DHANCHAKSHU METHODOLOGY

Last Updated: [13 Sep. 2026]

DhanChakshu - A product by Just Adbhut, uses a rules-based, math-first approach to compare credit cards.

Our objective is simple:

Don't just ask which card gives the most rewards.

Ask which card provides the most meaningful value for the way a particular user actually spends.

1. OUR CURRENT CARD UNIVERSE

The current DhanChakshu V1 database contains exactly:

50 Verified Credit Cards

These cards are selected from major Indian card issuers and cover different reward structures, fee levels, spending categories and use cases.

The card universe may change over time as products are added, removed, replaced or materially changed by issuers.

2. DATA SOURCES

DhanChakshu primarily uses publicly available issuer information and product documentation, including where available:

• Official card pages
• Official terms and conditions
• MITC documents
• Fee schedules
• Reward-program information
• Official issuer communications
• Other relevant issuer documentation

Where appropriate, DhanChakshu records source, verification and freshness information internally.

3. DhanChakshu Score

The DhanChakshu Score is a product-level assessment of the overall quality and value characteristics of a credit card.

It is not personalized to one particular user.

The current generic scoring framework uses the following weighted components:

• Reward Competitiveness — 25%
• Fee / Value Efficiency — 20%
• Reward Coverage Breadth — 15%
• Benefits Quality — 15%
• Caps / Exclusions / Restrictions — 10%
• Reward Flexibility — 10%
• Operational Usability — 5%
• Total — 100%

4. DhanChakshu Match

The DhanChakshu Match is personalized to the user's spending pattern and preferences.

The current framework uses:

• Spending Fit — 45%
• Preference Fit — 20%
• Fee Fit — 15%
• Benefit Fit — 10%
• Usability Fit — 10%
• Total — 100%

5. SCORE VS MATCH

These are deliberately different.

A card can have a high DhanChakshu Score but still be a poor match for a particular user.

Likewise, a card with a lower generic Score can be a better match because its reward structure aligns more closely with the user's actual spending.

The recommendation therefore does not simply rank cards by their generic score.

6. REWARD CALCULATION

Where sufficient data is available, DhanChakshu converts rewards into an estimated rupee value.

For example:

Points Earned = Eligible Spend × Points Earn Rate

Estimated Reward Value = Points Earned × Estimated Rupee Value Per Point

Effective Reward Rate = Estimated Reward Value ÷ Eligible Spend × 100

7. REWARD VALUATION

Points are not automatically treated as equivalent to cash.

Different reward programs can have different redemption values.

DhanChakshu therefore attempts to use a documented valuation appropriate to the relevant reward program and redemption mechanism.

Cashback is generally valued at its actual cash value where ₹1 cashback equals ₹1.

Points, miles and other rewards may have different values depending on redemption.

8. WHY TRANSACTION ECONOMICS MATTER

A card transaction can earn rewards and still be financially unattractive.

For relevant transactions, DhanChakshu attempts to account for:

• Gross reward value
• Quantifiable benefits
• Surcharges
• GST on applicable surcharges
• Other mandatory transaction charges
• Applicable annual card fees

The core concept is:

Net Financial Value = Reward Value + Quantifiable Benefit Value − Transaction Charges − Applicable Taxes − Applicable Annual Fee

9. EXAMPLE OF A BAD REWARD TRANSACTION

Suppose:

• Reward earned = 1%
• Transaction surcharge = 1%
• GST on surcharge = 18%

The effective cost is:
1% + 0.18% = 1.18%

Therefore:
1.00% reward − 1.18% cost = −0.18%

In this situation, the transaction earns a reward but is still financially negative.

This is one of the important principles behind DhanChakshu.

10. SPENDING CATEGORIES

DhanChakshu uses a structured category system rather than requiring users to understand technical merchant-category terminology.

Users can enter spending using familiar categories such as:

• Shopping
• Dining
• Groceries
• Travel
• Utilities
• Insurance
• Education
• Rent
• Fuel
• Other relevant spending categories

The engine handles card-specific reward rules, exclusions, caps and payment-channel logic in the background.

11. MERCHANT-SPECIFIC SPENDING

Where a category contains merchant-specific information, the engine avoids counting the same spending twice.

For example:

Online Shopping = ₹10,000

If:
Flipkart = ₹4,000
Amazon = ₹3,000

then:
Merchant-specific spending = ₹7,000
Remaining general online spending = ₹3,000

The total remains ₹10,000.

Every rupee should be counted once.

12. PERIODIC EXPENSES

Expenses such as:

• Rent
• Insurance
• Education

may occur monthly, quarterly, half-yearly, yearly or as a one-time payment.

DhanChakshu uses the user's stated frequency to estimate annual economics where appropriate.

A one-time expense is not automatically treated as a recurring annual expense.

13. RENT, EDUCATION AND INSURANCE

These categories receive special treatment because credit-card rewards do not automatically make these transactions financially worthwhile.

The engine considers:

• Whether the card rewards the transaction
• Whether the payment channel is eligible
• Transaction fees
• Applicable taxes
• Reward caps
• Category exclusions
• Other card-specific restrictions

The goal is to distinguish:
"Does this card earn rewards?"
from:
"Is using this card financially worthwhile?"

14. CAPS AND EXCLUSIONS

Credit-card rewards often contain:

• Monthly caps
• Statement-cycle caps
• Category caps
• Shared caps
• Merchant restrictions
• MCC restrictions
• Payment-channel restrictions
• Reward exclusions

DhanChakshu attempts to apply the relevant rules before calculating the final value.

