claude skills / E-commerce pack / bilan-hebdomadaire / en français

Weekly Review.

One question on Monday morning, an answer in three sentences, and a decision with a name next to it.

See what it produces
  • Shopify
  • WooCommerce
  • Google Analytics 4
  • Meta Ads Manager
  • Email marketing tool
3 sentencesto answer: what changed, why, what we do
8metrics compared with the previous week and with last year
1alert at most, and one decision for Monday

orders of magnitude, depending on your volume

the problem

What gets in the way today.

  • 01The dashboard is too long, and nobody reads it.
  • 02A gap of a few orders triggers a meeting, when it is often just chance.
  • 03Actions decided three weeks ago have vanished, and nobody knows if they worked.

input, output

You give this. You get this.

This week's exports and what happened

Atelier Lisière, Shopify. Week of 9 to 15 November, the week before and the same week in 2025.
ventes_semaine.csv, sessions_semaine.csv, tunnel_appareil.csv, screenshot of the ads manager.
What happened: free shipping threshold shown on mobile product pages on 12/11.
Fig stock: 41 units, threshold set at 50 for Black Friday.

$ claude "How did the week go?"

Week of 9 to 15 November 2026example

What changed this week?

Net sales rose by 18% (€3,186 vs €2,700), thanks to the conversion rate climbing back to 1.45% after the free shipping threshold was shown on mobile product pages (inference). Traffic is stable. We keep the setting and prepare stock for Black Friday week.

Alert: the Fig candle is down to 41 units, below the threshold of 50 set for Black Friday.

Decision for Monday: pour a batch of 120 Fig candles, owner: the workshop, we'll track stock and Fig sales.

The metrics

Net sales: €3,186, +18.0% week on week, +9.5% vs last year

Orders: 59, +9.3%

Sessions: 4,070, -1.9%

Conversion rate: 1.45%, +11.5% relative

Average order value: €54.00, +8.0%

Ad cost per order: €21, -12.5%

What explains the gap

Fact: on mobile, cart to checkout is 47% vs 39%.

Inference: the threshold shown since 12/11 explains the rise, nothing else changed in the funnel.

Low volume: 59 orders, to be confirmed over 4 weeks.

Action log

Shipping threshold visible on mobile, done 12/11, cart to checkout from 39% to 47%

Abandoned cart email, done 07/11, 14 emails, 3 orders recovered

Instagram ad retargeted, in progress, cost per order from €24 to €21

Missing export: sessions by source for the same week in 2025.

the method

What Claude does, in order.

  1. step 1Check

    Same dates in every export, test orders removed, missing exports named.

  2. step 2Calculate the 8 metrics

    Sales, orders, sessions, conversion, average order, returning customers, top source, cost per order.

  3. step 3Find what changed

    Traffic, conversion or average order, then the source, product or device, and stop once the cause is found.

  4. step 4Answer and decide

    Three sentences, an alert if a threshold is crossed, a decision with an owner, and the action log.

demo in French / click, try it

Your turn.

An interactive demo of the skill. It is in French, the pack works in any language: ask Claude to answer in English.

La question

the film / 44 seconds, with sound

The pack film.

the pack film

In 8 seconds.

the guardrails

What the skill never does.

  • Only the store's numbers: a missing figure is noted "not provided" with the export to add.
  • A gap is only discussed if it exceeds the threshold agreed with you on first use.
  • Small volumes and peak weeks are flagged before any conclusion.
  • One alert at most, one decision with an owner, and last week's actions carried over.

Questions

Which alert thresholds does the skill use?

Starting settings to approve on first use, for example plus or minus 15% on sales and 10% on conversion. They tighten as volume grows.

What if I don't have last year's history?

The review compares with the previous week and explains why the second comparison is missing.

Can I change the 8 metrics?

Yes. You can replace a metric, for example email open rate or reviews received, keeping 8 at most.

What if the week reveals a real problem?

The review suggests running the Store Audit skill for a deeper analysis.

the pack

In the pack: E-commerce pack.