COMMERCIAL PROPOSAL — VERSION 7.1

Release 1 — Commerce Foundation & Premium Storefront

A commercially complete Release 1: a premium customer storefront built on structured product data, Product Attribute Management and a launch-ready migration of MetroMart's existing attribute catalogue. The enterprise backend remains a separate later engagement.

Commerce FoundationAttribute ManagementCatalog MigrationBoG + TBC Payments1C Stock & Price ImportAnalytics Ready
Release 1 — Build Your Package
One-time Project:$20,000

Base is the $20k committed MVP storefront. Tick any optional module to build your Release 1 package — each is priced and contracted separately.

Optional Modules

Project ScopeRelease 1 Storefront
MVP Timeline~12–16 weeks
Payment StructureDiscovery + 40/30/25/5
Support Retainer40 hrs/mo, optional

MVP Payment Breakdown

Discovery is a separate, paid first step. The $16k build (40/30/25/5) is committed only after Discovery delivers the locked SOW. Release 1 total stays $20k.

0. Discovery & Architecture (separate, on signing):$4,000
1. Build kickoff (40% of build):$6,400
2. Design & spec signoff (30%):$4,800
3. Staging delivery (25%):$4,000
4. Storefront launch (5%):$800
Start with Discovery — $4,000
1. Executive Summary

Commerce Foundation MVP

The MetroMart Blueprint describes an enterprise-class e-commerce ecosystem. Our recommendation remains a phased launch: begin with a commercially complete Release 1, learn from real customer, payment and operational data, then expand the backend in controlled stages.

Version 7.1 clarifies the complete Product Attribute commitment. Product Attributes and their management are part of the Release 1 platform foundation, and the existing attribute/value dataset will be migrated and normalised to the agreed launch-ready standard. An optional AI Catalog Intelligence Pack adds a deeper enrichment and ongoing quality layer.

Release 1 remains a single USD 20,000 commitment — Discovery USD 4,000 plus Build USD 16,000. Product Attribute Management, migration and launch-readiness normalisation are included within this price. Optional growth modules and support remain separately contracted.

Go-LiveSTAGE · 01

Commerce Foundation MVP

Commercially usable first version: premium storefront, Product Attribute Management, launch-ready catalog migration, checkout, BoG + TBC payments, and a one-way 1C stock/price import.

OptionalSTAGE · 02

Growth & Conversion Modules

Three à-la-carte add-ons, each priced separately: Critical Commerce Pack (+$12k), Stabilization Sprint (+$9k), and the optional AI Catalog Intelligence Pack (+$7.5k).

OperationalSTAGE · 03

Operational Support

Optional post-launch retainer ($2,500/mo): monitoring, incident resolution, payment-callback analysis, and up to 40 hours per month.

2. Scope Matrix

Deliverables by stage

The USD 20,000 Release 1 includes Product Attribute Management plus migration and launch-readiness normalisation of MetroMart's existing attribute/value dataset. Optional modules remain separately contracted.

Commerce Foundation MVP Scope

Premium Storefront: Responsive mobile-first Homepage, Category, Brands, PLP and PDP experiences.
Product Attribute Management: Reusable attribute definitions, groups, data types, allowed values, category templates/mapping, product-level values, display ordering and core required-field/type validation.
Attribute-Driven Discovery: Category, brand, price and attribute filters; product specifications on PDPs; an attribute structure prepared for comparison and richer search.
Catalog Administration: Laravel/Filament admin workspace for categories, brands, products, product media, attribute groups, attribute values and basic order management.
Catalog Migration & Launch Readiness: Migration of MetroMart's existing attribute/value dataset using agreed mappings, including deduplication, format and unit normalisation, category assignment, import testing and launch-readiness corrections.
Checkout & Ordering: Shopping cart, linear checkout form, and automated customer email confirmations.
1C Stock & Price Import: One-way scheduled import from a client-provided 1C export in an agreed format. Product master-data and attribute feeds are included only if available in an agreed structured export. Write-back, real-time bidirectional sync and rollback fail-safe remain outside Release 1.
Payments: Bank of Georgia AND TBC standard payment/installment callbacks. Both are included; each gateway's timeline depends on the client providing working sandbox credentials and merchant approval.
Setup & Launch: GA4/GTM analytics, Meta Pixel tags, Cloudflare setup, production deployment, and 30-day bug warranty.
Bilingual storefront: EN/KA interface framework with Georgian UI. Content translation and general copywriting are provided by the client.
Import Templates: Agreed Excel/CSV templates for product and attribute data imports.

Product Attribute Management & Migration — included foundation

The detailed taxonomy, mapping rules and acceptance criteria will be finalised during Discovery. Release 1 includes migration of MetroMart's existing attribute/value dataset, technical normalisation and the corrections required for working specifications, filters and daily catalog management. MetroMart will provide the source dataset and business decisions for ambiguous or conflicting values.