Shared caps are not independently applied to every category when the issuer's rule indicates that they share the same limit.

15. ELIGIBILITY

A card can be financially attractive but still unavailable to a particular user.

DhanChakshu therefore considers relevant eligibility information separately from the reward calculation.

Eligibility is not the same as value.

The engine may identify:

• Eligible
• Potentially eligible / uncertain
• Not eligible
• Information unavailable

depending on the available card rules and user inputs.

16. CONFIDENCE

Confidence is intended to communicate how reliable the recommendation is based on the completeness and quality of the underlying information.

Confidence may be affected by:

• Missing product rules
• Unclear issuer terms
• Incomplete valuation information
• Ambiguous transaction treatment
• Data freshness
• Uncertain eligibility information

Where important data is not sufficiently reliable, DhanChakshu may avoid making a definitive recommendation rather than inventing a value.

17. DATA AVAILABILITY VS ZERO REWARD

DhanChakshu distinguishes between:

"No reward"
and:
"We do not have sufficiently reliable information to calculate the reward."

These are not the same.

A confirmed zero-reward rule can be used in the calculation.

An unresolved rule may instead result in a value being marked unavailable or a card being blocked from recommendation for that scenario.

18. ANNUAL VALUE

Annual value is an estimate based on the spending and assumptions supplied by the user.

Where precision would create false confidence, DhanChakshu may display a range rather than an exact amount.

For example:
₹22,000–₹26,500/year
rather than implying that the user will definitely receive exactly ₹24,183.

19. EXISTING-CARD OPTIMIZATION

When a user already has credit cards, DhanChakshu can compare the existing card against alternatives.

The analysis considers the estimated value of the current card under the user's spending pattern.

For a potential replacement:
Potential Additional Value = Recommended Card Value − Current Card Value

20. WHEN A SWITCH MAY NOT BE WORTHWHILE

DhanChakshu does not recommend changing cards merely because another card has a higher generic score.

As a V1 decision rule, a replacement is generally considered worthwhile only when the estimated improvement is meaningful.

The current threshold is:
At least ₹1,000 additional annual value

If the estimated improvement is below this threshold, DhanChakshu may tell the user that switching is not worthwhile based on the available assumptions.

21. OPTIMIZING EXISTING CARDS

For users with multiple cards, DhanChakshu can attempt to optimize spending across the existing portfolio.

The engine routes spending toward cards that provide the strongest incremental net value while considering:

• Reward rates
• Caps
• Exclusions
• Transaction costs
• Annual fees
• Eligibility
• Card-specific rules
• Portfolio interactions

22. ADDITIONAL CARD RECOMMENDATION

DhanChakshu does not recommend adding another card simply because another card has attractive rewards.

The additional card must provide meaningful incremental value over the existing portfolio.

The current V1 decision framework requires:
Incremental annual value ≥ ₹1,000
AND
Incremental annual value ≥ 1.5 × the effective annual cost of the additional card

The effective annual cost considers the annual fee, applicable fee waiver and only reliably quantifiable guaranteed offsets.

23. TWO-CARD SMART COMBO

The current V1 Smart Combo feature evaluates combinations of two cards.

It attempts to determine whether two cards used strategically across different spending categories can provide more value than relying on one card.

The current implementation is limited to two-card combinations.

Advanced three-card or larger portfolio optimization is a future capability and is not represented as a current V1 feature.

24. NO COMMISSION-BASED RANKING

Affiliate compensation does not enter the scoring formula.

The engine does not rank cards according to the commission DhanChakshu may receive.

This separation is fundamental to the DhanChakshu methodology.

25. RESPONSIBLE RECOMMENDATION

DhanChakshu is designed to optimize the value of spending that a user would make anyway.

It is not designed to encourage unnecessary spending merely to earn rewards.

Users should never spend beyond their means simply to achieve:

• Reward milestones
• Welcome bonuses
• Cashback
• Points
• Miles
• Fee-waiver thresholds

26. WHAT THE MODEL DOES NOT KNOW

No recommendation engine can know everything about a financial product or a user.

The analysis may not fully account for:

• Future issuer changes
• Merchant-specific processing behavior
• Changes in MCC classification
• Temporary promotions
• Personal circumstances not entered into the analyzer
• Changes in redemption value
• Application outcomes
• Future spending behavior

27. ISSUER TERMS ALWAYS PREVAIL

DhanChakshu's calculations are based on the information available to the engine.

If the issuer's current official terms differ from information displayed by DhanChakshu, the issuer's current terms prevail.

Users should verify important fees, rewards, exclusions and eligibility directly with the issuer before applying.

28. METHODOLOGY UPDATES

The DhanChakshu methodology may evolve as:

• Card products change
• Issuers change reward programs
• Better source information becomes available
• New transaction rules emerge
• The product's analytical capabilities improve

Material methodology changes may be reflected in future versions of this page.

29. CURRENT V1 ARCHITECTURE

The current DhanChakshu V1 analyzer is local-first.

The financial calculations are performed in the user's browser.

The current V1 does not provide:

• User accounts
• Cloud-saved analyses
• Server-side financial-profile storage
• Public share links
• Three-card-or-more portfolio optimization
• User behavior analytics

Future versions may introduce additional capabilities.

30. FINAL PRINCIPLE

DhanChakshu follows one simple principle:

A credit card is not valuable because it has the biggest reward number.

It is valuable when the rewards, benefits and costs make sense for the way you actually use it.

That is the "Right Nazar" behind DhanChakshu.