> ## Documentation Index > Fetch the complete documentation index at: https://docs.vortexiq.ai/llms.txt > Use this file to discover all available pages before exploring further. # Open Chargebacks, CyberSource > Open Chargebacks for CyberSource stores. Tracked live in Vortex IQ Nerve Centre. How to read it, why it matters, and how to act on it. **Metrics type:** [Key Metrics](/nerve-centre/overview#metrics-types-explained) • **Category:** [Payment Gateway](/nerve-centre/connectors#connectors-by-type) ## At a glance > Live count of chargeback / dispute cases currently in the merchant's queue requiring response action. The "queue depth" view: how many cases are open right now, how many are within their response deadline, how many are about to time out. The single most operationally urgent payments card on the dashboard for an enterprise merchant; missing the response deadline on a dispute means automatic loss regardless of evidence quality. | | | | ------------------------------------------ | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | **What it counts** | Every chargeback record from `/reporting/v3/chargebacks` whose status is `OPEN`, `PENDING_RESPONSE`, `IN_PROGRESS`, or `PRE_ARBITRATION_OPEN`. Excludes `WON`, `LOST`, `WITHDRAWN`, `CLOSED_NO_ACTION`. | | **Why open queue is operationally urgent** | Visa response deadlines run **7-14 days** depending on reason code (4853 fraud disputes typically 7 days; 4855 non-receipt typically 14 days). Mastercard runs **7-14 days**. Amex runs **20 days**. Missing the deadline means automatic loss regardless of how strong the evidence pack is. This card surfaces queue items approaching deadline. | | **Statuses included** | `OPEN` (just received, no response yet), `PENDING_RESPONSE` (response in draft / acknowledged), `IN_PROGRESS` (representment / evidence pack submitted, awaiting issuer decision), `PRE_ARBITRATION_OPEN` (issuer appealed merchant's win, second round). | | **Statuses excluded** | `WON` (merchant won the dispute), `LOST` (merchant lost), `WITHDRAWN` (customer withdrew), `CLOSED_NO_ACTION` (case closed without merchant action required). | | **Pre-arbitration handling** | Pre-arb cases are open queue items with their own deadline (typically 30 days) and are counted here. They are flagged separately in the [Open Chargebacks drilldown](/nerve-centre/kpi-cards/cybersource/open-chargebacks). Win rate on pre-arb runs 30, 40% so they need stronger evidence packs. | | **Reason-code breakdown** | The total count is the headline; the drilldown splits by reason code (Visa 4853 fraud, Visa 4855 non-receipt, Mastercard 4837 unauthorised, etc.) so the response team can prioritise by deadline-and-difficulty. | | **Currency** | n/a (count). Each open case carries its own currency in the drilldown. | | **Refunds** | Refunds issued *to close a dispute* don't subtract from this count instantly, the case status flips to `CLOSED_NO_ACTION` only after CyberSource confirms the issuer accepted the refund as resolution, typically 1-3 business days later. During the lag the case is still in this count. | | **Page cap** | `/reporting/v3/chargebacks` has no hard cap but typical enterprise merchant queues run 50, 500 open cases. The endpoint paginates if needed; full queue is typically pulled in 1, 5 API calls. | | **Time window** | `RT` (real-time, refresh every sync cycle, typically 5, 15 min). | | **Reporting API extraction lag** | Real-time. Chargeback API does NOT run an overnight batch like the conversion-detail report; it surfaces new cases as soon as the issuer files them. | | **Alert trigger** | `> 0`. Any open queue is a "look at this" signal because each item is on a clock. | | **Roles** | owner, finance, operations, dispute-ops | ## Calculation Calculated automatically from your CyberSource data. See the At a glance summary above for what the metric tracks and the worked example below for a typical reading. ## Worked example A US-based subscription streaming media merchant running CyberSource for monthly renewal billing across 14M active subscribers. State at 12 Apr 26 09:00 UTC, dispute-ops team daily standup. The card reads **243 open chargebacks**. The queue breakdown by status and reason: | Status | Reason | Count | Deadline median | Notes | | --------------------------------------- | ---------------------------------------- | ------- | ----------------------- | ------------------------------------------------------------------------------------------ | | `OPEN` (no response yet) | Visa 4863 (cardholder doesn't recognise) | 87 | 12 days | Largest bucket; "I don't recognise this charge" | | `OPEN` | Visa 4853 (fraud / cardholder dispute) | 41 | 6 days | Genuine fraud + friendly fraud mix | | `PENDING_RESPONSE` | Visa 4855 (non-receipt) | 28 | 8 days | Streaming = digital, so 4855 is rare; mostly customers claiming subscription was cancelled | | `IN_PROGRESS` | Visa 4853 | 33 | 22 days (issuer review) | Evidence packs submitted, awaiting decision | | `IN_PROGRESS` | Mastercard 4837 (no auth) | 19 | 18 days | Awaiting issuer decision | | `PRE_ARBITRATION_OPEN` | Visa 4853 | 14 | 21 days | Issuer appealed merchant's win | | Other (Amex, Discover, smaller buckets) | various | 21 | various | | | **Total open** | | **243** | | | Five things worth noticing for an enterprise dispute-ops lead: 1. **128 cases need a response within 7-14 days.** Of the 243 open, the ones in `OPEN` or `PENDING_RESPONSE` status (115 + 28 = 143) plus a portion of those in `IN_PROGRESS` are on a response clock. At a typical evidence-pack drafting rate of 8 cases per ops-rep per day, this queue takes one rep \~16 days to clear or four reps \~4 days. Most enterprise merchants run 3, 6 dispute-ops reps; a queue of 243 is operationally normal but needs steady throughput. 2. **The 87 Visa 4863 "doesn't recognise" cases are the highest-volume bucket and the easiest wins.** 4863 is typically resolved by sending the issuer a clear billing-descriptor refresh + customer email confirmation showing the recognised brand name. Win rate on well-defended 4863 cases runs 65, 80%. These should be batch-processed first because they're cheapest. 3. **41 Visa 4853 fraud disputes are the hardest defends.** Fraud disputes (genuine fraud + friendly fraud) require strong evidence packs: IP+device fingerprint, customer order history, communication logs, proof-of-delivery (n/a for streaming). Win rate on streaming-media 4853 cases runs 25, 40% because the digital-delivery proof is harder to compile vs physical-goods proof-of-delivery. Budget more time per case. 4. **14 pre-arbitration cases are time-sensitive AND high-stakes.** Pre-arb means the merchant won round one, the issuer appealed, and now there's a 30-day clock to file representment. Pre-arb win rate runs 30, 40% so \~9 of these will be lost regardless of effort. The ones that DO win require strongest evidence; allocate the senior dispute-ops reps to these. 5. **Heavy Visa-4863 share is a customer-experience signal.** When >35% of the open queue is "doesn't recognise" (this merchant: 87 ÷ 243 = 35.8%), the merchant has a billing-descriptor problem: customers don't connect the charge on their statement to the subscription they signed up for. Fixing the descriptor (clear brand name, support phone number, descriptor format `BRAND.COM 800-555-0100`) usually drops 4863 volume 30, 60% within 60 days because customers can self-resolve via support call without filing a dispute. This merchant's monthly chargeback volume runs \~750 (which feeds [Dispute Rate](/nerve-centre/kpi-cards/cybersource/dispute-rate) and [Chargeback Value](/nerve-centre/kpi-cards/cybersource/chargeback-value)). The 243 open queue represents roughly 30, 35% of monthly volume in the active state at any given time, which is operationally typical: cases enter the queue, sit in `OPEN` for 1, 3 days, move to `IN_PROGRESS` after evidence-pack submission, then resolve `WON` / `LOST` 14, 60 days later. A queue suddenly above 400 with the same monthly volume signals an ops-throughput problem (rep capacity, evidence-pack quality) not a dispute-volume problem. ## Sibling cards merchants should reference together | Card | Why pair it with Open Chargebacks | | ------------------------------------------------------------------------------------------ | -------------------------------------------------------------------------------------------------------------- | | [Dispute Rate](/nerve-centre/kpi-cards/cybersource/dispute-rate) | The rate-based view (vs this count-based view). Together they show queue depth + regulatory risk. | | [Chargeback Reason Codes](/nerve-centre/kpi-cards/cybersource/chargeback-reason-codes) | The reason-code split feeding this queue. | | [Chargeback Value (30d)](/nerve-centre/kpi-cards/cybersource/chargeback-value) | The dollar exposure of the queue. | | [Avg Dispute Response Time](/nerve-centre/kpi-cards/cybersource/avg-dispute-response-time) | How fast the team currently responds; if response time creeps up, the queue grows. | | [Dispute Win Rate](/nerve-centre/kpi-cards/cybersource/dispute-win-rate) | Of cases that resolved, what's the win rate? Below 35% = evidence packs need work. | | [Dispute Threshold Watch](/nerve-centre/kpi-cards/cybersource/dispute-threshold-watch) | Real-time alert at 90% of Visa cap. | | [Stripe Disputes Open (peer)](/nerve-centre/stripe/str_disputes_open) | Cross-processor peer for multi-acquirer enterprises. | | [PayPal Disputes Open (peer)](/nerve-centre/kpi-cards/paypal/open-disputes) | PayPal peer; PayPal disputes have different deadline structure (typically 10 days for buyer-protection cases). | ## Reconciling against the vendor's own dashboard **Where to look in CyberSource Business Center (EBC2):** * [Decisions → Chargebacks → Case Management](https://ebc2.cybersource.com/ebc2/app/RiskManagement/ChargebackManagement), the canonical row-level dispute case list. Filter by status `Open` / `Pending Response` / `In Progress` / `Pre-Arbitration Open`. * [Decisions → Chargebacks → Workflow View](https://ebc2.cybersource.com/ebc2/app/RiskManagement/ChargebackManagement), a Kanban-style view by deadline / reason / status that mirrors how dispute-ops teams actually work. **Why our number may legitimately differ from EBC2:** | Reason | Direction | What to do | | --------------------------------------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------- | | **Refresh lag**. We refresh on the standard 5-15 min sync cycle. EBC2 is real-time. | Vortex IQ may be 5-15 min stale on most-recent case openings / closings. | For a real-time live queue, EBC2 is authoritative. | | **Status mapping**. CyberSource exposes \~12 internal case statuses; we collapse to four queue states. | A case in transitional status (e.g. "evidence acknowledged, not yet acted on") may be in our `IN_PROGRESS` while EBC2 shows a different sub-status. | Tiny discrepancy possible; the case is genuinely open in both views. | | **Time zone**. EBC2 uses merchant-account tz; we use UTC. The deadline display can shift by ±1 day at boundary. | Tiny drift on deadline-display. | Real deadline is on the case record, not the display. | | **Pre-arb classification**. We surface pre-arb cases under `PRE_ARBITRATION_OPEN`; EBC2 sometimes splits "first round pre-arb" vs "second pre-arb". | Tiny drift in sub-grouping; total open count matches. | n/a. | **Cross-connector reconciliation:** | Comparison | Expected relationship | When divergence is legitimate | | --------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------ | | `cs_chargebacks_open` ↔ [`stripe.str_disputes_open`](/nerve-centre/stripe/str_disputes_open) | Independent counts (different processors). Sum gives total open queue across processors for a multi-acquirer merchant. | n/a. | | `cs_chargebacks_open` ↔ [`cs_dispute_rate`](/nerve-centre/kpi-cards/cybersource/dispute-rate) | The rate is a 30D denominator-aware view; this is a count-only point-in-time. They don't equate but both should grow / shrink together over time. | n/a (different metrics). | | `cs_chargebacks_open` ↔ EBC2 case-management count | Should agree within ±5 cases at any moment due to refresh lag. | Larger gap: investigate; usually a status-mapping edge case. | ***
Documentation cross-reference (for multi-processor enterprise merchants) The Stripe peer is [`stripe.str_disputes_open`](/nerve-centre/stripe/str_disputes_open); the PayPal peer is [`paypal.pp_disputes_open`](/nerve-centre/kpi-cards/paypal/open-disputes). Multi-processor enterprises should sum all three for the true cross-processor open-queue depth, since each processor's deadlines are independent.
