Skip to content

AI Fit Sprint · 30-Day Assessment

The 30-Day AI Readiness Assessment for Ecommerce Businesses

We connect to your Shopify, Amazon, or GA4 data, identify 3–5 AI opportunities ranked by ROI and data feasibility, and deliver a production-ready roadmap in 30 days. Working documents with milestones and owners — not slides.

30 days, fixed One month. Start to finish. Then it's done.
3–5 ranked use cases Ranked by what your data supports and what's worth building — not gut feel
0 filler Every deliverable is a working document. No slide decks.
Built by Sunil Kumar, former Kohl's & Sears AI lead Fixed scope. Fixed timeline. Fixed fee. Roadmap is yours — take it anywhere Honest verdict, even if the answer is 'not yet'

Who It's For

Who needs an AI readiness assessment?

The sprint is for businesses doing $1M–$20M who already know AI could help, but haven't found a clear, low-risk way to start.

What is an AI readiness assessment?

An AI readiness assessment evaluates whether your data, tech stack, and business processes are in position to support a custom AI build — and identifies which problems are worth solving first. The AI Fit Sprint is TwoDots's fixed-scope version: 30 days, five specific deliverables, one honest verdict on where AI fits in your business.

  • You want AI but don't know where to start

    You've read enough to know AI could help. You just don't know which problem to solve first, or whether your data is good enough to start.

  • You've tried tools that didn't stick

    You've installed Shopify apps, run a few experiments. Nothing landed. You're not sure if AI is genuinely useful for your business or just overhyped.

  • You've talked to a consultant and got a deck

    You paid someone to assess your AI readiness. They sent a 40-slide presentation. Nothing shipped. You want someone who delivers something you can actually build from.

  • You have data but no one's reading it

    You have 360+ days of orders, returns, and sessions sitting in Shopify, Amazon, or GA4. You know there are patterns in there. You just need someone to surface them.

What AI typically solves

Problems we find in most ecommerce businesses

In our audits of ecommerce businesses doing $1M–$20M, the average business has at least three solvable AI problems sitting in their existing data. Most never find them because nobody is looking. The sprint figures out which ones are worth solving first — and whether your data is ready to support it.

Stockouts and overstock eating into margin

AI can: Demand forecasting — predict what to order and when, based on your actual sales patterns

Most ecommerce businesses we audit lose 4–8% of annual revenue to preventable inventory mismatches.

Product recommendations that don't convert

AI can: Personalisation model — surface items people actually buy together, not just what's trending

Returns eroding profit you can't see coming

AI can: Returns risk model — flag high-return orders before they ship so you can act early

In our audits, return rates above 15% are almost always predictable from existing order data — the signals are there; nobody is reading them.

Reconciliation taking days every month

AI can: Automated matching — close your books in minutes instead of chasing transactions manually

Static prices in a market that moves daily

AI can: Dynamic pricing engine — adjust in real time based on demand, competition, and stock levels

Customers churning with no warning

AI can: Churn prediction — identify at-risk customers early and trigger retention before they leave

How we evaluate readiness

The three things that determine whether AI is worth building

Not every business is ready for AI, and not every use case is worth building. Before we recommend anything, the sprint runs three checks.

01

Data volume and quality

We check whether you have enough structured transaction history to train a reliable model. Demand forecasting typically needs 12–18 months of SKU-level sales data. A returns risk model can work with 6. We tell you exactly what you have and what is missing.

02

Use case feasibility

We map every potential AI opportunity against your actual data — not a generic framework. An inventory model built on incomplete warehouse records produces false confidence in bad numbers. We flag those before any build begins.

03

ROI justification

Before recommending any build, we model the projected impact: revenue recovered, cost reduced, hours saved per week. If the numbers do not justify a custom build at your scale, we say so clearly — not after you have committed to a build.

The 30 Days

The 30-Day AI Readiness Assessment Process — Week by Week

Fixed scope. Fixed timeline. No drifting briefs or open-ended engagements.

1

Week 1

Data Audit

We connect to your platforms — Shopify, Amazon, GA4, your warehouse system — and assess what data you have, what's clean, and what's missing.

Data readiness report
2

Week 2

Use Case Mapping

We identify 3–5 AI opportunities in your business. Each one is assessed for data feasibility and potential ROI before we go any further.

Ranked use case list
3

Week 3

Opportunity Sizing

We model the projected impact of each use case — revenue recovered, cost reduced, hours saved. You see the numbers before any build begins.

