For retailers, D2C brands and marketplace sellers.
We build campaign templates that keep every variant on brand, and product data that search engines and AI assistants can read. Our checkouts show shoppers the full price before they pay.
DPDP Act 2023 — Digital Personal Data Protection Act, 2023GDPR — General Data Protection Regulation (EU) 2016/679PCI DSS v4.0.1 — Payment Card Industry Data Security StandardCore Web Vitals — Loading, interactivity and visual stabilityWCAG 2.2 AA — Web Content Accessibility Guidelines
Retail & commerceThe store is now one screen among many.
SecuritySecurity & incidents: Often shapes the work
AccessAccessibility: Often shapes the work
AIAI governance: Sometimes shapes the work
What is shifting
The shelf became a thumbnail, and the checkout became regulated.
Discovery moved into AI answers and feeds, and the rules followed shoppers to the cart. Each shift names what it demands, with a fact you can check.
Quick commerce compresses the decision to minutes.
01
Discovery moved from search results to AI answers, marketplaces and social feeds.
What it requiresProduct data and content structured for search engines and AI assistants.
EvidenceAI assistants now answer product questions from structured data, reviews and policy pages.
02
Quick commerce reduces the decision to a thumbnail and a price.
What it requiresCampaign templates that produce thousands of variants, each one on brand.
EvidenceOnline listings must still show the maker, net quantity, MRP and country of origin.
03
Dark-pattern and consent rules now reach the checkout and the cookie banner.
What it requiresCheckouts that convert without drip pricing, false urgency or forced consent.
EvidenceIn November 2023, the CCPA named thirteen dark patterns as unfair trade practices.
04
Imported products will need a country-of-origin filter on every e-commerce site.
What it requiresProduct data models that hold every declaration before a listing can go live.
EvidenceUnder the April 2026 Legal Metrology amendment, the filter applies from 1 July 2027.
The rulebook
Honest prices, complete listings, and consent every tool respects.
Consumer-protection rules, the dark-pattern guidelines, packaging declarations, card security and the data law, read for a typical storefront. Your counsel keeps sign-off; our templates carry the required fields by design.
Frameworks we build to
DPDP Act 2023 — Digital Personal Data Protection Act, 2023
GDPR — General Data Protection Regulation (EU) 2016/679
PCI DSS v4.0.1 — Payment Card Industry Data Security Standard
Core Web Vitals — Loading, interactivity and visual stability
WCAG 2.2 AA — Web Content Accessibility Guidelines
We design and build to these frameworks; this is not a claim of certification.
Advertising & claims
Often shapes the work
Central Consumer Protection Authority
CCPA dark-patterns guidelines, 2023
Names thirteen dark patterns, including false urgency, basket sneaking, drip pricing and confirm-shaming, as unfair trade practices.
We designEvery checkout is audited against all thirteen patterns before launch.
E-commerce listings must show the pack’s mandatory declarations; from 1 July 2027, imported products must also be searchable and sortable by country of origin.
We designProduct data models hold every declaration, so a listing cannot publish without them.
LMPC Rules · Rule 6(10); Rule 6(10A) as substituted by G.S.R. 312(E), 27 April 2026
Payments
Shapes most of the work
PCI Security Standards Council
PCI DSS v4.0.1
Card-data security, including the v4 rules on payment-page scripts (6.4.3) and tamper detection (11.6.1), in force since 31 March 2025.
We designCheckout pages go live with a script inventory and change detection.
Compiled 2 October 2026. Our reading of typical programmes, not legal advice. Rules change; your counsel confirms how each one applies to you.
Challenges
A price shoppers trust, at the speed of the season.
What shoppers expect from the listing to the doorstep, and what it takes your teams to release thousands of variants without a gap.
AYour customers
A1
A price they can trust
Totals shown from the first screen, with no fee added at the last step.
A2
Found where they search
Products that answer questions in search, on marketplaces and in AI assistants.
A3
Accurate stock and delivery
Stock, delivery slot and price stay the same from the listing to the doorstep.
Every listing is only as good as the product data behind it.
BYour operation
B1
Thousands of variants
Every season multiplies formats, sizes, markets and channels.
B2
Complete product data
Declarations, attributes and imagery must be complete before a listing goes live.
B3
Consent across every tool
Analytics, ads and CRM must all honour the choice the shopper made.
What we build
A production line and a storefront that launch a season in days.
Production and platform carry most of a retail programme, with product, AI and CRM close behind. Every line links to the capability that does the work.
Audit the checkout first, then build the production line.
The way in tests every step from listing to payment against the dark patterns, the E-Commerce Rules and the card-security controls, then ranks the fixes. Lengths are typical, never promised.
The way in
Checkout and consent audit
We test every step from listing to payment against the dark-pattern guidelines, the E-Commerce Rules and PCI DSS script controls, then rank the fixes.
TaskGenerates channel, size and market variants of an approved campaign.
CheckLogo, colour, type and price checks run before any variant can be exported.
ApproverA designer approves each batch.
LogEach variant stores its source, prompt and approval.
✓Goes live only with its evaluation set, automatic checks and a named owner
Use case 02
Product data enrichment
Measured byListings complete at launch
Tested forOWASP LLM09 · Misinformation
agent / product-data-enrichmentGuarded
TaskFills missing attributes and declarations from supplier sheets.
CheckNever invents a declaration: a missing MRP or country of origin blocks the listing.
ApproverThe category manager approves new attributes.
LogField-level history of every change.
✓Goes live only with its evaluation set, automatic checks and a named owner
Use case 03
Shopping assistant
Measured byAssisted conversion, with the return rate held flat
Tested forOWASP LLM01 · Prompt injection
agent / shopping-assistantGuarded
TaskAnswers product, size and delivery questions from the catalogue and policy pages.
CheckQuotes prices only from the live catalogue and makes no urgency claim it cannot verify.
ApproverOrders, refunds and complaints go to customer care.
LogAnswers stored with the source of every price and policy.
✓Goes live only with its evaluation set, automatic checks and a named owner
A deliverable, in working code
Audit a checkout against the patterns that apply.
Mark what you find in your own checkout and the audit recounts what must be fixed before launch. The patterns shown are the ones a typical storefront can fall into.
Checkout audit · dark patternsIllustrative
Checkout v3, tested against the CCPA’s thirteen patterns
5 passed · , 2 fixed before launch
PassDrip pricingDelivery and platform fees shown on the product page
PassFalse urgency“Only 2 left” is tied to live stock
PassBasket sneakingNothing joins the cart without a tap
FixedConfirm-shamingOpt-out copy rewritten in neutral words
FixedSubscription trapCancel takes as many taps as sign-up
PassInterface interferenceAccept and decline carry equal weight
PassForced actionGuest checkout without an account
✓Seven of the thirteen patterns apply to this checkout.
What it shows
An illustrative checkout audit: of the seven dark patterns that apply, five pass and two are marked to fix before launch. Mark what your own checkout does and the audit recounts what is left to fix.
Yes. The checkout audit tests every step against the thirteen patterns in the CCPA’s 2023 guidelines and the E-Commerce Rules, and ranks the fixes by risk and by conversion.
Q.02Do you work on marketplaces as well as our own site?
Yes. Product data, content and retail media are built to work on your storefront, on marketplaces and in AI answers.
Q.03How fast can we launch a season?
The aim is days rather than weeks. That depends on having templates, product data and approvals in place, which we set up with you.
Q.04Will generative AI replace our design team?
No. It produces variants of approved work, with brand checks; your designers direct, review and approve every batch.
Keep exploring
Others in consumer & retail, and the rest of the map.