Included in Release 1

  • Existing attribute/value dataset migration
  • Mapping, deduplication and unit/format normalisation
  • Category assignment, core validation and launch corrections
  • PDP specifications, filters and admin management

Defined during Discovery

  • Final taxonomy and category templates
  • Source-of-truth and ambiguity rules
  • Sample testing and measurable acceptance criteria
  • AI Pack priorities, sources and review workflow if selected

Separate / at cost

  • Manual reconstruction of missing specifications
  • External research and unbounded item-by-item verification
  • Complex custom rule engines beyond the agreed scope
  • Ongoing catalog data-entry operations
3. Delivery Phases

A structured execution timeline

We divide execution into structured engineering phases, plus support and the optional AI Catalog Intelligence Pack. Click on any phase to review objectives and activities.

Objectives & Key Deliverables:
Objective: De-risk the build before any fixed-price commitment. Paid separately; the MVP build is committed only after this phase.
Catalog and attribute audit: Review the Blueprint, current category structure, the complete available attribute/value dataset, representative products, data quality and operational ownership.
Attribute model & migration: Define groups, types, category templates, required/optional rules, display/filter/search flags, mappings, normalisation rules, ambiguity handling and launch acceptance criteria.
AI Pack scope, if selected: Agree priority categories, permitted source materials, enrichment targets, confidence thresholds, administrator review workflow and measurable acceptance criteria.
1C/ERP mapping: Confirm stock/price feed feasibility and whether product master-data or attribute data can be consumed from an agreed structured export.
Confirm Bank of Georgia & TBC sandbox availability and merchant-approval status.
Locked SOW: Produce the architecture, migration acceptance criteria and confirmed Build scope. Materially different integration or data conditions may re-baseline scope, price or timeline.
4. Technology Stack

Modern, fast, and maintainable

We design bespoke architectures to optimize user load capacity, SEO ranking, and data synchronization.

Frontend Layer

Next.js (React)

Server-rendered, responsive storefront architecture with accessible components and search-engine-readable product and category pages.

Backend Layer

Laravel or NestJS

Highly robust frameworks featuring enterprise queue systems. Safely schedules catalog synchronizations and parses webhooks in the background.

Product Data Layer

PostgreSQL & Redis

A relational model for categories, products, attribute definitions, allowed values, category templates and product attribute values. It supports specifications, filters, indexing and future comparison without duplicating product logic across the storefront.

Search & Filters

Meilisearch / Typesense

Attribute facets, typo tolerance and fast catalog discovery.

API & Admin

Custom / Filament Admin

Role-aware management of categories, products, media, attribute groups, attribute definitions, values, import mappings and orders.

Static Media & CDN

S3 & Cloudflare

Cloudflare manages edge-caching, dynamic file optimizations, and active DDoS mitigations, shielding the production cluster.

5. Pricing Details

Budget breakdown and schedules

MVP discovery, design, and engineering milestones are committed as fixed fees in USD. Custom modules are invoiced according to the deliverables checklist.

Discovery & Architecture (separate, pre-build)~2 weeks. Delivers the locked SOW; the build is committed only after.
$4,000100% on signing
Commerce Foundation MVP Build (committed after Discovery)Release 1 total, including Product Attribute Management, migration and launch-readiness normalisation: $20,000.
$16,000Milestone payments (40/30/25/5)
Critical Commerce Pack (SEO, Merchandising, Analytics)Quoted at $15,000 if contracted separately post-launch.
$12,00050% start / 50% integration
Stabilization & Enhancement Sprint6–8 weeks duration. Planned optimization pass on real transaction data.
$9,000Post-sprint completion
AI Catalog Intelligence PackAI-assisted enrichment and quality control: extract missing specifications from agreed sources, detect duplicates/anomalies, normalise terminology and units, suggest mappings, score confidence and route changes through admin review; also includes descriptions, metadata and image alt text.
$7,500Start of pack
Post-Launch Operational Support RetainerUp to 40 hours/month. Recommended initial term: 6 months. Renewable after review.
$2,500/moMonthly in advance
Optional enrichment

AI Catalog Intelligence Pack — optional enrichment layer

Release 1 makes the existing catalogue operational and launch-ready. If the AI Catalog Intelligence Pack is selected, the catalogue is taken to an agreed enriched and quality-controlled standard through AI-assisted extraction, anomaly detection, terminology and unit normalisation, confidence scoring and administrator review. Priority categories, permitted source materials and acceptance criteria are locked during Discovery; no AI-generated or corrected data is published without review.

