Integrated IT services E-commerce vertical Book a call

The defense layer for online commerce —
fraud, bots, chargebacks, abuse.

Mandala IT stops the revenue leakage that mid-market e-commerce silently bleeds: card testing, bot scraping, chargebacks, promo abuse, account takeover. Performance-based billing — pay against losses prevented and disputes won.

Get a proposal View capabilities
  • 0 E-com fraud losses by 2029 (USD)
  • 0 Annual growth (industry band)
01 · Market

A $100B+ leakage problem, compounding with every online channel.

Online fraud is no longer a back-office line item — it's a top-of-funnel margin drain.

E-com fraud losses today

0

Approximate global online-fraud losses in 2024 (industry research, $40–48B band).

E-com fraud losses by 2029

0

Projected annual losses — driven by AI bots, real-time payments, and mobile commerce.

Chargebacks as % GMV

0

Industry average across mid-market merchants — fees stack on top of the principal.

Why losses keep growing

Bot tooling is now cheap and AI-augmented. Every quarter the attacker side improves; defense has to compound to match.

Why the model is shifting

Performance-based billing — pay per chargeback won, per fraud loss prevented — flips the incentives. That is exactly where Mandala IT is positioned.

02 · Fraud surface

The fraud landscape Mandala IT is built to defend against.

Eight overlapping abuse categories — every one is growing in scale, automation, and AI sophistication.

03 · The company

Mandala IT — purpose-built for online-commerce risk.

Mandala IT operates inside the rapidly expanding e-commerce defense sector, delivering an integrated detection-and-evidence layer designed to block fraud, stop bots, and win chargebacks in real time.

  • Real-time fraud scoring on every checkout and login
  • Bot & scraping defense at the inventory and search endpoints
  • Chargeback dispute evidence packaging — automated
  • Promo / loyalty abuse detection with device-level linkage
  • Continuous tuning to balance conversion vs. fraud cost

“The right e-commerce defense engine raises conversion and lowers loss simultaneously — they are not in tension if the signals are good enough.”

04 · Differentiator

Performance-based defense. You pay against chargebacks won, false declines reduced, and bot traffic stopped — line items your CFO can audit.

Mandala IT introduces a model where merchants primarily pay against chargebacks won, fraud losses prevented, and bot traffic stopped.

01

Lower barrier to entry

Adoption is easier because payment tracks actual loss prevented.

02

Aligned vendor incentives

Provider revenue rises when fraud rate drops and chargeback wins compound.

03

Conversion-aware tuning

False-decline reduction is a billable KPI. Our team actively tunes against over-rejection so merchant revenue stays optimized while losses fall.

04

Recurring revenue potential

Continuous traffic = continuous scoring = continuous billing.

05

Mid-market opportunity

Shopify+ and BigCommerce-tier merchants lack scalable in-house fraud teams.

06

Fraud-scoring engine

Every session and every transaction gets a 0–100 risk score. Each prevented loss is a billable event.

05 · Pillars

What we cover for you.

Five pillars under one engine — the bundle every mid-market merchant buys together.

Defense pillars

  • Payment fraud (CNP + card-testing)
  • Bot & scraping defense
  • Chargeback mitigation
  • Promo & loyalty abuse

What it prevents

  • Stolen-card use at checkout
  • Inventory hoarding and denial-of-inventory
  • Friendly fraud and repeat-disputer losses
  • Multi-account promo / referral abuse

How it operates

  • Real-time scoring at checkout and login
  • Device, network and behavioral fingerprinting
  • Continuous conversion-vs-fraud tuning
  • Performance-based commercial model
Case study · anonymised under client NDA

How we lifted a Shopify Plus merchant's chargeback win-rate from 41% to 73%.

Buyer profile

D2C apparel · USD 400M GMV

Platform
Shopify Plus
Engagement
Fraud + chargeback dispute desk · 12 months
Team deployed
6 fraud analysts + 2 dispute specialists
Stack
Riskified (kept and retuned)
  1. 73% Chargeback win-rate ↑ from 41%
  2. −0.8pp False-decline rate on legitimate buyers
  3. $6.2M Annual loss prevented audited by CFO

The situation

The merchant was losing 1.4% of revenue to chargeback fees and friendly fraud, and false-decline rate on legitimate buyers was 4.8% — the CFO's biggest line-item complaint to the CMO. The existing Riskified deployment was set too conservatively.

What we did

  1. 01Took over the dispute desk and rebuilt evidence packaging templates aligned to Visa CE 3.0 and Mastercard MCMP.
  2. 02Retuned Riskified decision thresholds against measured false-decline cost — not the vendor's default risk score.
  3. 03Added device + behavioural linkage on multi-account promo abuse rings.
  4. 04Reported weekly to the CMO on conversion vs. fraud — a single number both sides could action.
06 · Architecture

A six-layer commerce-defense stack — verdict + dispute evidence on every event.