## Known limitations / merchant FAQs **My queue suddenly jumped from 200 to 350 overnight. What happened?** Three common causes: (1) **issuer batch upload**, some issuers (large US banks especially) upload disputes in nightly batches; the day after the batch the queue can grow 30, 50%; (2) **fraud incident**, a card-testing or account-takeover attack against the merchant's checkout; check [Fraud Velocity Alert](/nerve-centre/kpi-cards/cybersource/fraud-velocity-alert) for clustering; (3) **descriptor change**, if the merchant recently changed their billing descriptor, a wave of "doesn't recognise" (4863) disputes follows for 30, 60 days as customers see the new descriptor on statements. **What's the typical response deadline?** Visa: 7-14 days depending on reason code (4853 fraud usually 7 days; 4855 non-receipt usually 14 days). Mastercard: 7-14 days. Amex: 20 days. Discover: 14 days. Pre-arbitration: 30 days. The card surfaces the per-case deadline in the drilldown so dispute-ops can prioritise. **What happens if I miss the deadline?** Automatic loss. The issuer's system flips the case to LOST regardless of whether the merchant has evidence to defend. The merchant pays the chargeback amount + the chargeback fee ($25-$50 typical). For a high-AOV enterprise merchant, a single missed deadline can be a \$5,000+ avoidable loss. This is why the queue card is so operationally important. **My win rate is fine but the queue keeps growing. What's wrong?** Throughput, not quality. A growing queue with stable win rate means cases enter faster than the team responds. Causes: (1) under-staffing on dispute-ops; (2) evidence-pack drafting taking too long (check [Avg Dispute Response Time](/nerve-centre/kpi-cards/cybersource/avg-dispute-response-time)); (3) too many manual steps in evidence collection (proof-of-delivery from carrier APIs, customer-support transcript exports, fraud-screening logs). Most enterprise dispute-ops teams should target 24-48 hour evidence-pack drafting; teams running 5+ days are usually doing too much manually. **Should I refund-to-close cases instead of defending?** Depends on the case. The decision matrix: (1) **refund-to-close** when (a) the customer's complaint is legitimate (broken product, non-delivery), (b) the win probability is \< 50% (e.g. lost 4855 / 4837 / Amex F30), (c) the dispute amount is \< customer LTV; (2) **defend** when (a) clear evidence of fraud / friendly-fraud, (b) high-value transaction, (c) precedent matters (high-volume disputes from similar pattern). Most enterprise merchants run 30, 50% refund-to-close on their open queue. **Why are 4863 "doesn't recognise" disputes the easiest to win?** Because they're typically not adversarial, the customer genuinely doesn't recognise the descriptor on their statement, and a clear evidence pack showing the brand name + recognised customer-facing identifiers usually resolves it. Win rates run 65-80% with well-designed evidence packs. This is also why a high 4863 share signals a billing-descriptor problem: clearer descriptors prevent the dispute from filing in the first place. **How does pre-arbitration work?** The merchant won round one. The issuer disagrees and files pre-arbitration with new evidence (e.g. customer affidavit). The merchant has 30 days to file a representment. If the issuer wins pre-arb, the merchant loses double-damage (chargeback + pre-arb fee, typically $50-$100). Pre-arb win rates run 30-40%, materially lower than first-round disputes (50-65% industry average), because the issuer wouldn't have escalated without new evidence. **Can disputes be filed long after the original transaction?** Yes. Visa allows up to **120 days** from transaction date for most reason codes (some recurring-billing variants extend to 540 days). Mastercard up to **120 days** typical, with extensions for merchandise non-receipt (start the clock at expected-delivery date). This is why dispute counts on long historical periods can be misleading, a transaction from 60 days ago can still file a dispute today. **My multi-currency global merchant, does this card sum across currencies?** The count is currency-neutral (one case = one count regardless of currency). The drilldown displays each case in its own currency. The aggregate dollar exposure (in [Chargeback Value](/nerve-centre/kpi-cards/cybersource/chargeback-value)) is summed by currency without FX conversion, matching the reporting convention. *** ### Tracked live in Vortex IQ Nerve Centre *Open Chargebacks* is one of hundreds of KPI pulses Vortex IQ tracks across CyberSource and 70+ other ecommerce connectors. Nerve Centre runs the detection layer; Vortex Mind investigates the cause when something moves; Ask Viq lets you interrogate any number in plain English. [Start for free](https://app.vortexiq.ai/login) or [book a demo](https://www.vortexiq.ai/contact-us) to see this metric running on your own data.