Skip to content

Case studies

The real problems apparel operators asked us to solve.

Four engagements. Four AI disciplines. Every one triggered by a specific question a founder, COO or CFO had already asked themselves in a Monday meeting. TwoDots ships in weeks, not quarters, and lands in the tools the team already uses.

Every case study on this page is an illustrative composite of TwoDots and HappySellers engagements. Specific client names are anonymised at operator request. Figures are drawn from real work.

Case studies

Demand forecasting Apparel · $6M · Shopify Plus · 900 SKUs

Stockouts 11% to 2.4%

Hero SKUs died in the first 72 hours of every drop. COO rebuilt the reorder spreadsheet every Sunday night.

"Every Sunday night my COO rebuilds the reorder sheet from scratch. Why can't the system just tell us what to buy on Monday?"
Founder, women's contemporary apparel

~$180K

recovered monthly

Timeline
6 weeks
Cost
$68K
Read the case study
Returns prediction Women's fashion · $12M · Shopify Plus

Returns 41% to 30%

Returns at 41% eating 14% of gross margin. Size charts and 3D try-on had moved the number by less than 2 points.

"We already know a specific customer buying a specific dress will return it. So why do we ship it exactly the same way?"
CFO, women's fashion D2C brand

$420K

annualised savings

Timeline
14 weeks
Cost
$82K
Read the case study
Content at scale Contemporary apparel · $4M · Shopify

1,400 pages in 11 weeks

1,800 thin SKU pages. Agency quoted $200K for 6 months to rewrite. Founder priced that against Q1 inventory and said no.

"An agency wanted six months and $200K to rewrite our catalog. That's more than we spent on inventory last quarter."
Founder, contemporary apparel D2C brand

$22K

total cost (vs $200K quote)

Timeline
11 weeks
Cost
$22K
Read the case study
Markdown optimization Contemporary apparel · $9M · Shopify + wholesale

Full-price sell-through +11 pts

Markdowns called by feel in a Monday meeting. Fast movers cut too early. Slow movers sat past the season. Wholesale conflicts every quarter.

"Every markdown feels like a coin flip. When do I cut, how deep, and how long before I cut again? Nobody in this building actually knows."
COO, contemporary apparel brand

+$340K

per season, recovered margin

Timeline
8 weeks
Cost
$58K
Read the case study

What repeats

The pattern across every engagement.

Four different AI disciplines, one recurring shape.

The buyer is the operator, not the CIO

Every engagement in this library was scoped, approved and lived through by the founder, COO or CFO. Not a data leader. That is the shape of the market we serve.

A Fit Sprint before any model code

Four weeks of discovery. Data audit. Intervention design. If the math does not work in the Fit Sprint, we say so and stop. Roughly one in five Fit Sprints ends there. That is the point.

The model is 30% of the work

The other 70% is intervention design, integration into the tools the team already uses, and the change-management around a Monday morning meeting. Any consultancy can build a classifier.

One screen per week, not one dashboard

Every case study lands in the same place: one screen the operator opens once a week, with the recommendations, the reasoning, and an override button. Dashboards get ignored. Weekly screens get used.

Common questions

Frequently asked

Are these real client engagements?

Every case study on this page is labelled illustrative until the specific client involved signs consent to publish their name. The figures, the intervention shapes, and the operator quotes are drawn from real TwoDots and HappySellers engagements. The naming and identifying detail is anonymised. As real clients complete engagements and sign consent, we replace the illustrative composite with the named version.

Why only apparel?

Because that is where the volume of our current work sits, and because apparel operators recognise apparel case studies more easily than composite ones. TwoDots ships in D2C beauty, home goods, industrial B2B and marketplace sellers too. Ask on a call if you want a walkthrough of a comparable engagement in your category.

How were the case studies picked?

They cover the four AI disciplines an apparel D2C brand doing $3M to $20M is most likely to need: demand forecasting, returns prediction, GenAI content, and markdown optimization. Together they map to roughly 80% of the inbound requests we take in a given quarter.

What does a typical engagement cost?

The four engagements on this page range from $22K to $82K, including the AI Fit Sprint. That covers a Fit Sprint at $10K plus implementation in the $12K to $72K range depending on scope. For an active retainer after implementation, the range is $5K to $15K per month. We scope tightly and quote before starting.

What if my brand does not look like these?

Book a call. We turn away brands regularly when the data is not there or the scope is wrong for our engagement model. That conversation is worth having whether you become a client or not. Our first call has no pitch deck.

Recognise the shape?

Book a 30-minute call.

We will tell you honestly which of these four engagements would move a real number in your business right now, and what the shape of a Fit Sprint would look like for you.

The Retail AI Implementation Weekly

Practical AI implementation for e-commerce operators. No hype.