Skip to content
case study · brand x · fashion & apparel

what five years of fashion data actually tells you

we built a synthetic dataset for a £1.5m apparel brand and ran our full analysis against it. here is what the data surfaces - and what each tier of the fte engagement gives you.

freedom to exist · data analysis · 8 min read

we get asked a lot: what does an fte analysis actually look like? what will you find in my data?

the honest answer is: it depends on your data. but the patterns we look for are consistent. to show you what we mean, we built brand x - a fictional fashion and apparel brand with five years of synthetic order data, modelled on the buying behaviours we see repeatedly across shopify stores in the £500k to £3m revenue range.

brand x is not a real client. the customer names, email addresses, and order history are all generated. but the patterns are real - we designed them to reflect what we actually find when we run the fte analysis on a genuine store.

the brand x dataset

brand x sells fifteen apparel and accessories products - tees, knitwear, outerwear, dresses, trousers. classic, considered pieces. the kind of range where a customer who buys one thing is a good candidate for buying two.

brand x · dataset overview
13,000
customers
30,018
orders (5yr)
£144
avg order value
2020£326,0352,283 orders
2021£581,6683,927 orders
2022£860,2505,882 orders
2023£1,069,8757,509 orders
2024£1,500,00010,417 orders

healthy growth. four and a half times revenue over five years. if you are this brand’s founder, you are probably pleased with the trajectory. what you might not be looking at - because the headline number is moving in the right direction - is what is happening inside the data.

what the data reveals

four patterns show up immediately when you run the fte analysis. none of them are visible in a revenue dashboard. all of them represent actionable opportunity.

+151%
average order value uplift when a customer adds a second item. single-item orders average £95. multi-item orders average £238.

the multi-item gap is the most consistent finding we see across shopify stores. customers who buy two items are not twice as valuable - they are two and a half times as valuable. and the majority of orders are single-item. at brand x, 65% of orders contain only one product.

the second pattern is the lapsed customer base. 560 brand x customers - 4.3% of the total - have not purchased in over 18 months. their combined historical spend is £332,922. they already know the brand. reactivating them costs a fraction of acquiring someone new.

third: 386 customers have only ever purchased on a discount code across two or more orders. the margin cost of that pattern compounds quietly.

fourth: the top 20% of customers account for 54% of five-year revenue. the bottom 50% account for 16%. knowing who your top customers are is the foundation of every retention decision worth making.

the three report tiers

fte produces analysis at three levels of depth. the reports below are built on the brand x dataset. they show you what you receive at each tier - the actual findings, the actual framing, and the actual recommendations.

view the sample reports

bronze

sales data analysis

revenue by year, category and product. seasonal patterns. aov and order frequency.

  • 5-year revenue breakdown
  • top products and categories
  • seasonal revenue patterns
  • aov and order volume trends
view sample report →
silver

sales + segmentation

everything in bronze, plus customer segmentation, multi-item uplift, and lapsed customer opportunity.

  • everything in bronze
  • customer segment breakdown
  • multi-item order uplift analysis
  • lapsed customer opportunity
  • discount dependency findings
view sample report →
gold

full analysis + recommendations

everything in silver, plus category cross-sell gaps, cohort retention, discount roi, and a prioritised action plan.

  • everything in silver
  • category cross-sell analysis
  • cohort retention by acquisition year
  • discount code margin impact
  • prioritised recommendation stack
view sample report →
a note on this dataset. brand x is a synthetic dataset. all customer names, email addresses, and transaction records are generated and do not represent real people or a real business. the patterns are modelled on findings from real fte engagements. the numbers are illustrative, not a guarantee of any specific outcome for your store.