Guides and answers for everything in ReplyLabs
ReplyLabs is available on the Google Workspace Marketplace. To install it, open the Marketplace and search for "ReplyLabs", or ask your Google Workspace administrator to install it for your domain. Click Install and accept the permission prompts. The permissions allow ReplyLabs to display a sidebar in Google Sheets, read and write data in the current spreadsheet, and make network requests to our servers. Once installed, you can access ReplyLabs from the Extensions menu in any Google Sheet.
Open any Google Sheet and go to Extensions → ReplyLabs → Open Sidebar. The sidebar appears on the right. You'll see a login form where you can enter your email and password, or click "Sign in with Google" for passwordless access. If you don't have a ReplyLabs account yet, click "Create account" to sign up. New accounts receive $20 in free credit immediately. After signing in, the sidebar loads your available AI models and current balance.
Every new ReplyLabs account receives $20 in free credit. This is real money that you can spend on any operation: AI prompts, web scraping, or email verification. With GPT-4o-mini, $20 gets you approximately 7,000 short prompts or 3,000 medium-length prompts. Credits never expire. You can check your current balance at any time in the sidebar (top-right corner) or on the dashboard at replylabs.io/dashboard/billing.
The ReplyLabs sidebar has five tabs at the top: AI Prompt (the main tab for running AI models), Scrape (for extracting data from websites), Verify (for email verification), History (view past batch jobs), and Settings (manage your account and API keys). Below the tabs, you'll see your current balance and the selected AI model. The model picker dropdown shows all 140+ available models sorted by cost, so the most affordable options appear first.
ReplyLabs offers 140+ AI models from six providers. Here's a quick guide:
The quality of your results depends entirely on your prompt. Proven techniques:
The system prompt sets the AI's identity and rules for the entire batch. It's like giving the AI a job description before it starts work. Enter your system prompt in the "System Prompt" field (click to expand it if hidden). Example: "You are a senior SDR at a B2B SaaS company specialising in supply chain software. You write brief, specific, personalised cold emails. You never use filler phrases. You always reference something concrete about the prospect." Every row in your batch will follow these rules.
AI pricing is based on tokens — roughly 4 characters per token, or about 750 words per 1,000 tokens. Each prompt has an input cost (your prompt + row data) and an output cost (the AI's response). The cost estimate shown in the sidebar before you run is calculated from the model's per-token pricing and the average size of your selected rows. Actual costs may vary slightly based on response length. You're never charged more than your available balance.
ReplyLabs can extract structured data from any public web page. Go to the Scrape tab in the sidebar. You'll need a column containing URLs in your spreadsheet. Select which fields you want to extract: page title, meta description, headings (H1/H2), body text, links, or contact information (emails and phone numbers found on the page). Select your rows and click Scrape. ReplyLabs visits each URL, parses the HTML, and returns the requested fields into adjacent columns.
Not all websites are equally scrapable. Sites that require JavaScript rendering (single-page applications) or that are behind login walls may return incomplete data. For best results, target public marketing pages, company about pages, blog posts, and product pages. If a URL returns no data, try the homepage instead of a deep link. Scraping respects robots.txt where applicable. Do not use scraping to extract data from websites whose terms of service prohibit it.
Each URL scraped costs a flat rate deducted from your balance. The cost per scrape is shown in the sidebar before you run the job. Scraping is generally cheaper than AI prompts because it doesn't use a language model — it simply fetches and parses HTML.
ReplyLabs validates email addresses by checking multiple signals: DNS MX record existence (does the domain accept email?), SMTP handshake response (does the mailbox exist?), disposable domain detection (is this a temporary email service?), catch-all detection (does the domain accept all addresses?), and syntax validation. Results are returned as a status: deliverable (safe to send), risky (may bounce), or undeliverable (will bounce). Verification uses the No2Bounce API under the hood.
Deliverable: the email address exists and is accepting mail. Safe to include in your outreach campaigns. Risky: the address may exist but showed uncertain signals. The domain might be a catch-all or the mailbox may be full. Consider sending but monitor bounce rates. Undeliverable: the address does not exist, the domain has no mail server, or the mailbox is permanently unavailable. Remove these from your list to protect your sender reputation.
High bounce rates damage your email sender reputation. Major email providers (Gmail, Outlook) track your bounce rate and will throttle or block your emails if it exceeds 2–3%. Verifying your list before outreach ensures you only send to real addresses. This protects your domain reputation, improves deliverability, and saves money on email sending tools.
