Pricing infrastructure for commerce.
Find the price changes worth making — before customers see them.
A 1% price increase can move operating profit ~9%1 — but commerce pricing is still run reactively.
Bestsellers remain untouched
Margin leaks because teams are afraid to touch winners.
Discounting destroys avoidable margin
Brands cut price to move stock, even when it destroys gross profit.
Decisions are delayed or made by instinct
Teams suspect a move is right, but lack the evidence to ship it.
Analytics
What happened?
Repricers
What does the market charge, and what rule should run?
Intelligems
Which selected price wins on live traffic?
Parbat
Which moves deserve action, what are they likely worth, and do they clear our constraints before exposure?
Every morning, rank the price moves most likely to add incremental gross profit, from the brand’s own traffic, conversion, margin, inventory and price history.
Test each candidate on brand-calibrated simulated shoppers at every price, and check it against the brand’s guardrails, before any customer sees it.
Only the brand ships. One click puts the price live and one click rolls it back. Every shipped move carries its Evidence Pack.
Measure each shipped move’s realized gross profit against its projection, misses included.
Feed measured outcomes back into finding and validation, so each brand’s projections become more accurate over time.
Jul 31, 2026
Shopify public app launched
Intentionally unlisted until beta concludes
1
Design partner in production
Operating live end-to-end
11
Live price changes shipped
Across 2 products with our design partner; a cohort of 20+ moves is next
8
Beta brands planned
Ready to onboard
Shopify provides a standardized data and execution surface
Modern demand modeling makes catalog-wide screening feasible
Governance, calibration, and accountability are now the bottlenecks
The missing layer is not another recommendation engine. It is a controlled system that validates and governs changes before exposure.
Initial ICP
~10,000 brands
Matching U.S. Shopify DTC brands*
× $6K–$12K ACV
Annual Contract Value
= $60–$120M ARR
Serviceable addressable market
Expansion paths
Global Shopify pricing
Multi-store and enterprise governance
Broader measurable commerce decisions
* Filtered for $2M–$20M Shopify revenue, order density (15+ products at 20+ monthly orders, proxied externally by store traffic), margin, and decision authority. Manually verified fit rate on a 100-store sample.
Initial pricing
Standard ACV
$12K / year — annual subscription
Beta conversion
~$6K in year one (50% pre-agreed beta discount)
Expansion
Additional stores, catalog complexity, review workflows, governance, and integrations
Customer break-even
$1,000
per month incremental gross profit to break even at $12K ACV
Our go-to-market strategy focuses on a disciplined, three-stage approach to acquire and grow customer relationships.
Founder-led outreach to design-partner referrals, heads of eCommerce, and brands in the beachhead segment.
Generate measurable pricing wins and turn them into strong case studies with quantified impact.
Grow through a robust Shopify-native presence, partner agencies, and customer proof points.
Every shipped move creates a proprietary intervention record:
Decision context
Candidate price
Projected outcome
Uncertainty
Policy state
Brand decision
Realized outcome
What improves
Parbat tracks forecast bias, error, and interval coverage by brand. As the system observes more interventions, brand-specific calibration should improve and uncertainty should narrow only where the evidence supports it.
Why switching has a cost
A new provider can access historical orders. It does not inherit Parbat's exact projection, policy, ship, rollback, and outcome lineage—or the calibrated model state produced by those interventions.
Parbat starts with Shopify pricing because pricing is frequent, high-leverage, measurable, and under-operationalized.
But the core system is not limited to price changes.
Predict the outcome of proposed commercial actions before they go live.
Ensure changes align with business guardrails and policies.
Launch only what is economically attractive and validated.
Continuously refine models based on realized results.
That same loop can extend from pricing into promotions, markdowns, merchandising, inventory, and other profit-sensitive commerce decisions.
Parbat did not start as a generic AI pricing idea. We spent the last three years embedded with a live design partner, observing how pricing decisions were actually made in practice: too much manual judgment, too little evidence, and no safe path to ship.
That led to the belief that brands do not need more pricing analytics. They need a system that helps them find, validate, and ship pricing decisions with confidence.
Founder and CTO (full-time)
Built production platforms at Cisco and Salesforce. Changing prices on a live store is a systems-trust problem before it’s a modeling problem — rollback, feed sync, failure modes at load — and that’s the work he's done for years.
Co-Founder (advisory → full-time post-raise)
Comes from Dynata and Ipsos, where measuring how people actually respond to things — rather than what they say they’ll do — was the entire job.
Milestones
connect@parbat.ai · www.parbat.ai/contact
Diligence appendix available on request: simulation architecture, measurement methodology, market-sizing method, beta design, security and governance, fundraising, intellectual property.