Estimate what it costs to run Generative AI, 5-provider Live Web Search, Decision Mode ($1.00/1k rows), email verification, multi-page website or LinkedIn enrichment, and 1-to-1 Hosted Links across Claude/ChatGPT MCP Hosted Lists and Google Sheets. Every new workspace starts with up to $20 free credit ($5 instant + $15 with Auto-Reload).
DEC-301 unified baseline: 1,000 input + 150 output tokens/row at 1.65x + $0.0035/row base fee.
AI enrichment
$3.90
$0.0039 / row
Email verification
$10.00
$0.01 / email
Enrich (Sorzio Enrich)
$6.00
Includes up to 6 pages at $0 extra
Only succeeded rows are charged ($0 on failed URLs or #not_found personal LinkedIn lookups). Sorzio Enrich ($0.006/URL) fetches up to 6 company website pages (Home, about, products, case_studies, pricing, and careers) at $0 extra. On Pro ($39/mo flat, 1 seat) and Scale ($99/mo flat, 3 seats), bring your own AI key so generative AI runs at 0% platform markup and no base fee (Decision Mode always bills from workspace balance). New accounts get up to $20 free credit ($5 instant + $15 with Auto-Reload).
How Sorzio per-row pricing works across Google Sheets and MCP
Spreadsheet formulas and credit-bundle GTM platforms make outbound unit economics hard to predict because cell recalculations, failed scrapes, and hidden provider multipliers silently burn credits. Sorzio uses deterministic, USD-denominated per-row formulas across both the Google Sheets sidebar and our 39-tool remote MCP server, and every adapter settles charges strictly on succeeded rows.
For Generative AI across 300+ models from OpenAI, Anthropic, Google Gemini, Mistral, and xAI, workspace-funded runs bill 1.65x actual provider token cost plus a $0.0035 per-row base fee. Pre-flight estimates use a unified 1,000 input token and 150 output token baseline so multi-column prompts and enriched company text are reflected honestly before execution. When you enable Live Web Search, Perplexity Sonar runs all-in at roughly $8.78 per 1,000 rows, while OpenAI, Anthropic, Gemini, and Grok web search tools add their published search call surcharge at 1.65x.
For high-volume ICP qualification, Decision Mode (typesafe-ai/jev-1.14.0) replaces expensive generative LLM calls with a calibrated non-generative evaluation model priced at $1.00 per 1,000 rows ($0.0010/row base fee plus 1.25x input tokens). Each row receives a calibrated probability score with 3-way keep, reject, and unsure routing so downstream steps only process qualified accounts.
For data enrichment and deliverability, Sorzio Enrich fetches up to 6 company website pages (homepage plus /about, /products, /case-studies, /pricing, and /careers at 6 pages at $0 extra) for $6.00 per 1,000 URLs ($0.006/URL), with optional Jina ($5.00/1k) and Firecrawl ($7.50/1k) engines. Company LinkedIn pages run at $7.50 per 1,000 rows, Personal LinkedIn (/in/) profiles run at $10.00 per 1,000 succeeded rows ($0.00 on miss), Email Verification runs at $10.00 per 1,000 emails ($0.01/email), and Hosted Links cost $0.00 to publish plus $0.15 once on the first verified human open.
Worked batch cost benchmarks: 1,000, 10,000, and 25,000 rows
Use the reference matrix below to compare standalone operation spend across 1,000, 10,000, and 25,000 succeeded rows before applying multi-step Flow filter gates or BYOK keys.
