> For the complete documentation index, see [llms.txt](https://help.ringtonic.app/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.ringtonic.app/guides/setup-workspace.md). # Setup Workspace ### What is a workspace? A workspace is an isolated environment for managing a single business or client. Each workspace has its own dedicated tracking numbers, campaigns, call logs, team members, and analytics reports. This separation is essential for agencies that need to keep client data completely private and organized. Think of it like having multiple separate accounts, but managed from one convenient dashboard. {% hint style="info" %} When you create an account, Ring Tonic automatically sets up a default workspace for you. {% endhint %} Here are a few quick steps to get your new workspace ready: ### 1. Set up Twilio for Tracking Calls Ring Tonic uses Twilio to provision phone numbers and track calls. To get started, create a Twilio account and connect it to your workspace using your Account SID and Auth Token. This lets Ring Tonic search for numbers, purchase tracking lines, and log incoming calls on your behalf. #### How to Get Your Twilio Credentials **Step 1: Create a Twilio Account** 1. Go to 2. Fill out the registration form with your information 3. Verify your email address and phone number 4. Complete the onboarding process {% hint style="info" %} Twilio offers a free trial with credits to get started. You can test Ring Tonic's features before committing to a paid plan. {% endhint %} {% hint style="warning" %} **Important:** During Twilio account setup, if asked about your account type or business model, select **"Direct"** (not ISV/reseller). Since you own the account and the API keys, Twilio considers you a direct customer. 'ISV' is a specific status for software platforms (like Ring Tonic) that resell connectivity. Choosing Direct makes your account verification and SMS registration much simpler! {% endhint %} **Step 2: Find Your Account SID and Auth Token** Once you're logged into Twilio: 1. Navigate to the [Twilio Console](https://console.twilio.com) 2. You'll see your **Account SID** and **Auth Token** displayed under the Account Info section 3. The Account SID starts with "AC" followed by 32 characters 4. Click "Show" next to the Auth Token to reveal it

Your Account SID and Auth Token are displayed on the Twilio Console dashboard

{% hint style="warning" %} **Important:** Your Auth Token is sensitive information. Never share it publicly or commit it to version control. Ring Tonic encrypts and stores these credentials securely. {% endhint %} **Step 3: Add Credentials to Your Workspace** 1. In Ring Tonic, navigate to your workspace settings 2. Scroll down to the **Twilio Credentials** section 3. Paste your **Account SID** into the "Account SID" field 4. Paste your **Auth Token** into the "Auth Token" field 5. Click "Save" or "Update Workspace" to save your credentials

Enter your Twilio credentials in the workspace settings

Once saved, Ring Tonic will be able to: * Search for available phone numbers in your desired area codes * Purchase tracking numbers on your behalf * Receive and log incoming calls * Provide call recordings and metadata {% hint style="success" %} **What's Next?** After connecting Twilio, you can start creating campaigns and purchasing tracking numbers. {% endhint %} ### 2. Call Control — Disable Calls for a Workspace (Agency Plan) {% hint style="info" %} This feature is available exclusively on the **Agency plan**. If you're on the Indie plan, you'll see the toggle but will be prompted to upgrade when you try to use it. {% endhint %} If you manage multiple workspaces for different clients, there may be times when you need to temporarily disable all calling for a specific workspace—for example, if a client stops paying or requests a pause on their campaigns. The **Call Control** toggle gives you a single master switch to turn off all inbound and outbound calls for a workspace instantly. When calling is disabled: * **Inbound calls** to all tracking numbers in the workspace are rejected (callers hear a busy signal or a custom message you define) * **Outbound dialer** is hidden and blocked for all team members in the workspace * **No call logs** are created for rejected inbound calls, so your analytics stay clean #### How to Disable Calls 1. Navigate to your workspace settings 2. Scroll down to the **Call Control** section (below Twilio Credentials) 3. Toggle **"Calling Enabled"** off

