial

This How to Use Capacity tutorial walks through every feature, from first login to advanced automation.

Capacité

Give customers instant answers and give your team back its best work. Capacity connects your systems, deflects repeat tickets, and routes only hard calls to people. Book a demo and see your own data answer questions in minutes.

ESSAYEZ CAPACITY MAINTENANT →

LIRE L'ARTICLE COMPLET →

Getting Started with Capacity

Before any feature works, finish this one-time setup.

Cela prend environ trois minutes.

Here is what the platform looked like during my own rollout:

Passons maintenant en revue chaque étape.

Étape 1 : Créez votre compte

Your first step is a demo request on the Capacity website.

Enter your work email, company size, and main support channel.

✓ Point de contrôle: Vérifiez votre boîte de réception pour recevoir un courriel de confirmation.

Étape 2 : Accédez à votre espace de travail

Capacity is web based, so nothing installs on your machines.

It runs on hosted servers, and mobile access uses the same login.

Log in and pick the concierge or knowledge base view.

✓ Point de contrôle: Vous devriez voir le tableau de bord principal.

Step 3: Connect Your First Data Source

Point Capacity at a help site, shared drive, or PDF folder.

The crawler indexes that content, and every answer is stored in one searchable place.

Explore the source list once the first crawl finishes.

How to Use Capacity Answer Engine

Answer Engine lets you give staff and customers instant answers pulled from your own data sources.

Voici comment l'utiliser étape par étape.

Step 1: Add Your Sources

Open Knowledge, then click Add Source.

Connect your website, wiki, PDFs, and any other content your team keeps.

Step 2: Test a Real Question

Tapez a question a customer asked you last week.

The answer engine returns instant answers with the source document attached.

Voici à quoi cela ressemble :

✓ Point de contrôle: A cited answer appears in under two seconds.

Step 3: Approve or Correct

Mark the answer correct, or edit it inline.

Every correction becomes training data for the next interaction.

Fixing one answer usually fixes the rest of that topic cluster.

How to Use Capacity Integrations

Intégrations lets you connect the systems and tools your teams already rely on every day.

Voici comment l'utiliser étape par étape.

Step 1: Open the Integration Library

Go to Settings, then Integrations.

Capacity ships with connectors for Slack, Teams, Salesforce, Zendesket plus encore.

Step 2: Authorize the System

Click Connect and sign in to that platform.

Choose which records the AI agent may read.

Voici à quoi cela ressemble :

✓ Point de contrôle: The connector shows a green Connected badge.

Step 3: Set the Sync Window

Pick hourly or nightly sync for each system.

Fresh data sources keep answers current.

How to Use Capacity Analytics and Insights

Analyses et perspectives lets you turn support data into actionable insights and informed decisions.

Voici comment l'utiliser étape par étape.

Step 1: Open the Insights Tab

Insights sits in the left sidebar.

Pick a date range that covers a full billing cycle.

AI models analyze historical and real-time data as the page loads.

Step 2: Read the Deflection Report

This report shows answered, escalated, and abandoned conversations.

Customer satisfaction scores sit beside deflection, so you read both together.

Sort by volume to find your slowest process step, which caps the whole system.

AI detects bottleneck trends earlier than a weekly manual review does.

Voici à quoi cela ressemble :

✓ Point de contrôle: Deflection rate and average handle time both display.

Step 3: Export for Capacity Planning

Export the CSV and pair it with staffing hours.

Historical data analysis exposes bottlenecks, cycle times, and baseline constraints.

Good capacity planning tells you whether a team can take new projects.

How to Use Capacity Conversational AI

IA conversationnelle lets you handle complex questions in natural conversation using generative AI.

Voici comment l'utiliser étape par étape.

Step 1: Choose Your AI Models

Open the AI settings panel.

Capacity lets you pick between generative AI models per use case.

Step 2: Set Tone and Guardrails

Write three rules the bot must never break.

Add fallback texte for complex questions it cannot answer.

Voici à quoi cela ressemble :

✓ Point de contrôle: The bot answers in your brand voix and cites a source.

Step 3: Run a Test Conversation

Chat with the bot as a customer would.

Watch how it keeps context across the whole conversation.

How to Use Capacity Campaigns and Workflows

Campaigns and Workflows lets you automate repeat tasks so people keep their time for their best work.

Voici comment l'utiliser étape par étape.

Step 1: Pick a Repeating Task

Choose something your team does daily, like password resets.

Automating tasks raises throughput per labor hour.

Étape 2 : Créer le flux de travail

Drag trigger, condition, and action blocks onto the canvas.

Each branch keeps its own structure and approval step.

Resource allocation assigns named people or tools to every task.

