From
Demand Radar, a Shopify app in development
To
You, every Monday at 7:00 in your store's timezone

Subject

What did people try to buy from you last week that you couldn't sell them, and what is it worth?

Searches that return nothing, back-in-stock requests, shoppers landing on sold-out pages from your ads, customers asking support for a size you don't stock. Demand Radar adds those signals up, puts a dollar range on each thing you could act on, and sends the answer as this report. Below is a sample.

Week of October 27 · sample data

Estimated demand you couldn't serve last week

$2,700–4,800

Across 5 lines, about the same as the four weeks before. A range, not a promise: it narrows as more kinds of signal agree.

1Restock now

Linen Shirt / Sage / L41 back-in-stock requests · 212 sold-out page views · 9 ad clicks landed on itabout $1,400–2,300/week

2Variants you don't carry

Linen Shirt in XXLasked 14 times in search and 6 times in supportabout $520–970/week

3Products you don't make

“linen trousers”87 searches with no results, grouped with “linen trouser” and “trousers linen”about $640–1,300/week

4Money leaking from ads

summer-linen · google9 paid clicks landed on sold-out itemscounted in Restock now

5What changed

Canvas Tote / Naturalrestocked Oct 21 · estimated 9–16 sold in 14 days · sold 12within the range

How to read the numbers

Each line's range comes from weighted signals, your store's own conversion rate for that kind of product, and its price. “3 of 5 signals agree” means three different kinds of evidence point at the same thing; a single stray view never becomes a line. When there isn't enough data yet, the report says so and shows the last 30 days instead.

Hear when it opens

How a line gets onto the report

  1. Capture

    A Shopify web pixel and an app embed, switched on once. No theme code is edited.

  2. Match

    Each signal is tied to a product and variant, or grouped with searches for the same missing thing.

  3. Estimate

    Weighted by signal, times your store's own conversion rate and the price. Always a range.

  4. Report

    Monday at 7 in your store's timezone, by email and in the app. Mark lines planned, done or ignored.

  5. Check

    After a restock, what sold in the next 14 days is set against the estimate.

The same lists, inside Shopify

Every line opens to its evidence: the counts by signal, the actual search strings, and when it was first seen. Give each one a status so the list stays clean. Ignored lines leave the headline and the email, and come back if you ask.

Drawn for this page with sample data, not a screenshot. Use the tabs.

Last week

$2,700–4,800

Estimated demand you couldn't serve last week, across 5 lines. A range, not a promise: it narrows as more kinds of signal agree.

Sep 29 Oct 6 Oct 13 Oct 20 Oct 27

About the same as the four weeks before.

  • Linen Shirt / Sage / L

    Sold out

    41 back-in-stock requests · 212 sold-out page views · 9 ad clicks landed on it

    $1,400–2,300/week

    3 of 5 signals agree

    New

  • Canvas Tote / Natural

    Sold out

    18 sold-out page views · 7 blocked adds · 4 support asks

    $480–860/week

    3 of 5 signals agree

    Planned

  • Linen Shirt / Navy / M

    Sold out

    26 sold-out page views · 5 back-in-stock requests

    $210–440/week

    2 of 5 signals agree

    New

  • Linen Shirt in XXL

    No variant with this size

    Asked 14 times in search and 6 times in support

    $520–970/week

    2 of 5 signals agree

    New

  • Canvas Tote in Black

    No variant with this colour

    Searched 11 times

    $90–210/week

    1 of 5 signals

    New

  • “linen trousers”

    No matching product

    87 searches with no results: “linen trousers”, “linen trouser”, “trousers linen”

    $640–1,300/week

    1 of 5 signals

    Ignored

CampaignSourcePaid clicks
summer-linengoogle9
tote-retargetmeta4

Paid clicks that landed on sold-out items. They already count in the restock lines above, so there's no second dollar figure here.

The fine print

On your storefront

The app embed passes events to Shopify's web pixel and never calls us itself. The pixel sends about one small request per page, carries no names, emails or phone numbers, and sends nothing before analytics consent where consent is required. Each visit gets a random id that isn't linked to a customer.

In your admin

Reads products, variants and inventory, and orders as daily totals per variant. Order details aren't kept. It doesn't edit your theme, products or inventory, and it never emails your customers.

Only if you connect them

Your back-in-stock app, your helpdesk, your ad accounts. Support messages are kept as the product, the size or colour asked for, and a short snippet with emails and phone numbers stripped.

Does it email my customers?

No. Demand Radar doesn't send back-in-stock emails, run your store search or manage inventory. It reads the signals your storefront and your other apps already produce and adds them up. Keep your back-in-stock app; Demand Radar makes its requests count for more.

Will it slow my store down?

No theme code is edited. Capture runs through a Shopify web pixel, which runs in Shopify's sandbox, and an app embed you switch on once. The embed stays under 5 KB and does nothing on pages other than product pages. Signals go out in about one small request per page.

What does it record about shoppers?

What they tried to buy, not who they are: a committed search and how many results it returned, a view of a sold-out variant, an add-to-cart that couldn't go through, and the campaign tags on a landing. No names, emails or phone numbers. Emails and phone numbers typed into search are removed. Each visit gets a random id that isn't linked to a customer or to other visits, and nothing is sent before analytics consent where the law requires it.

How is the dollar figure worked out?

Each signal is weighted (a support ask counts for more than a single sold-out view), multiplied by your own store's conversion rate for that kind of product, and by the price. It's always shown as a range, with how many of the five signals agree. A line needs corroboration, so one stray view never becomes a line.

My store is small. Will the first report be useful?

It will tell you plainly if there isn't enough data yet, and show the last 30 days instead of the last week. It won't dress up a thin week as a finding.

Does it read my support inbox?

Only if you connect it, by forwarding your support mail (Gmail, Outlook or any inbox) to your own Demand Radar address. Gorgias and Zendesk connections are coming. Messages are filtered by keyword first (“do you have”, “restock”, “size”) and the ones that pass are classified by an AI model. Demand Radar keeps the product, the size or colour asked for, and a short snippet with names, emails and phone numbers taken out. Disconnect, and the snippets are deleted.

Which search does it understand?

Shopify's own storefront search, with the number of results each search returned. Searches in Boost, Searchanise and Klevu count too, though without their result counts, since those apps draw their own results.

What will it cost?

Starter is $29 a month and Growth $59, billed through Shopify. Paying yearly costs ten months' price. Every plan has a 14-day free trial, long enough for the first Monday report to land with real data.

In development

Coming to the Shopify App Store.

Demand Radar is being built and tested on our own development store. Leave your email and we'll write once, when it opens.

One email when Demand Radar opens. Nothing else, and never shared.