The Call Control toggle in workspace settings

#### Choosing a Rejection Method When you disable calls, you can choose what callers hear when they dial one of your tracking numbers: | Method | What Callers Hear | Twilio Cost | Best For | | ------------------ | ------------------------------------------- | ---------------- | -------------------------------------------------------------------------------------- | | **Busy Signal** | A standard busy tone, then the call ends | $0 (no charge) | Temporary suspensions where you don't want to reveal any details | | **Custom Message** | Your custom message, then the call hangs up | \~$0.01 per call | Professional communication, letting callers know the number is temporarily unavailable | **To configure the rejection method:** 1. After toggling calls off, select either **"Busy Signal"** or **"Custom Message"** 2. If you chose **Custom Message**, enter your message in the text field (up to 500 characters) 3. Click **"Save"** or **"Update Workspace"** **Example custom messages:** * "The number you have called is temporarily unavailable. Please try again later." * "This line is currently not accepting calls. Please contact us at our main office number." * "Service for this account has been temporarily suspended. Please contact your account manager." {% hint style="warning" %} **Important:** When calls are disabled, ALL inbound calls to every tracking number in the workspace will be rejected, and the outbound dialer will be completely hidden. Make sure this is what you intend before saving. {% endhint %} #### Re-enabling Calls To restore calling, simply toggle **"Calling Enabled"** back on and save. All inbound call tracking and the outbound dialer will resume immediately—no need to reconfigure anything. #### Per-Number Call Control Need more granular control? Instead of disabling an entire workspace, you can disable calling for **individual phone numbers**. This is useful when you want to suspend just one line while keeping the rest of the workspace active. Per-number call control is managed from the **Phone Numbers** page. See the [Phone Numbers guide](/guides/phone-numbers.md#per-number-call-control-agency-plan) for full details. ### 3. Set up Call Transcription & Sentiment Analysis Call transcription turns recorded conversations into searchable text, making it easy to find calls, analyze patterns, and spot keywords without replaying full recordings.

Transcription with AssemblyAI

Ring Tonic supports Deepgram and AssemblyAI—both offer automatic transcription with speaker identification, but differ in speed, accuracy, and pricing. {% hint style="info" %} **Don't know which provider to choose?** We recommend **AssemblyAI** for most users. It provides the best accuracy with 30% improvement in noisy environments, making it ideal for phone call transcription where audio quality can vary. {% endhint %} #### Choosing a Transcription Provider Ring Tonic integrates with two powerful transcription providers. Here's how they compare: | Feature | Deepgram | AssemblyAI | | ----------------------- | ------------------------------------------ | ------------------------------------------------ | | **Processing Speed** | 10x faster than real-time | Standard processing | | **Accuracy** | Excellent | Best-in-class (30% better in noisy environments) | | **Speaker Diarization** | Up to 100,000+ speakers | Up to 50 speakers | | **Language Support** | 30+ languages | 100+ languages | | **Best For** | High-volume call centers, batch processing | Maximum accuracy, challenging audio conditions | | **Pricing** | \~$0.0043/minute | \~$0.00065/minute | {% tabs %} {% tab title="AssemblyAI" %} **How to Set Up AssemblyAI** **Step 1: Create an AssemblyAI Account** 1. Visit 2. Sign up for a new account 3. Verify your email address 4. Complete the onboarding questionnaire {% hint style="info" %} AssemblyAI provides $50 free credits to get started, allowing you to test their transcription quality with your actual call recordings. {% endhint %} **Step 2: Get Your AssemblyAI API Key** 1. After logging in, you'll be taken to the AssemblyAI Dashboard 2. Your API key is displayed prominently on the dashboard homepage 3. Click "Copy" to copy your API key to your clipboard 4. Alternatively, you can access your API key at any time from the [API Keys page](https://assemblyai.com/dashboard/api-keys)

Your AssemblyAI API key is shown on the dashboard

**Step 3: Add AssemblyAI Credentials to Your Workspace** 1. In Ring Tonic, navigate to your workspace settings 2. Go to the **Transcription** tab 3. Select **AssemblyAI** as your transcription provider 4. Paste your API key into the "AssemblyAI API Key" field 5. Enable "Automatically transcribe call recordings" if you want transcriptions to run after every call 6. Enable "Enable speaker diarization" to identify different speakers (Agent vs Customer) 7. Optionally select a language, or leave it on "Auto-detect" for automatic language detection 8. Click "Save" or "Update Workspace"

Configure AssemblyAI in your workspace transcription settings

{% endtab %} {% tab title="Deepgram" %} **How to Set Up Deepgram** **Step 1: Create a Deepgram Account** 1. Visit 2. Sign up for a new account 3. Verify your email address 4. Complete the onboarding process {% hint style="info" %} Deepgram offers $200 in free credits to get started, which is approximately 46,500 minutes of transcription. {% endhint %} **Step 2: Get Your Deepgram API Key** 1. Once logged in, navigate to the [API Keys page](https://console.deepgram.com/project/default/settings/api-keys) in your Deepgram Console 2. Click "Create a New API Key" 3. Give your key a descriptive name (e.g., "Ring Tonic Production") 4. Set appropriate permissions (at minimum, you'll need "Usage" and "Transcription" permissions) 5. Click "Create Key" and copy the API key immediately (you won't be able to see it again)