ROI model per use case
4

Week 4

Roadmap Delivery

We hand you a prioritised production roadmap: specific milestones, owners, timelines, dependencies, and a build-vs-buy recommendation per use case.

Production roadmap (yours to keep)

What You Get

What you get at the end of the AI Readiness Assessment

At the end of 30 days you have a complete picture of where AI fits in your business — and a plan specific enough to hand to an engineering team.

  • Data readiness report — what's usable, what needs fixing, what's missing
  • 3–5 ranked AI use cases with projected ROI for each
  • Production roadmap with milestones, owners, and timelines
  • Build vs buy recommendation per use case
  • An honest verdict: is AI worth pursuing right now, or not

In practice

What a sprint looks like in practice

A Shopify operator doing $6M/year came to us after a failed pilot with a third-party recommendation tool. In 30 days, the sprint identified two viable use cases — demand forecasting and returns risk prediction — with a combined projected annual impact of $380K.

They went into implementation with a ranked build order, an ROI model per use case, and a clear data infrastructure gap identified before a line of code was written.

Anonymised. Apparel brand on Shopify, $1M–$20M revenue tier.

  • Business size $6M annual revenue
  • Use cases identified 2 viable — demand forecasting + returns prediction
  • Projected annual impact $380K combined
  • Time to roadmap 30 days
  • Starting point Failed third-party tool pilot

After the Sprint

Three honest outcomes

The sprint ends with a clear verdict — not a vague recommendation to "explore further."

Outcome 1

Strong AI opportunities found

Move into AI Implementation — we scope the first milestone and start building.

Learn more →

Outcome 2

Data infrastructure gaps

Fix the foundations first with The Plumbing, then build the AI on top.

Learn more →

Outcome 3

AI isn't right yet

We tell you that clearly, explain what needs to change, and don't bill for the sprint.

Common questions

Common questions

What is an AI readiness assessment?

An AI readiness assessment evaluates whether your data, tech stack, and business processes are in position to support a custom AI build — and identifies which problems are worth solving first. It produces a ranked list of AI opportunities, a data readiness report, and a production roadmap. The AI Fit Sprint is TwoDots' fixed-scope version: 30 days, five specific deliverables, one honest verdict.

What data do I need before the sprint starts?

At minimum: 12 months of order history and access to your primary sales platform (Shopify, Amazon Seller Central, WooCommerce, or similar). We work with what you have and flag gaps in Week 1.

How much of my time does this take?

Around 3–4 hours total. A kick-off call at the start, a mid-sprint check-in in Week 2, and a roadmap walkthrough at the end. We do the work; you review the output.

Can I take the roadmap to a different agency to build?

Yes. The roadmap is yours. We write it to be implementable by any competent AI or engineering team, not just ours.

What if you find AI isn't right for my business?

We say so clearly. If the sprint uncovers that your data isn't ready or that the ROI doesn't stack up, we won't push you into a build. We explain what needs to change and what to do next. You also don't pay for the sprint.

How is this different from a free discovery call?

A discovery call takes 30 minutes and ends with a follow-up email. The AI Fit Sprint takes 30 days and ends with five specific documents: a data readiness report, 3–5 ranked use cases with individual ROI models, a production roadmap with milestones and named owners, and a build-vs-buy recommendation per use case. One produces a conversation. The other produces something you can hand to an engineering team.

What platforms do you work with?

Shopify, WooCommerce, Amazon Seller Central, Magento, and custom ecommerce stacks. For analytics: GA4, Mixpanel, and direct database access. For accounting: QuickBooks, Xero, and NetSuite.

What size of business is this for?

The sprint is designed for ecommerce businesses doing $1M–$20M in annual revenue. Below that threshold, the data volume and ROI usually don't justify a custom AI build.

How much does an AI readiness assessment cost?

The AI Fit Sprint is a fixed-fee engagement — one price, 30 days, five deliverables. There are no retainers, no hourly billing, and no scope creep. If the sprint finds that AI isn't right for your business yet, you don't pay. Contact us for current pricing.

Ready to start?

Book a free 30-minute AI fit call

We'll tell you in the first call whether the sprint makes sense for your business. No pitch, no obligation.

Fixed-fee engagement. One price, 30 days, five deliverables. No retainers, no hourly billing.

The Retail AI Implementation Weekly

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