Operation
Unit formula
1,000 rows
10,000 rows
25,000 rows
Generative AI (GPT-4o mini, 1k in / 150 out)
~$0.0039 / row (1.65x + $0.0035)
$3.90
$38.96
$97.40
Pro / Scale BYOK AI (0% Sorzio markup)
$0.00 Sorzio fee (raw provider key)
$0.00 ($0.24 raw)
$0.00 ($2.40 raw)
$0.00 ($6.00 raw)
Live Web Search (Perplexity Sonar)
~$0.0088 / row (~$8.78 / 1k)
$8.78
$87.80
$219.50
Decision Mode ICP Gate (TypeSafe JEV 1.14)
$1.00 / 1k base + 1.25x input tokens
$1.01
$10.08
$25.20
Sorzio Enrich (6-page website bundle)
$6.00 / 1k ($0.006 / URL, 6 pages at $0 extra)
$6.00
$60.00
$150.00
Company LinkedIn (/company/)
$7.50 / 1k ($0.0075 / succeeded row)
$7.50
$75.00
$187.50
Personal LinkedIn (/in/)
$10.00 / 1k ($0.01 / succeeded row, $0 on miss)
$10.00
$100.00
$250.00
Email Verification
$10.00 / 1k ($0.01 / verified email)
$10.00
$100.00
$250.00
1-to-1 Hosted Links (example 20% human open rate)
$0.00 publish + $0.15 first human open
$30.00 (200 opens)
$99/mo Unlimited
$99/mo Unlimited
Multi-step Flow savings example: Suppose you upload 10,000 raw prospects. Instead of running Personal LinkedIn enrichment ($100) and Generative AI ($38.96) on all 10,000 rows, run Sorzio Enrich 6-page websites ($60.00) + Decision Mode ICP gate ($10.08) first. If 35% of accounts pass your ICP threshold (3,500 rows), verifying emails ($35.00), enriching Personal LinkedIn ($35.00), and generating personalized copy on Pro BYOK ($0 Sorzio fee, ~$0.84 raw OpenAI) brings the entire 10,000-lead funnel to roughly $140.92 in usage.
When to stay on Free vs upgrade to Pro ($39/mo) or Scale ($99/mo)
Every Sorzio workspace starts on the credit-first Free plan ($0/month) with up to 5,000 rows per run, all 39 MCP tools, 2-hour Hosted Lists retention, and up to $20 welcome credit ($5 instant at signup + $15 when you turn on Auto-Reload with a verified card or deposit $20+). Upgrading to Pro ($39/month flat, 1 seat) unlocks 0% Sorzio markup on your own AI API keys, 10,000 rows per run, Scheduled Flows, and 7-day Hosted Lists, paying for itself around 10,500 generative AI rows per month. Scale ($99/month flat, 3 seats included, +$19/mo per extra seat) adds 25,000 rows per run, Forever Hosted Lists retention, and BYOK across supported enrichment and verification providers, plus a +10% bonus credit on single balance reloads of $1,000 or more.
Frequently asked questions about AI and enrichment batch costs
How does Sorzio calculate Generative AI cost per row?
On workspace balance, Generative AI rows bill 1.65x the underlying model provider's actual input and output token cost plus a $0.0035 per-row base fee. Pre-flight estimates use a realistic 1,000 input token and 150 output token baseline so enriched website text and system instructions are accounted for before you click Run. On Pro ($39/month flat) and Scale ($99/month flat), connecting your own API key (BYOK) removes both the 1.65x multiplier and the $0.0035 base fee (0% Sorzio markup).
What happens if an email verification, website scrape, or LinkedIn lookup fails?
You pay $0.00. Every adapter in Sorzio bills succeeded rows only. Dead company domains, HTTP timeouts, undeliverable network errors, and #not_found Personal LinkedIn (/in/) lookups are automatically refunded or excluded from final settlement.
How much does Decision Mode cost compared with standard LLM qualification?
Decision Mode runs on typesafe-ai/jev-1.14.0 at a $0.0010 per-row base fee ($1.00 per 1,000 rows) plus 1.25x input tokens (around $1.01 per 1,000 rows total). Because it produces calibrated 0 to 1 probabilities with keep, reject, and unsure thresholds, placing a Decision Mode gate early in a multi-step Flow lets you filter out non-ICP accounts before spending on Personal LinkedIn enrichment or personalized copy.
How are 1-to-1 Hosted Links billed?
Publishing personalized Hosted Links from a Hosted List or Google Sheet costs $0.00 upfront. Without the Unlimited Add-On, you are charged $0.15 once per link only when a verified human opens the page (bot and scanner prefetches are filtered out). High-volume teams can add Hosted Links Unlimited + 1 Custom CNAME Domain to any plan for $99/month.