Create a new API key in the Deepgram Console

{% hint style="warning" %} **Important:** Store your API key securely. Deepgram only shows the key once when you create it. If you lose it, you'll need to generate a new one. {% endhint %} **Step 3: Add Deepgram Credentials to Your Workspace** 1. In Ring Tonic, navigate to your workspace settings 2. Go to the **Transcription** tab 3. Select **Deepgram** as your transcription provider 4. Paste your API key into the "Deepgram API Key" field 5. Enable "Automatically transcribe call recordings" if you want transcriptions to run after every call 6. Enable "Enable speaker diarization" to identify different speakers (Agent vs Customer) 7. Optionally select a language, or leave it on "Auto-detect" for automatic language detection 8. Click "Save" or "Update Workspace"

Configure Deepgram in your workspace transcription settings

{% endtab %} {% endtabs %} #### Understanding Transcription Settings After choosing your provider and entering your API key, you'll need to configure a few important settings: **Automatically Transcribe Call Recordings**

Automatically Transcribe Call Recordings

When enabled, Ring Tonic will automatically send each call recording to your chosen transcription provider as soon as the call ends. The transcription usually completes within a few minutes, and you'll be able to view the full text conversation in your call log. If disabled, you'll need to manually trigger transcription for each call you want transcribed. {% hint style="success" %} **Recommendation:** Keep this enabled. Automatic transcription ensures you never miss capturing important conversation details, and you can always search through past calls by keywords. {% endhint %} **Enable Speaker Diarization** Speaker diarization is the process of identifying and labeling different speakers in the conversation.

Enable Speaker Diarization

When enabled, your transcription will show: * **Agent:** The person who answered or made the call (usually your team member) * **Customer:** The caller * **Speaker C, D, E...:** If more than two people participate in the call Example transcription with speaker diarization: ``` Agent: Thanks for calling ABC Plumbing, this is Sarah. How can I help you today? Customer: Hi Sarah, I have a leaking pipe under my kitchen sink and it's getting worse. Agent: I understand, that sounds urgent. Let me check our schedule for you. ``` Without speaker diarization, the entire conversation would appear as a single block of text, making it harder to follow. {% hint style="info" %} Speaker diarization adds minimal cost to your transcription and significantly improves readability. We recommend keeping it enabled. {% endhint %} **Language Selection** Both Deepgram and AssemblyAI support automatic language detection, which is the default setting. However, if you know your calls will always be in a specific language, selecting it manually can slightly improve accuracy and processing speed.
Ring Tonic supports: * **Deepgram:** 30+ languages including English, Spanish, French, German, Chinese, Japanese, and more * **AssemblyAI:** 100+ languages covering most global markets {% hint style="info" %} Leave the language setting on "Auto-detect" unless you exclusively handle calls in a single language. The auto-detection is highly accurate and adapts to different accents and dialects. {% endhint %} #### Live AI Call Transcription (AI Agent flows only) If your call flows use the [AI Agent node](/guides/ai-call-answering.md), the Transcription tab has a second section: **AI call transcription (live)**. This controls the speech-to-text engine that understands the caller **during** the live AI conversation — separate from the post-call transcription provider above. A few things make it different from post-call transcription: * It runs on your **Twilio account** — no separate API key needed. * Only **Google** and **Deepgram** are available for live calls. * The default, **Automatic — let Twilio pick the default (recommended)**, uses whichever engine Twilio selects for your account.

The AI call transcription (live) section on the Transcription tab

{% hint style="info" %} **Leave this on Automatic** unless a specific engine understands your callers better — accents and industry vocabulary occasionally favor one provider. If you never use the AI Agent node, this setting has no effect. {% endhint %} #### Set up Google Natural Language API for Sentiment Analysis (Optional) Sentiment analysis examines the emotional tone of your call transcriptions, helping you identify whether conversations were positive, negative, or neutral. This is incredibly valuable for: * Flagging potentially unhappy customers for follow-up * Identifying calls that might need manager review * Tracking overall customer satisfaction trends * Spotting training opportunities for your team
Ring Tonic uses Google's Natural Language API to analyze sentiment from your transcriptions. The API examines the entire conversation and returns a sentiment score ranging from -1.0 (very negative) to +1.0 (very positive). **How Sentiment Categories Work** Ring Tonic categorizes sentiment into three groups: | Sentiment Score | Category | Meaning | | ----------------- | ------------ | ------------------------------------------------------------------ | | Greater than 0.25 | **Positive** | Customer expressed satisfaction, enthusiasm, or approval | | -0.25 to 0.25 | **Neutral** | Factual conversation with balanced or mixed emotions | | Less than -0.25 | **Negative** | Customer expressed frustration, disappointment, or dissatisfaction | {% hint style="success" %} **Use Case:** Set up alerts for calls with negative sentiment so managers can review them promptly and reach out to unhappy customers before the situation escalates. {% endhint %} **Step 1: Create a Google Cloud Account** 1. Visit 2. Sign in with your Google account (or create one if needed) 3. Accept the terms of service 4. You may need to set up billing, but Google offers $300 in free credits for new users **Step 2: Create a New Project** 1. In the Google Cloud Console, click the project dropdown at the top of the page 2. Click "New Project" 3. Enter a project name (e.g., "Ring Tonic Sentiment Analysis") 4. Click "Create" 5. Wait for the project to be created, then select it from the project dropdown

