Sales Data & Competitor Research

Turn exported numbers and scattered research into something you can act on

Claude doesn’t have your sales data — you bring it

There’s no live connection between your Shopify or WooCommerce analytics and Claude. What you can do is export a report (weekly sales by product, traffic by source, whatever your platform gives you), paste the relevant numbers in, and ask Claude to help you see the pattern faster than scanning rows yourself would. This is genuinely one of the highest-leverage uses in this whole course, because most store owners have more data than time to look at it.

Preparing data before you paste it

Two things make analysis prompts work well: aggregating first (don’t paste 2,000 raw order rows — paste weekly or product-level totals), and stripping PII per Lesson 2 before anything goes in.

Here's my weekly sales by product category for the last 8 weeks
(units sold, revenue):

Week 1: Tents 12/$2,388, Sleeping Bags 34/$3,060, Backpacks 8/$960
Week 2: Tents 15/$2,985, Sleeping Bags 29/$2,610, Backpacks 11/$1,320
...
Week 8: Tents 22/$4,378, Sleeping Bags 18/$1,620, Backpacks 19/$2,280

What patterns do you notice? Which category is trending up or down,
and is there anything that looks like it might be seasonal versus a
real shift?

Claude is good at spotting trends, ratios, and outliers in data like this — but treat its read as a hypothesis to check against what you actually know about your business (a promotion you ran, a seasonal shift, a stockout), not a verified conclusion.

Warning: Claude can’t verify your numbers against anything else — it only sees what you paste in. If a row has a typo or a category is mislabeled, Claude will analyze the mistake as if it were true. Double-check your source data before drawing conclusions from the analysis.

Questions worth asking of your own sales data

  • “Which products have high traffic but low conversion — worth investigating for pricing or listing issues?”
  • “Is there a pattern in what gets returned or refunded most?”
  • “Based on this data, which 3 products would you prioritize restocking first?”
  • “Does this look like normal week-to-week variation, or an actual trend?”

Structuring competitor research

Claude doesn’t automatically know what your competitors are doing right now — it can’t browse their live site on its own. What it’s genuinely useful for is turning research you’ve gathered (screenshots, pasted text, notes from browsing their site) into a structured comparison you can actually use, rather than a pile of open tabs.

I'm comparing my store to 2 competitors selling similar camping gear.
Here's what I found on each (pasted from their sites):

MY STORE: Tents $180-$340, free shipping over $75, 30-day returns
COMPETITOR A: Tents $150-$310, free shipping over $50, 60-day returns,
  heavier emphasis on sustainability messaging
COMPETITOR B: Tents $200-$400, no free shipping threshold mentioned,
  30-day returns, strong influencer/UGC presence on product pages

Build a comparison table across: price range, shipping policy, return
policy, and positioning/messaging angle. Then flag 2-3 gaps or
opportunities based on what neither competitor seems to be doing.

The value here isn’t Claude “knowing” your market — it’s Claude organizing what you already found into a format that makes the gaps obvious.

A repeatable competitor-tracking habit

Rather than a one-off deep dive, keep a running doc where you paste in new observations (a price change you noticed, a new promotion) every month or so, and periodically ask Claude to summarize what’s changed since your last review.

Here's my competitor notes doc from the last 3 months: [paste notes].
Summarize what's changed since the last entry, and note anything that
looks like a meaningful shift in their strategy versus a one-off
promotion.

Tip: Ask Claude to separate “things I should act on now” from “things worth watching” — not every competitor move or data blip deserves a response, and that distinction is easy to lose when you’re staring at a long list.

Key principle: Claude is a fast, tireless second pair of eyes on data and research you’ve already gathered — not a live source of truth about your sales or your market. The quality of the input always caps the quality of the analysis.

Try it: Export one month of sales-by-product data from your store, aggregate it to weekly totals, and paste it in asking what patterns stand out. Compare Claude’s read against what you already suspected was true.