Every relevant event is evaluated by six cooperating layers in under two seconds.

  1. L1

    Identity

    Decide who is on the other end — and whether they can buy.

    • Device fingerprinting and step-up on cart-value triggers
    • Account-takeover detection on every authenticated session
    • Multi-account linkage (same device, shared addresses, shared cards)
  2. L2

    Behavioral monitoring

    Decide whether the shopper is real, in real time.

    • Basket anomalies, mouse path, typing cadence
    • Geography / device / IP heuristics
    • Velocity, value-deviation and dormant-to-active detection
  3. L3

    Web & API protection

    Stop the bot or payload at the door of the store.

    • Inventory-endpoint and search-endpoint bot mitigation
    • Continuous scanning of public-facing endpoints
    • API abuse detection — scraping, enumeration, business-logic abuse
  4. L4

    Threat intelligence

    Decide whether the source is already known to be hostile.

    • Stolen-card BIN lists and fraud-ring IP ranges
    • Live phishing-domain and bot-network feeds
    • Cross-tenant signal sharing
  5. L5

    Control plane

    Your team's cockpit. Where our hours and your per-event fees are reconciled each month.

    • Multi-store administration with per-brand isolation
    • Real-time dashboards — fraud blocked, chargebacks open, win rate
    • Multi-channel alerting and billing
  6. L6

    Chargeback evidence packaging

    Produce the dispute package the issuer / acquirer will accept.

    • Automated evidence collection — device, session, delivery, communication
    • Issuer-bank-formatted dispute responses
    • Win-rate analytics and dispute-pattern learning
A suspicious checkout, end to end · under 2 seconds
  1. 01Auth
  2. 02Basket shape
  3. 03Bot check
  4. 04BIN reputation
  5. 05Risk score
  6. 06Verdict
  7. 07Approve · 3DS · decline
08 · Engage

Engage Mandala IT across the defense surface of online commerce.

Engagement terms

Issuer
Mandala IT (IT services consultancy)
Engagement model
Project · Managed · Retainer · Staff-aug
Minimum engagement
From USD 10,000 per engagement
Sector
E-commerce defense · fraud · bot & abuse
Pricing model
Outcome-aligned hybrid
Status
Open
Request proposal Book a 30-min discovery call instead

Tell us about your project

Leave your details — a Mandala IT consultant will follow up within one business day\.

How we handle your data — Privacy policy

Note. Scope, deliverables, timelines, and SLA tiers are agreed in a mutual Statement of Work. Commitments on this page are illustrative; binding terms live in the engagement contract.

09 · FAQ

Quick answers.

How does Mandala IT plug into our existing checkout?

Lightweight JS tag + server-side webhook. Compatible with Shopify Plus, BigCommerce, custom carts.

Does Mandala IT guarantee a chargeback win rate?

No vendor can. Mandala IT commits to measurable improvement vs. your current win rate, billed against actual wins.

Will tight fraud rules kill conversion?

False-decline reduction is a billable KPI. Our team actively tunes against over-rejection so merchant revenue stays optimized.

Which merchant tier does Mandala IT target?

Mid-market online merchants — Shopify Plus / BigCommerce / Magento tier.

Do you guarantee outcomes?

We commit to measurable improvement against your baseline — quantified per engagement in the Statement of Work. SLA-backed commitments and clear remedies, not vague guarantees.

How fast can you start?

Scoping call within 2 business days. Signed Statement of Work typically within 7–14 days. Monitoring live within 30 days of SOW for standard engagements; emergency incident-response retainer can be activated within 24 hours.

Do you sub-contract any of the work?

We deliver primarily with directly-employed analysts and engineers. Where a vertical needs specialist coverage (forensics, firmware analysis, jurisdiction-specific filings), named partners are disclosed in the SOW before signature — never silently white-labelled.

What if we're already using Signifyd / Forter / Riskified / DataDome?

We integrate with your existing fraud / bot stack rather than displace it. Our team tunes decision thresholds against measured false-decline cost, runs the chargeback dispute workflow, and owns the win-rate analytics. If a tool is genuinely under-performing your conversion / loss targets, we say so in writing.

Where does our data live? Can we keep it in-region?

Region-specific options — EU, UK, US, Israel, GCC — are scoped per engagement. BAA (US healthcare), DPA (EU), and ISO 27001-aligned controls are issued under the engagement contract. Production data and PII do not leave your designated region without written consent.

Can we see reference clients?

After the first scoping call, under mutual NDA. Most of our clients are regulated and contractually cannot be named publicly. Reference calls with comparable-size buyers in your vertical are arranged before SOW signature.