Create a new project in Google Cloud Console

**Step 3: Enable the Natural Language API** 1. In the left sidebar, navigate to "APIs & Services" > "Library" 2. Search for "Cloud Natural Language API" 3. Click on "Cloud Natural Language API" from the results 4. Click the "Enable" button 5. Wait for the API to be enabled (usually takes a few seconds)

Enable the Cloud Natural Language API

**Step 4: Create an API Key** 1. Navigate to "APIs & Services" > "Credentials" 2. Click "Create Credentials" at the top 3. Select "API key" from the dropdown 4. Your new API key will be displayed (it starts with "AIza...") 5. Click "Copy" to copy the key 6. **Optional but recommended:** Click "Restrict Key" and add API restrictions to limit the key to only the Natural Language API

Create a new API key for the Natural Language API

{% hint style="warning" %} **Security Best Practice:** Restrict your API key to only the Natural Language API to prevent unauthorized use. You can do this by clicking "Restrict Key" after creating it, then under "API restrictions," select "Restrict key" and choose "Cloud Natural Language API." {% endhint %} **Step 5: Add Google NLP API Key to Your Workspace** 1. In Ring Tonic, navigate to your workspace settings 2. Go to the **Transcription** tab 3. Scroll down to the "Sentiment Analysis (Optional)" section 4. Paste your API key into the "Google NLP API Key" field 5. Enable "Automatically analyze sentiment from transcriptions" 6. Click "Save" or "Update Workspace"

Configure Google NLP for sentiment analysis

Once configured, Ring Tonic will automatically analyze sentiment for every transcribed call and display the results in your call log. **Supported Languages for Sentiment Analysis**
Google's Natural Language API supports sentiment analysis in the following languages: 1. Arabic (ar) 2. Chinese (Simplified) (zh) 3. Chinese (Traditional) (zh-Hant) 4. Dutch (nl) 5. English (en) 6. French (fr) 7. German (de) 8. Indonesian (id) 9. Italian (it) 10. Japanese (ja) 11. Korean (ko) 12. Portuguese (pt) 13. Spanish (es) 14. Thai (th) 15. Turkish (tr) 16. Vietnamese (vi)
{% hint style="info" %} If your transcription is in a language not supported for sentiment analysis, the sentiment will be marked as "Unsupported" in your call log. The transcription will still work normally. {% endhint %} {% hint style="success" %} **What's Next?** After setting up transcription and sentiment analysis, Ring Tonic will automatically process your call recordings and provide searchable transcripts with sentiment scores. You can view all this data in your call logs and use it to improve customer service, train your team, and qualify leads more effectively. {% endhint %} ### 4. Set up AI Automation with OpenAI AI automation supercharges your call analytics by automatically extracting insights from every conversation. With OpenAI integration, Ring Tonic can identify important keywords, categorize calls, qualify leads, and even estimate deal values—all without manual review.

Keyword Spotting

This is incredibly powerful for: * Automatically identifying hot leads ready to buy * Spotting keywords that indicate urgency or specific needs * Auto-tagging calls for easy filtering ("Ready to Buy", "Needs Escalation", "Price Shopping") * Estimating potential revenue from each call * Training your team by highlighting what customers care about Ring Tonic uses OpenAI's GPT models to analyze your call transcriptions and provide actionable intelligence on every conversation. {% hint style="success" %} **The same key powers AI Call Answering.** Once your OpenAI key is saved here, the [AI Agent node](/guides/ai-call-answering.md) can answer calls live — greeting, qualifying, and booking on your behalf. Usage bills directly to your OpenAI account. {% endhint %} {% hint style="info" %} **Important:** AI automation requires transcription to be set up first. Make sure you've completed [Section 3: Transcription](#id-3.-set-up-call-transcription-and-sentiment-analysis) before configuring AI features. {% endhint %} #### How to Set Up OpenAI **Step 1: Create an OpenAI Account** 1. Visit 2. Sign up with your email or Google account 3. Verify your email address 4. Complete the onboarding process **Step 2: Get Your OpenAI API Key** 1. After logging in, navigate to the [API Keys page](https://platform.openai.com/api-keys) 2. Click "Create new secret key" 3. Give your key a name (e.g., "Ring Tonic Production") 4. Optionally restrict permissions to only what's needed 5. Click "Create secret key" 6. Copy the API key immediately (it starts with "sk-" and you won't be able to see it again)

