automations.help
Book a teardown

Reviews & UGC

Flag negative reviews for follow-up

A one-star review is a customer telling you something's wrong in the most public place they have. This watches every review as it's submitted, catches the low ones, and drops them in a Slack channel with the rating, the product, what they wrote, and who they are — so someone reaches out while it still matters instead of finding it a week later in the app's dashboard. Build it yourself with the steps below, or we build it, wire the routing, and hand you the keys so you own it with no monthly fee. Free teardown of your current review setup first.

Difficulty
Intermediate
Time to build
1-3 hrs
Updated
  • Slack
  • Klaviyo

The manual way today

Reviews land inside whatever app collects them — Judge.me, Loox, Okendo, Yotpo — and the only way anyone sees a bad one is by opening that dashboard and scrolling. Most teams don't, daily. So a one-star review with a real complaint sits there, public on the product page, until someone happens to check, a customer emails about it separately, or it shows up in a weekly digest nobody fully reads.

Two things bleed here. The first is the customer: someone unhappy enough to leave one star is often still recoverable on day one and gone by day seven — a fast, human 'we saw this, let us fix it' turns a detractor into a repeat buyer surprisingly often, but only if you reach them before they've moved on. The second is the next buyer: that review is sitting on the product page shaping the purchase decision of everyone who reads it, and a complaint you could have resolved or responded to is instead just sitting there, unanswered, doing quiet damage to conversion. Neither shows up in any report, because nobody knows the clock started.

What running looks like

The moment a review is submitted at or below your threshold — say two stars or fewer — a formatted card posts to a dedicated Slack channel: the star rating, the product, the review title and body, the customer's name and email, and a link straight to the review in your app. One review pings once. The worst ones (one-star, verified buyer, repeat customer) route louder than a borderline two-star. A person still owns the reply and the outreach — the automation just makes sure they know in minutes, not days.

01The build

How to build it

  1. Pick where reviews live, and set the bar on paper first

    Needs a human

    Confirm which app actually collects your reviews — Judge.me, Loox, Okendo, Yotpo, and Stamped all qualify — because that app is the source of truth this whole build hangs off. Then decide the threshold before you wire anything: one and two stars is the safe default for 'flag it', three stars is a judgment call (a three-star with an angry body is worth catching; a three-star 'it's fine' usually isn't). Write the rule down — rating is at most 2 — so the flow filter later is just typing in a decision you already made, not making one.

  2. Pipe every review into Klaviyo as a metric event

    Runs itself

    In your reviews app's integrations settings, turn on the Klaviyo connection. Once it's live, each submitted review shows up in Klaviyo as a metric — the exact name depends on the app ('Created Review', 'Submitted Review', 'Reviewed Product') — carrying the properties you need: the star rating, the product name, the review title and body, and the reviewer's email. Submit a test review and confirm it lands under Klaviyo → Analytics → Metrics with the rating attached. If the rating isn't on the event, stop here and fix the integration mapping — every step after this depends on being able to filter by it.

  3. Trigger a Klaviyo flow and filter down to the low ratings

    Runs itself

    Create a metric-triggered flow (Flows → Create flow → choose the review metric as the trigger). Add a trigger filter so the flow only enters profiles where the rating property is at most 2 — set this on the trigger, not as a later branch, so high-star reviews never enter the flow at all and you're not paying flow volume on five-star events you'll discard. This filter is the on-paper rule from step one, made real.

  4. Stand up a Slack Incoming Webhook for one channel

    Runs itself

    Create a Slack app (api.slack.com/apps → Create New App → From scratch), enable Incoming Webhooks, and add one to a dedicated channel like #reviews-followup — not a general channel where it'll get buried. Copy the webhook URL it generates; that single URL is the entire destination Klaviyo needs. Keep it out of any shared doc — anyone with the URL can post to that channel.

  5. Post a triage-ready card to Slack from the flow

    Runs itself

    Add a Webhook action to the flow: method POST, URL set to your Slack Incoming Webhook, and a JSON request body. The simple version is one field — {"text": "..."} — built from Klaviyo event variables: "{{ event.Rating }}★ on {{ event.ProductName }} from {{ person.first_name }} ({{ person.email }}): {{ event.ReviewBody }}". For a card a person can act on at a glance, use Slack Block Kit blocks instead and include a button linking to the review in your app. Give whoever reads it everything they need without opening another tab: rating, product, the actual words, and who said them.

  6. Fire once per review, and only on genuinely new ones

    Runs itself

    Some review apps emit an event again when a customer edits a review or when a merchant replies — you don't want a second Slack ping for the same complaint. Trigger only on the 'created' metric (not an 'updated' one if your app splits them), and set the flow to send once per event rather than re-enqueueing the same profile. Add a trigger filter to exclude your own test and staff emails so QA submissions don't ping the team. One review, one alert.

  7. Assign the follow-up before any automatic reply goes out

    Needs a human

    The point of this is a human reaching out, so the Slack card should make ownership obvious — who's on reviews this week, or a simple 'react with ✅ when you've got it' convention. Resist the urge to fire an instant automated apology: a templated 'sorry you're unhappy' before anyone has read the actual complaint reads worse than silence. If you do want Klaviyo to send a 'we saw your review, can we make it right?' email, put a several-hour delay and a flow filter in front of it so a person can intervene first — or skip the auto-email entirely and keep the outreach genuinely human.

