Hazel reads every Fairing response (attribution answers, NPS scores, write-ins, custom questions) and answers business questions in plain English. No dashboards to build.
May 23, 2026
Hazel is an AI coworker that connects to your Fairing account and answers business questions about your post-purchase survey data the way an analyst would. Once connected, Hazel ingests every Fairing response. "how did you hear about us?" attribution answers, NPS scores, custom question responses, and free-text write-ins, and refreshes every few hours. Ask Hazel how podcast attribution trended week over week, what percentage of customers selected "other" and what they actually wrote in, or how Fairing-reported channels compare to your UTM and ad-platform numbers, and you get an answer with the numbers behind it. Hazel joins your Fairing responses to Shopify orders, so attribution answers tie back to real revenue, AOV, and LTV, not just response counts. For consumer brands running Fairing, Hazel replaces the work of exporting survey CSVs and pivoting them in a spreadsheet, and it pairs Fairing data with your ad platforms, email, and subscriptions so cross-source attribution questions are a single prompt.
Hazel reads the Fairing data that matters for analyzing attribution and customer sentiment, and lets you query it conversationally:
Hazel keeps every response historically, so YTD comparisons and multi-month attribution trends are one prompt.
Common questions consumer brands ask about their Fairing data:
"What is our Fairing response rate for the last full month, and how is it trending?"
"How did Fairing channel-reported data change week over week?"
"Look at the breakdown of new customer acquisition by channel and compare UTM source vs. the Fairing 'how did you hear about us?' answer. Where do the two diverge?"
"I want to look at users who selected 'other' for 'how did you hear about us?'. Of those, give me the distribution of write-in responses broken down by month, ranked by frequency."
"For customers who attributed to a specific event or pop-up in the Fairing write-in, pull their LTV cohort and compare it to UTM-attributed customers from paid social."
"Which influencer codes show up most in Fairing write-ins, and what's the 90-day repeat-purchase rate for those customers vs. paid-social-attributed customers?"
"Of customers who said 'in-store' as their acquisition channel, what percentage went on to place a second order online within 60 days?"
"Which Fairing survey responses have the highest repeat-purchase rate at 90 days? I want to see which acquisition channels actually retain."
Grab your Fairing API key from Account Settings in Fairing, paste it into Hazel, and you're done. No app install, no OAuth handshake, no Shopify side-config required.
Every survey response, including the question shown, the answer selected, any free-text write-in, the order ID it's tied to, and the timestamp. Hazel pulls the full historical record on first sync, so YTD and multi-year trends are queryable immediately.
Roughly every 6 hours, automatically.
Hazel ties every Fairing response back to the underlying Shopify order, so you can analyze AOV, repeat rate, and LTV by attributed channel, not just response counts. If Fairing data also lands in your Shopify order metafields (some brands route it that way), Hazel reads it from both sources and reconciles.
API key, stored encrypted. Hazel only reads from Fairing; it never sends survey questions, edits responses, or writes back to your account.
Two community Fairing MCPs ship today. Xmayanksehgal's open-source `fairing_mcp` on GitHub and Vinkius's hosted Fairing MCP (12 tools, Cursor/Claude/LlamaIndex compatible). Both connect an LLM directly to your Fairing API for raw survey access: useful for one-off questions, less useful if you need cross-source joins to Shopify, Meta, or Klaviyo.
Rather than connecting an LLM directly to Fairing's API and writing the join-to-Shopify logic yourself, Hazel ingests Fairing responses into an analytical store, joins them to your Shopify orders so every response carries AOV and LTV context, and joins both to your ad platforms, email, subscriptions, and reviews. The same agent answers cross-source questions like "how does Fairing-reported podcast attribution compare to Meta-attributed conversions on the same week". Which a Fairing-only MCP can't reach.
If you specifically want MCP access to Hazel itself, that's available too. Ping us at https://calendly.com/clint-dunn/clint-hazel-intro.
Fairing ships two AI and data surfaces worth knowing about. First, AI Insights: a weekly attribution brief that scans 40+ data points and surfaces week-over-week changes automatically. That's the right tool for a recurring pulse check inside Fairing. Second, Fairing's Total Impact Attribution Model: an ML model that combines first-party pixel data with post-purchase survey data for blended attribution. Hazel complements both: it gives your team on-demand, cross-source answers (not weekly summaries) and joins Fairing-attributed channels to Shopify revenue, ad spend, email, and subscription data in a single prompt.
Fairing data now syncs into Shopify Analytics natively. Useful for basic grouping by product type or country. Hazel extends that: cross-source joins to Meta, TikTok, Klaviyo, and Recharge; full write-in text analysis; multi-month trend reads; and LTV per attributed channel that Shopify Analytics doesn't model.
Yes. Paste your Fairing API key into Hazel and your full response history syncs within hours. No app install required.
Fairing's AI Insights summarizes attribution patterns inside Fairing on a weekly cadence. Hazel is conversational and cross-source: ask any question about your survey data on demand, join it to Shopify orders so you see revenue and LTV by channel (not just response counts), and compare Fairing-attributed channels to your UTM, Meta, TikTok, and Google Ads data in the same answer.
Total Impact is Fairing's ML model that blends pixel data with survey data for attribution scoring. Hazel doesn't replace it. Hazel is the cross-source analytics layer that lets your team ask questions about Total Impact's outputs alongside Shopify revenue, ad spend, email engagement, and subscription data.
Polar, Triple Whale, and Shopify Analytics are dashboard-first. They show Fairing data in a chart. Hazel is analyst-first: you ask a question in plain English, Hazel writes the analysis, joins Fairing responses to Shopify orders and your ad platforms, and returns a written answer with the numbers behind it. If your team wants a Fairing attribution dashboard, use Polar or Triple Whale. If your team wants to ask "why did podcast attribution drop 15% this month and what happened to the LTV of those customers," use Hazel.
Roughly every 6 hours.
No. Hazel only reads. We never send questions, edit responses, or modify your Fairing setup.
Yes. Write-ins are queryable. You can ask Hazel to pull the distribution of write-in responses by month, summarize them into themes, or find PR and influencer mentions buried in "other."
Yes. Hazel reads the raw write-in text and surfaces specific named mentions, so if customers typed "I heard about you on Lex Fridman's podcast," Hazel returns "Lex Fridman," not "podcast (write-in)."
Yes. Every response is linked to the underlying Shopify order, so you can analyze AOV, repeat rate, and LTV by Fairing-attributed channel, not just response volume.
Hazel reads both. Some brands route Fairing's "how did you hear about us?" answer into a Shopify order metafield; Hazel picks up the metafield version automatically and reconciles it with the direct Fairing pull.
Yes. If you also connect Meta, TikTok, Google Ads, or your UTM data, Hazel can show Fairing-reported channel splits next to platform-claimed conversions for the same week, product, or cohort, and surface where the two diverge, so your team can investigate discrepancies rather than just staring at two different numbers.
Book a call and we'll walk through your Fairing setup, the attribution questions you want to answer, and what the rollout looks like.
Hazel helps your team uncover new revenue, catch problems early, and make smarter, faster decisions before you even know what to ask.