Create a new API key in the OpenAI Platform

{% hint style="warning" %} **Important:** Store your API key securely. OpenAI only shows the full key once when you create it. If you lose it, you'll need to generate a new one and update Ring Tonic. {% endhint %} **Step 3: Add OpenAI API Key to Your Workspace** 1. In Ring Tonic, navigate to your workspace settings 2. Go to the **AI Automation** tab 3. Paste your API key into the "OpenAI API Key" field 4. Configure your AI automation settings (explained below) 5. Click "Save" or "Update Workspace"

Configure OpenAI in your workspace AI automation settings

#### Automatically Analyze Keywords and Qualify Leads When enabled, Ring Tonic's AI will analyze every transcribed call to: 1. **Extract Important Keywords** - Identifies meaningful words and phrases that indicate customer intent, needs, or concerns 2. **Categorize Each Keyword** - Labels keywords by type (sales, scheduling, support, urgent, pricing, escalation, etc.) 3. **Identify the Speaker** - Knows whether the Agent or Customer said each keyword 4. **Auto-Tag Calls** - Applies tags like "Ready to Buy", "Needs Escalation", "Price Shopping", "Product Inquiry" for easy filtering 5. **Qualify Leads Automatically** - Determines if the call represents a qualified lead based on your criteria

Auto-tagging and keyword spotting

**Example: How It Works** Imagine a customer calls and says: > "Hi, I need someone to fix my water heater today. It's leaking everywhere and I'm worried about water damage. Can you give me a quote and come out this afternoon?" The AI would extract keywords like: * **"water heater"** (category: product, importance: 8, speaker: Customer) * **"today"** (category: urgent, importance: 9, speaker: Customer) * **"leaking"** (category: support, importance: 7, speaker: Customer) * **"quote"** (category: pricing, importance: 9, speaker: Customer) * **"this afternoon"** (category: scheduling, importance: 10, speaker: Customer) Auto-tags applied: "Ready to Buy", "Urgent", "Needs Quote" Lead qualification: **QUALIFIED** (95% confidence) because the customer requested a quote, showed urgency, and wanted to schedule service.

Auto-qualify Leads

{% hint style="success" %} **Recommendation:** Enable this feature. It saves hours of manual call review and ensures no qualified leads slip through the cracks. The AI is remarkably accurate at identifying buying intent. {% endhint %} #### Custom Qualification Criteria (Optional) Every business defines "qualified leads" differently. Ring Tonic provides sensible defaults, but you can customize the qualification criteria to match your specific business needs. **Default Qualification Criteria** Out of the box, Ring Tonic considers a call qualified if the customer: 1. Requests a quote or pricing information 2. Wants to schedule an appointment within 2 weeks 3. Has an urgent problem that needs immediate attention 4. Mentions they are ready to move forward with the service

Default Qualification Criteria

**Customizing for Your Business** You can override the default criteria with your own business-specific guidelines. For example: **For a High-End Law Firm:** ``` A lead is qualified if: - Case value exceeds $50,000 - Client has immediate legal needs - Decision-maker is on the call - Case type matches our practice areas (corporate, real estate, litigation) ``` **For a Roofing Company:** ``` A lead is qualified if: - Customer owns the property (not renters) - Roof damage is current, not future planning - Customer requests inspection or estimate - Property is within our service area - Customer has decision-making authority ``` **For a SaaS Business:** ``` A lead is qualified if: - Company size is 50+ employees - Customer discusses budget or pricing - Decision-maker or influencer is on the call - Timeline is within 3 months - Specific use case mentioned that matches our product ``` The AI will incorporate your custom criteria when analyzing calls and determining qualification status. {% hint style="info" %} Don't worry about making it perfect. The AI is smart enough to understand natural language guidelines. Write your criteria as if you're explaining it to a new team member. {% endhint %} #### Confidence Threshold The confidence threshold determines how certain the AI must be before automatically qualifying a lead. This is a crucial setting that balances quantity versus quality.