Pricing & Tax Note: All commercial figures are stated in USD; the GEL equivalent, if required, is calculated at the exchange rate at invoice. The quoted price is the final amount payable — no VAT is added to these figures. They exclude only third-party SaaS/provider bills (hosting, Cloudflare subscriptions, domain licenses, transactional SMS packs, 1C developer retainers).
6. Shared Expectations

Keeping the build predictable

We operate under transparent alignment policies. Explore how we split integration uncertainty and what we need from you.

Capped Risk-Shared Operations

A small subset of database tasks carry extreme unknowns due to target system documentation gaps. To protect you from inflated fixed-price bids, these modules are confirmed during Discovery and handled against agreed caps or separate SOWs:

Complex 1C/ERP synchronization rulesConfiguring transactional write-backs, catalog variant tables, and failover sync scripts.
Bank access dependency (BoG & TBC)Both gateways are in the MVP, but each one's timeline depends on the client supplying working sandbox credentials and merchant approval. Client-side delays pause that gateway's timeline (not a breach); if access is unavailable by staging, it ships immediately post-launch without blocking go-live.
Extended installment & write-back logicInterest/installment edge-cases beyond standard callbacks, and any 1C write-back, are confirmed during Discovery and handled against agreed caps.
Automated refunds & external listings (future scope)Programmatic bank-return webhooks and Wolt/Onoff catalog feeds belong to the separate enterprise engagement, not this contract.

Contractual protections (both sides)

Clear boundaries keep the engagement predictable and protect both parties. These terms are written into the agreement.

Separate, paid DiscoveryDiscovery ($4,000) is its own contract and produces the locked SOW. The $16k build is committed only afterwards. If Discovery reveals materially different integration conditions (1C feed quality, bank API constraints), scope, price and timeline are re-baselined — or the client stops there, having paid only $4,000.
Limitation of liabilityTotal liability is capped at fees paid. No liability for indirect or consequential loss, for third-party outages (banks, 1C, Cloudflare), or for loss arising from client-supplied ERP data.
Cancellation & IPMilestone payments are non-refundable; work to date is billable on cancellation. Intellectual property transfers to the client on full payment.
Warranty scopeThe 30-day warranty covers defects in delivered, committed scope only — not new features, change requests, or breakage originating on the bank or 1C side.
Migration and data responsibilitiesRelease 1 includes migration and launch-readiness normalisation of MetroMart's existing attribute/value dataset. MetroMart provides the source data and timely business decisions for ambiguous, conflicting or commercially sensitive values.
Agreed migration boundaryDiscovery locks mappings, normalisation rules, sample tests and acceptance criteria. Manual reconstruction of missing specifications, external research and ongoing catalog operations are separately estimated unless explicitly included in the AI Catalog Intelligence Pack scope.
AI Catalog Intelligence PackIf selected, AI-assisted enrichment is limited to the agreed source materials, priority categories, review workflow and acceptance criteria. It cannot guarantee facts that are absent or unverifiable in the available sources.
Explicitly out of MVPLoad/stress testing, DR rollback fail-safe, full WAF/DDoS hardening & rate-limiting, and CI/E2E pipelines are not in Release 1. "Cloudflare setup" means baseline CDN/SSL routing only.
Phase 2 · Future Direction

Statement of readiness — the enterprise backend

Not part of this proposal

The MetroMart blueprint also describes a deep enterprise backend. We are fully equipped and ready to build it — but it is intentionally excluded from this contract. It is a separate engagement, scoped, priced and negotiated on its own once Release 1 is live and proven on real data. Nothing here implies a commitment or a figure today.

Automated bank refunds triggered directly from the admin panel.
Bidirectional, hardened 1C/ERP synchronization with validation and rollback fail-safe.
Warehouse stock-ledger tracking and RMA / returns cycles tied to 1C stock levels.
Advanced CRM flows, customer profiles, and automated abandoned-cart re-engagement.
Multi-bank installment logic beyond standard callbacks.
External listing feeds — Wolt, My.ge, Onoff.ge — and marketplace stock sharing.
End-to-end automated testing pipelines and high-capacity load/stress testing.
Full edge security: WAF, DDoS hardening, rate-limiting, and DR backup strategy.

When MetroMart is ready, this becomes a dedicated Statement of Work with its own discovery, scope and pricing. Treat this section purely as a declaration of capability — a signal that the Release 1 foundation is built to grow into it.

Kickoff Step

Begin with Discovery & Architecture

Begin Discovery & Architecture — USD 4,000. The phase produces the locked Statement of Work, confirms the Product Attribute model and migration acceptance criteria, defines the optional AI enrichment boundary if selected, validates 1C and bank dependencies, and de-risks the USD 16,000 build commitment.

Begin Phase 1Discovery Phase · $4,000 · ~2 weeks