  8. Route the worst reviews louder than the borderline ones

    Needs a human

    Not every flagged review is equal. Branch the flow on the rating and the customer: a one-star from a verified buyer or a repeat/high-LTV customer can post with an @here to an escalations channel, while a borderline two-star posts quietly to #reviews-followup. If you sell across distinct product lines, route by product so the right owner sees their own complaints. Routing is automated; who picks it up stays a person. Then spend ten minutes a week scanning what got flagged — is the bar catching real complaints without crying wolf, and is someone actually replying — and nudge the threshold or the routing from there.

02The stack

What it’s built on

Klaviyo
The hub the review event flows through. Your reviews app already pushes each submission into Klaviyo as a metric, so a metric-triggered flow with a rating filter does the catching, and a Webhook action does the posting — no separate glue tool needed for the core build.Klaviyo's free tier covers a small contact list and includes flows and the Webhook action, so you can build and test this without upgrading. You only outgrow free on email/SMS volume, which this automation doesn't touch.
Slack
Where the flagged review lands and the team acts on it. A dedicated #reviews-followup channel keeps complaints out of general chatter; Block Kit gives you the context card and a deep link straight to the review.Incoming Webhooks are free on every Slack plan, including the free tier — there's no upcharge to receive these alerts.

03Questions

Before you build

My reviews app doesn't push to Klaviyo — can I still build this?

Yes. If the app has a native webhook for new reviews, point it at n8n (or Zapier), filter to low ratings there, and post to the same Slack Incoming Webhook — the Slack half of this recipe is identical. You can also have n8n receive the review and create the event in Klaviyo via its API if you'd rather keep everything centered there. The only hard requirement is that the review's star rating travels with the event so something can filter on it.

Should the customer get an automatic apology email when they leave one star?

Not as the first move. An instant templated apology fired before anyone has read the specific complaint usually lands worse than a thoughtful human reply a few hours later. Let the Slack alert get a person on it first. If you do want an automated touch, add it as a delayed, optional step in the same flow with a filter that lets a human override — don't make the robot apologize before you know what for.

What star rating should count as 'negative'?

One and two stars is the safe default — it's almost always a real problem worth a human reply. Three stars is the judgment zone: a three-star with a frustrated paragraph is worth flagging, a three-star 'it's fine, does the job' usually isn't. Start at two-or-fewer, watch the channel for a couple of weeks, and only widen the net to three if you find you're missing complaints worth catching.

Won't editing a review or replying to it re-trigger the alert?

Not if you set it up to fire on the 'created' event only and send once per event. Some apps emit a separate event on edit or merchant reply — you trigger on the new-review metric specifically and skip those, so a customer tweaking their wording or you posting a public response doesn't ping the channel a second time for a complaint you've already seen.

Does this work for Shopify's built-in reviews or the Shop app?

It works for any review app that emits a per-review event with a rating — which covers Judge.me, Loox, Okendo, Yotpo, Stamped, and most others. Shopify itself doesn't ship a first-party product-reviews engine you'd hook into this way (its old Product Reviews app was retired), and Shop app / store ratings are a separate surface that doesn't expose individual reviews as events. So the rule of thumb is: if your reviews live in a dedicated app, you can flag them; if they only live as an aggregate store rating, there's nothing per-review to catch.

Reviews & UGC

Post new 5-star reviews to Slack

Your worst reviews get found fast — a customer emails, a ticket opens, someone scrolls the dashboard. Your best ones just sit in Judge.me or Loox where nobody looks. This watches every review as it's submitted, catches the five-star ones, and posts them to a Slack channel with the product, the words, the customer, and the photo if there is one — so the team sees the wins as they happen, and the quotable ones get caught before they're buried. Build it yourself with the steps below, or we build it, wire the routing, and hand you the keys so you own it with no monthly fee. Free teardown of your current review setup first.

  • Slack
  • n8n

Intermediate1-3 hrs

Support & returns

Auto-escalate angry support tickets to Slack

A customer threatening a chargeback at 9am shouldn't be discovered at 2pm. This watches every inbound Gorgias ticket, decides which ones are actually angry — not just mildly annoyed — and drops a Slack alert with the message, the customer's history, and a one-click link into the ticket, so the right person sees it in seconds instead of when the queue gets around to it. Build it yourself with the steps below — or we build it, tune the noise out, and hand it over so you own it outright with no monthly fee. Free teardown of your current setup first.

  • Gorgias
  • Slack

Intermediate2-4 hrs

Reviews & UGC

Export product reviews to Google Sheets

Your reviews live inside Judge.me, Loox, Okendo, or Yotpo, where you can read them one at a time but can't easily pivot by product, filter to the angry ones, or hand marketing a list of quotable lines. This pipes every review — the ones you already have and every new one after — into a Google Sheet with a clean column per field, so you can sort, filter, pivot, and share them like any other data. Build it yourself with the steps below, or we build it, backfill your history, and hand you the keys so you own it with no monthly fee. Free teardown of your current review setup first.

  • Google Sheets
  • n8n

Intermediate2-4 hrs