Confidence threshold field

**How It Works** When the AI analyzes a call, it assigns a qualification confidence score from 0-100: * **90-100%:** Very confident this is a qualified lead * **75-89%:** Confident, but some uncertainty * **60-74%:** Moderate confidence * **Below 60%:** Low confidence Your workspace confidence threshold (default: 80%) acts as a filter. Only calls meeting or exceeding this threshold will be automatically marked as qualified. **Choosing Your Threshold** | Threshold | Effect | Best For | | ----------- | --------------------------------------------------- | ------------------------------------------------------------------------ | | **50-65%** | More leads auto-qualified, lower accuracy | High-volume businesses that manually review all leads anyway | | **70-80%** | Balanced approach, good accuracy | Most businesses (recommended) | | **85-100%** | Only very obvious qualified leads, highest accuracy | Businesses with limited sales resources, want to focus only on hot leads | **Example Scenario** With an 80% threshold: * Call A: 95% confidence → Auto-qualified ✅ * Call B: 82% confidence → Auto-qualified ✅ * Call C: 75% confidence → Not auto-qualified (you can manually review) ❌ * Call D: 60% confidence → Not auto-qualified ❌ {% hint style="success" %} **Recommendation:** Start with 80% (the default) and adjust based on your results. If you find the AI is missing qualified leads, lower it to 70-75%. If you're getting too many false positives, raise it to 85-90%. {% endhint %} #### Re-qualify Existing Leads Edits to your qualification criteria or confidence threshold apply to new calls going forward. To bring your **existing** leads in line with the updated criteria, re-qualify them. Ring Tonic re-runs the AI analysis on calls you've already received, using your latest criteria and threshold — no new phone calls are made, it reads the transcripts you already have. There are two ways to start, depending on whether you want to re-qualify everyone or just a selection. {% tabs %} {% tab title="All leads (from this page)" %} Click **"Re-qualify existing leads"** directly beneath the Custom Qualification Criteria box. This re-runs qualification for every eligible lead in the workspace — the natural next step right after you edit your criteria or threshold.

The "Re-qualify existing leads" button beneath the Custom Qualification Criteria editor

{% endtab %} {% tab title="Selected leads (from Call Activity)" %} Go to **Analytics > Call Activity**, tick the calls you want, then choose **Actions > Re-qualify Leads**. To re-qualify the whole workspace from here, use the header checkbox to select all.

The "Re-qualify Leads" bulk action in the Call Activity Actions menu

{% endtab %} {% endtabs %} **Choosing how the re-run behaves** Before it starts, a dialog lets you adjust three options: | Option | Default | What it does | | ---------------------------------- | ------- | --------------------------------------------------------------------------------------------------------------------------------------- | | **Allow downgrades** | On | Lets stricter criteria move an already-qualified lead back to *not qualified*. Turn off to only upgrade leads and never downgrade them. | | **Trigger webhooks & automations** | On | Re-fires lead webhooks and notifications for leads that *newly* become qualified. Turn off for a silent update. | | **Include voicemails** | On | Also re-qualifies missed calls that left a transcribed voicemail. |

Re-qualification options dialog with the allow-downgrades, webhooks, and voicemails toggles

{% hint style="info" %} Re-qualification runs in the background and can take a while on large workspaces. Ring Tonic emails you a summary when it finishes, showing how many leads ended up qualified, not qualified, still pending, or flagged as spam. {% endhint %} {% hint style="warning" %} Only calls with a transcription — or a transcribed voicemail — are re-qualified; calls without one are skipped. Leads you've set by hand are always preserved and never changed by a re-run. {% endhint %} #### Automatically Estimate Deal Value Deal value estimation helps you prioritize leads based on potential revenue. When enabled, Ring Tonic's AI analyzes customer conversations and automatically estimates the potential deal value using your Products & Services pricing catalog.

Automatically Estimate Deal Value field

**How It Works** When a customer mentions specific products or services during a call, the AI: 1. Matches their needs to items in your catalog 2. Considers context (urgency, scope, complexity) 3. Estimates a deal value within your pricing ranges 4. Displays the estimate in your call log **Example:** A customer calls about a broken water heater needing same-day replacement. The AI identifies "Water Heater Replacement" ($1,200-$3,500) plus "Emergency Service" ($150) from your catalog and estimates the deal at $1,500-$3,650. **Setting Up Deal Value Estimation** To use this feature: 1. **Enable the feature** - Toggle "Automatically estimate deal value" in the AI Automation tab 2. **Optionally build your catalog** - Add Products & Services with pricing ranges for more accurate estimates **How Deal Values Are Estimated:** **With a Products & Services Catalog (Recommended):** * AI matches customer needs to your catalog items * Estimates within your defined price ranges * More accurate and consistent estimations **Without a Catalog:** * AI can still estimate if customers explicitly mention budgets or prices * Example: "I'm looking to spend around $5,000" → AI estimates $5,000 * Less reliable as it depends on customers volunteering pricing information See the complete [Products & Services Guide](/guides/products-and-services.md) for detailed instructions on building your catalog for best results. {% hint style="info" %} **Recommendation:** Set up your Products & Services catalog for the most accurate deal value estimates. The AI works best when it has your actual pricing data to reference. {% endhint %} #### Automatically Detect Caller Name

Automatically Detect Caller Name setting

When enabled, the AI extracts caller names from transcriptions when callers introduce themselves (e.g., "Hi, this is John Smith..."). This supplements [CNAM lookup](/guides/phone-numbers.md#toggling-cnam-lookup) from phone carriers. If both are enabled, AI only overrides the CNAM name when detection confidence exceeds 80%. #### Voice & Audio — ElevenLabs Voices (Optional) At the bottom of the **AI Automation** tab, you'll find a **Voice & Audio** section where you can connect your [ElevenLabs](https://elevenlabs.io) account. ElevenLabs provides ultra-realistic AI voices that Ring Tonic uses for call-flow greetings and IVR prompts — a big upgrade from Twilio's built-in voices.