Go to replylabs.io/dashboard/billing and click "Add Funds". Choose an amount and complete the checkout via Stripe. Funds appear in your balance immediately. All payments are processed securely by Stripe — we never see or store your card details.
Starter (free, pay-as-you-go): $20 free credit on signup. Operations priced at cost + small platform fee. Up to 3 team members. Perfect for trying out ReplyLabs.
Pro ($19/month): Everything in Starter, plus bring your own AI keys (BYOK) for free AI operations, up to 10 team members, and priority email support. Best for individuals and small teams doing regular outbound.
Scale ($39/month): Everything in Pro, plus BYOK for all services (AI, scraping, verification), unlimited team members, and dedicated support. Best for sales teams running high-volume campaigns.
On Pro and Scale plans, go to replylabs.io/dashboard/settings. Click "Add API Key" and select the provider (OpenAI, Anthropic, Google, DeepSeek, Mistral, or xAI). Paste your key and click Save. The key is encrypted with AES-256-GCM before storage — we can never see the raw key after you save it. When you run a prompt using a provider for which you have a key, the operation is routed through your key and no credits are deducted from your ReplyLabs balance. You pay the provider directly at their standard rates.
Enable auto-recharge in the billing dashboard to automatically add funds when your balance drops below a threshold you set. This prevents batch jobs from stopping mid-way due to insufficient balance. You choose the recharge amount and the threshold. Auto-recharge is processed via the same Stripe payment method on file.
Go to replylabs.io/dashboard/team and enter the email address of the person you want to invite. Choose their role: Admin (can manage billing and settings) or Member (can use the extension only). They'll receive an email invitation with a link to join your organisation. Once they accept, they'll share your organisation's credit balance, API keys, and usage history.
Update your name, email, or password at replylabs.io/dashboard/settings. If you signed up with Google, you can add a password for email-based login from the same page.
To delete your account, go to replylabs.io/dashboard/settings, scroll to the bottom, and click "Delete Account". This permanently removes your profile, API keys, usage history, and transaction records. If you are the only member of your organisation, the organisation and its balance will also be deleted. This action cannot be undone.
If ReplyLabs doesn't appear in the Extensions menu: try refreshing the Google Sheets page (Cmd+R or Ctrl+R). If it still doesn't appear, go to Extensions → Add-ons → Manage add-ons and verify that ReplyLabs is listed and enabled. If it's not listed, reinstall from the Google Workspace Marketplace. Note that add-ons can take a few seconds to load after opening a sheet.
This usually means the sidebar JavaScript failed to load. Try closing the sidebar and reopening it (Extensions → ReplyLabs → Open Sidebar). If the issue persists, clear your browser cache (Cmd+Shift+Delete) and reload the page. If you're using a corporate network, your IT department may be blocking connections to replylabs.io — ask them to whitelist the domain.
If you can't sign in: verify your email and password are correct. Try the "Forgot password" flow at replylabs.io/forgot-password. If you signed up with Google, use the "Sign in with Google" button instead. If you see "Too many login attempts", wait 15 minutes and try again (this is rate limiting to protect your account). If nothing works, email hello@replylabs.io with your account email.
If a batch job appears stuck: check your internet connection. Large batches (1,000+ rows) can take several minutes. The progress bar should update every few seconds. If it stays at 0% for more than 2 minutes, close the sidebar, reopen it, and check the History tab — the job may still be processing in the background. If the job shows as "failed", check the error message. Common causes: insufficient balance (add funds), rate limit exceeded (wait a minute and retry), or the AI model is temporarily overloaded (try a different model).
If the batch completes but results don't appear in your sheet: check that you haven't moved or renamed the columns since starting the job. Results are written to the column immediately to the right of the selected input column. If you've added or removed columns in the meantime, the results may have been written to the wrong column. Also check that you haven't accidentally protected the target cells — protected ranges prevent the extension from writing.
AI costs depend on both input tokens (your prompt + row data) and output tokens (the AI's response). If your rows contain long text (like full article bodies), the input cost per row will be higher. The cost estimate shown before running is based on the average row size in your selection. To reduce costs: use a cheaper model (GPT-4o-mini), shorten your prompt, select fewer output columns, or add "Keep your response under 50 words" to your prompt.
Our team responds within 24 hours. Email us at hello@replylabs.io or visit our support page.