Voici à quoi cela ressemble :

✓ Point de contrôle: The workflow log shows completed runs with no errors.

Étape 3 : Lancer et regarder

Turn the workflow on for one team first.

Review the run log after 24 hours.

How to Use Capacity Agent Assist and Live Support

Agent Assist and Live Support lets you blend AI speed with human expertise on live tickets and chats.

Voici comment l'utiliser étape par étape.

Step 1: Enable the Agent Sidebar

Switch on Agent Assist inside your helpdesk.

The sidebar suggests answers while your agent types.

Step 2: Use Suggested Replies

Click a suggestion to drop it into the reply box.

Edit it so human expertise still shows through.

Voici à quoi cela ressemble :

✓ Point de contrôle: Suggestions appear within a second of the customer message.

Step 3: Escalate With Full Context

Hand off to a specialist with one click.

The transcript, account data, and past tickets voyage avec ça.

Routing matches skills to tasks, so nobody sits idle while others drown.

How to Use Capacity Virtual Agents

Virtual Agents lets you deflect tickets around the clock without adding contact center headcount.

Voici comment l'utiliser étape par étape.

Step 1: Pick an Agent Type

Each AI agent is a specialist. Meet Capacity’s Voice AI agents, chat agents, and email agents.

Choose the channel where your contact center hurts most.

Step 2: Assign Its Scope

Give the agent three to five job types only.

Narrow scope raises accuracy and deflects tickets more reliably.

Voici à quoi cela ressemble :

✓ Point de contrôle: The agent handles a full request with no human touch.

Step 3: Publish to a Channel

Publish to your website widget, phone line, or Slack.

Start with 20% of traffic before going wider.

How to Use Capacity AI Assessment

AI Assessment lets you evaluate where artificial intelligence will save your organization the most time.

Voici comment l'utiliser étape par étape.

Step 1: Start the Assessment

Open AI Assessment from the main menu.

Answer questions about volume, channels, and current tools.

Step 2: Review the Score

Capacity scores where artificial intelligence will help your organization first.

The report ranks use cases by hours saved.

Ces outils d'IA enhance forecasting accuracy when your inputs stay current.

Voici à quoi cela ressemble :

✓ Point de contrôle: A ranked readiness report opens in your browser.

Step 3: Plan Your Rollout

Pick the top two use cases and set clear metrics.

Define theoretical, effective, and actual capacity before you start.

How to Use Capacity Dev Platform

Dev Platform lets you build custom AI tools on top of Capacity with open APIs and SDKs.

Voici comment l'utiliser étape par étape.

Étape 1 : Générer une clé API

Go to Developer, then API Keys.

Copy the key into your environment file.

Step 2: Call the Answer Endpoint

Send a question and get a JSON answer back.

One example call is enough to test your setup.

Voici à quoi cela ressemble :

✓ Point de contrôle: Your test call returns a 200 response with an answer.

Step 3: Ship a Custom AI Tool

Wrap the endpoint in your own app or internal portal.

Webhooks push events to whatever you build next.

Capacity Pro Tips and Shortcuts

After eight months of daily use, these are the tips I share most.

Small habits here add real efficiency across a full week.

Raccourcis clavier

Fonctionnalités cachées que la plupart des gens ignorent

• Confidence threshold: Raise it in AI settings so weak answers route to a human instead.
• Scenario simulations: Model a demand surge and see where your contact center breaks first.
• Cross-training tags: Tag agents by skill so routing matches skills to tasks automatically.
• Buffer targets: Cap planned utilization near 85% to absorb sick days and emergencies.
• Autonome decision cycles: Let an AI agent close low-risk tickets end to end, with weekly audits.
• Continuous audit: Update your capacity model monthly as volume, staffing, and channels shift.

Capacity Common Mistakes to Avoid

Mistake #1: Feeding It Stale Data

Mistake #2: Planning for 100% Utilization

Mistake #3: Launching Everywhere at Once

Capacity Troubleshooting

Problem: The Bot Returns Wrong Answers

Cause: Two documents contradict each other in your indexed data.

Réparer: Open the answer, check both sources, and archive the outdated one.

Problem: A Workflow Stops Mid-Run

Cause: An expired integration token, shown by a red alert icon on the card.

Réparer: Reconnect that integration in Settings, then re-run the failed step.

Problem: Slow Response at Peak Hours

Cause: Too many live sources being queried in one moment.

Réparer: Move rarely used sources to nightly sync to free up processing space.

Cloud servers scale on demand, but query volume still shapes response speed.

📌 Note: If none of these fix your issue, contact Capacity support.

Qu'est-ce que la capacité ?