The Voice & Audio section at the bottom of the AI Automation tab

This is a **bring-your-own-key** integration, so you pay ElevenLabs directly based on your own plan and usage. Leaving the field empty simply hides the ElevenLabs option in the call-flow builder — Twilio's built-in voices remain available. {% hint style="success" %} **Also the AI Agent's voice.** The [AI Agent node](/guides/ai-call-answering.md) speaks in your ElevenLabs voices — the node's Voice tab loads them from this key. Each supported language ships with a curated default voice, and you can pick any voice from your own ElevenLabs library. {% endhint %} **Step 1: Create an ElevenLabs Account** 1. Visit 2. Sign up with your email or Google account 3. Verify your email address 4. Choose a plan that matches your expected usage (free tier available for testing) {% hint style="info" %} ElevenLabs' free tier includes enough characters per month to test voices with a handful of short greetings. For production IVR flows, we recommend at least their Starter plan. {% endhint %} **Step 2: Get Your ElevenLabs API Key** 1. Once logged in, navigate to [Settings → API Keys](https://elevenlabs.io/app/settings/api-keys) 2. Click **"Create API Key"** 3. Give your key a descriptive name (e.g., "Ring Tonic Production") 4. Set permissions (see the table below) — leave everything else on **No Access** for safety 5. Copy the API key immediately — ElevenLabs only shows it once **Required permissions** Ring Tonic only needs two permissions. Keeping the rest disabled limits the damage if the key is ever exposed. | Permission | Setting | Why Ring Tonic needs it | | ------------------ | ------- | ---------------------------------------------------------------------------------------- | | **Text to Speech** | Access | Generates the MP3 audio for each greeting and IVR prompt when you publish a call flow | | **Voices** | Read | Lists the voices on your ElevenLabs account so you can pick one in the call-flow builder | {% hint style="info" %} **Voices only needs Read, not Write.** Ring Tonic never creates, clones, or deletes voices on your ElevenLabs account — it only reads the list of voices you already have. {% endhint %} All other permissions — Speech to Speech, Speech to Text, Sound Effects, Audio Isolation, Music Generation, Dubbing, ElevenAgents, Projects, Audio Native, Voice Generation, Forced Alignment — should stay on **No Access**.

