How We Test & Analyze Every Deal
At Zaryaft, transparency is not optional — it's foundational. This page explains exactly how we evaluate, score, and recommend shopping deals across Pakistan's retail landscape.
Our Commitment to Transparency
The Pakistani online retail market has grown exponentially, but with that growth comes a proliferation of misleading discounts, inflated original prices, and manufactured urgency. Consumers deserve tools that cut through the noise and provide honest, data-driven guidance.
Zaryaft was built to be that tool. Unlike promotional platforms that earn revenue by driving volume regardless of value, our business model depends on trust. Every recommendation we publish is backed by verifiable data, and this methodology page exists so you can hold us accountable. We believe that an informed consumer is a confident consumer, and confidence starts with understanding how the information you receive is generated.
The Deal Score Algorithm
Every product listed on Zaryaft is evaluated by our proprietary Deal Score algorithm — a composite rating from 0 to 10 that represents the overall value proposition of a deal. Here's how it works:
1. Discount Depth (Weight: 30%)
How deep is the markdown relative to the product's historical retail price? A 60% discount scores significantly higher than a 15% promotional reduction. We measure this against both the stated original price and our independently tracked price history to ensure accuracy.
2. Brand Reputation Tier (Weight: 20%)
Brands are classified into tiers — luxury, premium, mid-range, and value — based on market positioning, fabric quality, and consumer sentiment. A 30% discount from a luxury brand like Sana Safinaz carries more weight than the same percentage from a value-tier label, as premium brands discount less frequently.
3. Category Average Comparison (Weight: 25%)
We benchmark every deal against the average discount in its product category. If lawn suits are averaging 25% off platform-wide, a 45% discount on a comparable item is flagged as above-average value. This contextual comparison prevents consumers from being impressed by high percentages that are actually standard for the category.
4. Price History Volatility (Weight: 15%)
Our system tracks price fluctuations over time. Products with stable pricing that suddenly receive a deep discount score higher than items whose prices fluctuate constantly. This factor helps identify genuine sales events versus brands that cycle through artificial "sales" to create false urgency.
5. Availability & Timing Signals (Weight: 10%)
Seasonal relevance, stock patterns, and the timing of the discount relative to product lifecycle all contribute to the final score. An end-of-season clearance on winter wear in February scores differently than the same product discounted at the start of winter.
Score Interpretation Guide
8-10
Exceptional
6-7.9
Strong Value
4-5.9
Fair Value
0-3.9
Below Average
Price History Tracking
One of the most common tactics in online retail is inflating the "original price" before announcing a sale. A product listed at Rs 5,000 that was "previously Rs 10,000" sounds like a 50% discount — but if the product was never actually sold at Rs 10,000, the discount is misleading.
Zaryaft combats this by maintaining independent price records for products across our tracked brands. Our system captures price snapshots at regular intervals, building a timeline that reveals genuine price movements versus artificial inflation. When we display a price history chart on a deal page, it reflects our independently collected data — not the brand's self-reported pricing.
Currently, we track prices across 60+ Pakistani brands with data collection cycles running every four hours. This frequency ensures that flash sales, limited-time offers, and sudden price changes are captured with minimal delay. Products with insufficient historical data are flagged accordingly, so consumers understand when our analysis is based on limited information.
Editorial Review Process
Beyond algorithmic scoring, every deal page on Zaryaft features an editorial analysis section. These reviews are generated using a combination of quantitative data and qualitative frameworks developed by our editorial team. Each review covers five key dimensions:
- →Market Context: Current trends in the product's category and seasonal relevance.
- →Brand Assessment: The brand's quality reputation, target audience, and typical pricing behavior.
- →Price Intelligence: How the current price compares to historical levels and category averages.
- →Shopping Tips: Practical advice on sizing, returns, and optimal purchase timing.
- →The Verdict: A clear Buy, Wait, or Skip recommendation with supporting rationale.
Our editorial framework ensures consistency across all 34,000+ deal pages while maintaining product-specific uniqueness. Each review is tailored to the individual deal's data profile, meaning no two reviews are identical — even for products from the same brand or category.
Data Freshness & Accuracy
Accuracy is the foundation of trust. Our data pipeline operates on a four-hour refresh cycle, meaning deal information on Zaryaft is never more than four hours old during active monitoring periods. When a deal expires, is modified, or a product goes out of stock on the brand's official website, our system reflects these changes in the subsequent update cycle.
All data is sourced exclusively from official brand websites — we do not rely on third-party marketplaces, user submissions, or unverified sources. Every price, discount percentage, and product image displayed on Zaryaft can be traced back to its original source on the brand's official online store.
Our Independence Policy
Zaryaft operates as an independent shopping intelligence platform. Our commitment to neutrality is non-negotiable and is governed by the following principles:
- ✓No Paid Placements: Brands cannot pay to improve their Deal Score or receive favorable editorial coverage. Our algorithm treats all brands equally.
- ✓Affiliate Transparency: Some product links may generate affiliate commissions. This revenue helps sustain our platform but has zero influence on our scoring or editorial recommendations.
- ✓Data-Driven Decisions: Every recommendation is generated from quantitative data. Our editorial team follows strict guidelines that prioritize consumer value over commercial relationships.
Questions About Our Methodology?
We welcome scrutiny and believe transparency strengthens trust. If you have questions about how a specific deal was scored, or want to understand our methodology in more detail, our editorial team is available for direct communication.