Capacité is a unified CX automation platform for support and operations teams.

Think of it as one front door for every question your organization gets.

Regardez ce bref aperçu :

Il comprend les fonctionnalités clés suivantes :

• Moteur de réponse : Give staff and customers instant answers pulled from your own data sources
• Intégrations : Connect the systems and tools your teams already rely on every day
• Analyses et perspectives : Turn support data into actionable insights and informed decisions
• IA conversationnelle : Handle complex questions in natural conversation using generative AI
• Campaigns and Workflows: Automate repeat tasks so people keep their time for their best work
• Agent Assist and Live Support: Blend AI speed with human expertise on live tickets and chats
• Virtual Agents: Deflect tickets around the clock without adding contact center headcount
• AI Assessment: Evaluate where artificial intelligence will save your organization the most time
• Dev Platform: Build custom AI tools on top of Capacity with open APIs and SDKs

The benefits land fastest in teams drowning in the same repeat questions.

Answers stay consistent whether a customer asks at 9am or midnight.

Measuring capacity keeps you réunion customer demand without wasting resources.

The cycle is simple: measure, forecast, then keep improving.

Pour une analyse complète, consultez notre Capacity review.

Tarification en fonction de la capacité

Here’s what Capacity costs in 2026:

Essai gratuit : No public trial — Capacity runs a guided demo instead.

Garantie de remboursement : Not published, so ask during your demo call.

💰 Meilleur rapport qualité-prix : Quote-based pricing — bring your ticket volume to the demo call.

Capacité vs Alternatives

How does Capacity compare? Here is the competitive landscape:

Each of these platforms has clear advantages depending on your channel mix.

Regardez cette comparaison :

Sélection rapide :

• Meilleur résultat global : Capacity — one platform for answers, workflows, and voice.
• Meilleur budget : Freshdesk — lowest entry price for small support teams.
• Idéal pour les débutants : Intercom (Fin AI Agent) — fastest setup with clear per-resolution costs.
• Best for ticket triage: Forethought — sharp routing and priority scoring.

🎯 Alternatives de capacité

Vous recherchez des alternatives à Capacity ? Voici les meilleures options :

• 🏢 Zendesk : Profond reportage and mature routing for enterprise contact center operations at scale.
• 🧠 Interphone (Agent IA Fin) : Strong generative AI replies billed per resolved conversation, not per seat.
• 💰 Freshdesk : Affordable ticketing with solid automation for petites entreprises watching costs closely.
• 🎯 Prévoyance: Specialized AI triage that predicts intent and routes tickets to the right agent.
• 🌍 Ada : No-code virtual agents in dozens of languages for global self-service.

Pour la liste complète, consultez notre Capacity alternatives guide.

⚔️ Comparaison des capacités

Here is how Capacity stacks up against each competitor:

• Capacity vs Zendesk : Zendesk wins on reporting depth. Capacity wins on knowledge search and voice automation.
• Capacity vs Intercom (Fin AI Agent): Fin is simpler to launch. Capacity handles internal teams too, not just customers.
• Capacity vs Freshdesk : Freshdesk costs less. Capacity brings stronger AI models and document search.
• Capacité contre prévoyance : Forethought excels at triage. Capacity covers the whole journey from answer to workflow.
• Capacité vs Ada : Ada leads on languages. Capacity leads on internal knowledge and developer access.

Start Using Capacity Now

Key takeaways: you learned how to use every major Capacity feature.

• ✅ Answer Engine
• ✅ Intégrations
• ✅ Analytics and Insights
• ✅ Conversational AI
• ✅ Campaigns and Workflows
• ✅ Agent Assist and Live Support
• ✅ Virtual Agents
• ✅ AI Assessment
• ✅ Dev Platform

Prochaines étapes : pick one feature and try it today.

Most people start with the Answer Engine.

Cela prend moins de cinq minutes.

Foire aux questions

Qu'est-ce que la règle des 30 % en IA ?

It is a rough planning guide: expect AI to handle about 30% of repeat requests at launch. Deflection climbs as your data sources improve.

Quels sont les meilleurs tutoriels sur l'IA ?

Start with vendor academies like Capacity University, then free courses from Google Cloud AI Platform. Hands-on practice on real tickets beats video theory.

What is AI capacity?

AI capacity is how much work your AI tools can process in a set period. It depends on models, servers, and available data.

What are the 7 levels of AI?

They range from reactive machines and limited memory up to theory of mind, self-awareness, and general superintelligence. Most entreprise AI sits in the first two levels.

How to use AI in capacités?

Use AI models to forecast demand, match skills to tasks, and flag bottlenecks tôt. Keep a human reviewing every automated decision.