Create a new API key in the ElevenLabs dashboard

{% hint style="warning" %} **Important:** Store your API key securely. If you lose it, you'll need to generate a new one and update Ring Tonic. {% endhint %} **Step 3: Add ElevenLabs API Key to Your Workspace** 1. In Ring Tonic, navigate to your workspace settings 2. Go to the **AI Automation** tab 3. Scroll down to the **Voice & Audio** section 4. Paste your API key into the **ElevenLabs API Key** field 5. Click "Save" or "Update Workspace" Once saved, the ElevenLabs voice option becomes available inside the Greeting and IVR nodes of the call-flow builder. You can pick from your full ElevenLabs voice library when designing each greeting. {% hint style="info" %} **How audio is generated:** ElevenLabs audio is pre-generated when you **publish** a call flow (not on every keystroke), so you only consume API credits on real changes. The generated audio is cached, so published flows serve instantly without hitting ElevenLabs on each incoming call. {% endhint %} **Free tier limitation — "Free users cannot use library voices via the API"** If you're on the ElevenLabs free plan, the API can only synthesize voices that belong to your account. The default "library" voices (Rachel, Adam, Bella, and the other premade voices in the voice picker) require a paid plan to use via API, even though you can preview them in the ElevenLabs web app. You have three options: | Option | Cost | What you get | | --------------------------------- | ---------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | **Upgrade to ElevenLabs Starter** | \~$6/month | All library voices (Rachel, Adam, etc.) become usable via API, plus \~30,000 characters/month | | **Create your own voice (free)** | $0 | Use **Voice Design** in the ElevenLabs web app — describe the voice you want, generate it, save it to your library. Voices you own are API-usable on the free tier. | | **Skip ElevenLabs** | $0 | Switch the Greeting / IVR node's audio provider to **Twilio** or **Amazon Polly**. No ElevenLabs account required. | {% hint style="info" %} If you created your own voice in ElevenLabs but don't see it in Ring Tonic's voice picker, refresh the call-flow editor — Ring Tonic fetches your voice list when the page loads. {% endhint %} {% hint style="success" %} **Clearing the key:** To disconnect ElevenLabs, empty the field and save. The ElevenLabs option will disappear from the call-flow builder and your already-published flows keep working from cache. {% endhint %} {% hint style="success" %} **What's Next?** After setting up AI automation, Ring Tonic will automatically analyze every transcribed call, extract keywords, qualify leads, estimate deal values, and detect caller names. You can view all this intelligence in your call logs and use it to prioritize follow-ups, train your team, and close more deals. {% endhint %} ### 5. Configure Currency, Language & Branding, Timezone Beyond the core integrations, your workspace has a few additional settings that control how Ring Tonic displays information and represents your brand. #### Currency The currency setting determines how money values are displayed throughout Ring Tonic—in call logs, deal value estimates, pricing catalogs, and reports.
**What It Affects:** * **Deal value estimates** - Shows amounts in your chosen currency ($1,500 vs €1,500 vs ₫1,500,000) * **Products & Services pricing** - Your catalog displays in your currency * **Reports and exports** - All financial data formatted correctly #### Language & Region (Locale) The locale setting controls regional formatting preferences for dates, times, numbers, and currency display. It works together with your currency setting to ensure everything displays correctly for your region.
**What It Affects:** * **Date and time formats** - US format (MM/DD/YYYY, 12-hour) vs European (DD/MM/YYYY, 24-hour) * **Number formatting** - Decimal and thousands separators (1,500.00 vs 1.500,00) * **Currency symbol position** - Before ($1,500) or after (1.500€) * **Day/month names** - Language for calendar displays * **First day of week** - Sunday vs Monday in calendars {% hint style="info" %} **Currency + Locale Work Together:** Your currency determines what currency to use (USD, EUR, etc.), while your locale determines how to format it. For example, if you choose USD currency with the de\_DE locale, amounts will show as "1.500,00 $" using German formatting conventions. {% endhint %} #### Timezone The timezone setting controls how dates and times are displayed throughout Ring Tonic—in call logs, analytics, date ranges, and reports. Choose the timezone where your business operates for accurate reporting aligned with your business hours. **Default:** UTC
### 6. Configure Analytics Defaults The **Analytics** tab in workspace settings (**Settings** → **Workspace** → **Analytics**) controls two workspace-wide defaults: the missed-call alert threshold and Google Ads consent fallbacks. #### Missed Call Alert Threshold A slider from 0% to 100%. Tracking numbers whose missed-call rate exceeds this percentage are flagged with a warning icon in analytics — useful for spotting underperforming numbers at a glance. **Default:** 5% Lower the value to be more sensitive (alert on small missed-call rates); raise it to suppress noise from low-volume numbers. #### Default Ad Consent *(Relevant when you use* [*Form Submissions*](/guides/form-submissions.md) *and have Google Ads conversion uploads configured on the Agency plan.)* Google Ads' privacy framework (Consent Mode v2) requires every conversion upload to carry a consent signal for two flags: * **`ad_user_data`** — whether the user consented to their data being sent to Google * **`ad_personalization`** — whether that data can be used for personalized advertising Your customer's page *should* push these signals via `window.dataLayer` or Google Tag Manager at form-submit time. When the page does push them, those signals win. When the page *doesn't* — common on first integration — Ring Tonic falls back to the workspace defaults you set here. | Setting | Use when | | --------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | **Unspecified (no signal)** | You haven't determined the right default yet. Google treats this as the strictest case — no user-level matching, lower conversion quality. | | **Granted** | Your customer has gathered consent off-page (cookie banner, terms-of-service checkbox) and you're confident every visitor whose form lands has consented. | | **Denied** | You operate in a region where ad tracking is opt-in only and the form fill itself doesn't grant consent. Uploads still happen but Google won't match to user-level identifiers. | {% hint style="warning" %} Default to **Unspecified** until you have a documented basis (cookie banner, T\&C) for choosing Granted. Setting Granted without that basis is a compliance risk in GDPR / CCPA / LGPD jurisdictions. {% endhint %} {% hint style="info" %} Page-level signals always override the workspace default. Setting Granted here doesn't force-grant for visitors whose page explicitly pushed Denied. {% endhint %} *** {% hint style="success" %} **You're All Set!** After configuring these basic settings along with your Twilio, transcription, and AI integrations, your workspace is fully configured and ready to track calls, analyze conversations, and generate valuable insights. {% endhint %} {% hint style="info" %} **Agency Plan Users:** If you're on the Agency plan, you can also [invite team members to collaborate on your workspace](/guides/team-management.md). {% endhint %} --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://help.ringtonic.app/guides/setup-workspace.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.