Abandoned Cart Recovery via AI Voice Calls: The D2C India Playbook for Shopify & WooCommerce

India's D2C sector crossed ₹2,00,000 crore in gross merchandise value in 2025. An estimated ₹1,56,000 crore of that — roughly 78% — was added to carts and never purchased. Cart abandonment is not a minor conversion problem. For most Indian D2C brands, it is the single largest addressable revenue leak in their entire funnel, bigger than acquisition cost, bigger than RTO, and almost entirely ignored.

The average Indian D2C brand running 10,000 monthly sessions loses ₹3-5 lakh per month to cart abandonment — assuming a 2.2% baseline conversion rate, ₹1,200 average cart value, and a 79% abandonment rate. Email recovery nudges 3-5% of those carts into orders. SMS recovers 1-2%. WhatsApp — the channel everyone is excited about — recovers 8-12% when done well. AI voice calls recover 10-18%. And they do it in a way none of the other channels can: by actually having a conversation.

This playbook covers the full architecture for running an abandoned cart recovery programme via AI voice calls — from the moment a customer closes the browser without buying, to the TRAI-compliant three-call sequence, to Shopify and WooCommerce webhook setup, to the ROI math that makes the decision obvious. It is written for Indian retail and e-commerce operators running D2C brands on owned-channel infrastructure.

Why India's Cart Abandonment Problem Is Worse Than the Global Average

The global cart abandonment rate sits at approximately 70%. India's rate is 78-82%. The 8-12 percentage point gap is not random — it is structural, and it points directly to where AI calling can intervene.

COD unavailability is the single largest driver. A customer from Patna or Vijayawada shopping a D2C brand for the first time encounters a payment page with only Razorpay, UPI, and card options. COD is either absent or restricted to returning customers. They add to cart to save the item, intending to come back, and never do. This is fundamentally different from Western abandonment, where the customer usually has a usable payment method and abandons due to price or intent uncertainty. Indian abandonment is often a payment-friction problem dressed up as a conversion problem. (AI COD order confirmation calls address the other side of this: customers who do choose COD but need confirmation before dispatch.)

Price checking across multiple platforms is the second driver. An Indian D2C shopper who finds a kajal pencil or a phone case on a brand's own website reflexively opens Meesho, Amazon, and Flipkart in parallel tabs before completing the purchase. They're not disinterested — they're doing 90-second competitive research. If the brand's checkout doesn't give them confidence during that window, the cart is abandoned.

Payment friction at checkout — particularly OTP timeouts, Razorpay/PayU loading delays, and UPI VPA failures — accounts for an estimated 12-18% of Indian e-commerce cart abandonment. These are technically recoverable carts where the customer had payment intent but encountered a system failure.

Impulse add-to-cart behaviour is endemic to Indian D2C, particularly in fashion, beauty, and home categories. Customers browse Instagram, click through to product pages, add multiple items to cart while comparing, and close the browser. These carts are large (₹1,800-3,500 average) but represent genuine purchase intent that just needs a nudge.

The implication for recovery strategy: Indian cart abandonment recovery needs to handle COD enquiries, price objections, payment failure reassurance, and impulse-buyer nudges — not just a generic "you left something behind" message. Email cannot handle objections. SMS cannot answer questions. Only voice can.

Why AI Voice Calls Outperform All Other Recovery Channels

The reason AI voice calls recover 10-18% of abandoned carts while email recovers 3-5% is not volume or frequency — it is the fundamental nature of the medium.

Real-time objection handling. When a customer says "yaar, Amazon pe same cheez ₹80 sasti hai," an AI voice agent can respond: "Haan, Amazon pe third-party sellers ki listings hote hain jo refund nahi dete. Hamare paas 7-day no-questions-asked return policy hai aur direct brand warranty." That response cannot happen over email. It cannot happen over SMS. It requires a real-time conversation, and AI can now conduct that conversation in Hindi, Hinglish, Tamil, Kannada, Bengali, and 10 other Indian languages at scale.

Language match creates trust. A customer from Jaipur who abandoned a kurta cart responds to a call in Rajasthani-inflected Hindi at an entirely different emotional register than to a generic English email. The language of the communication signals whether the brand sees the customer as a person or as a transaction. Caller Digital's AI voice agents run in 14 Indian languages, and language detection at the time of calling means a Tamil customer in Chennai automatically gets a Tamil-language call.

The call itself signals intent. When a brand calls a customer to follow up on an abandoned cart, it communicates something no message can: this brand cares enough to actually call. That signal alone creates reciprocity. Customers who receive an outbound follow-up call from a brand are 2.3x more likely to make a repeat purchase within 60 days, regardless of whether the recovery call converts the original cart.

Urgency is real-time. "Yeh item sirf 3 piece left hai" lands differently over a voice call than over an email. The customer cannot dismiss it the way they dismiss a push notification. They are on the call, in the moment, and the urgency is physically present.

Recovery calls are upsell opportunities. A customer who came back to buy one item can be offered a complementary product in the same call. Brands using upselling and cross-selling automation on recovery calls see 18-24% of converting callers take an upsell, adding ₹200-600 to the average recovered order value.

Cart Value Segmentation: The Critical Decision Framework

Not all abandoned carts should receive the same recovery treatment. Applying a ₹25 AI call to a ₹400 cart is economically marginal; applying a fully automated AI call to a ₹25,000 cart and hoping for conversion without human involvement leaves money on the table. The right framework segments carts by value and assigns a recovery track accordingly.

Under ₹1,500 — Fully automated AI call sequence

At this cart value, a 10% conversion rate on calls costing ₹8-25 per attempt produces a positive ROI even before accounting for the lifetime value of the customer. The economics work cleanly: 100 calls at ₹20 average = ₹2,000 cost, 10 conversions at ₹1,200 average = ₹12,000 recovered. Cost-to-recovery ratio of 16.7%. Run a three-call sequence, fully automated, with no human escalation required.

₹1,500-₹8,000 — AI call first, human escalation on warm signal

In this bracket, the customer has shown meaningful purchase intent. The AI makes the first call and qualifies the objection. If the customer expresses a warm signal — asking about EMI, asking for a specific colour variant, mentioning that their spouse needs to see it — the AI flags the call for human follow-up within 2 hours. The warm-signal escalation converts at 35-45% when the human calls back promptly, versus 12-18% for a cold call.

Over ₹8,000 — AI qualification + human closer within 4 hours

High-value carts (electronics, jewellery, premium fashion) require a consultative close. The AI call confirms intent and gathers objections, and a human sales agent follows up within 4 hours with detailed product knowledge and the authority to offer a personalised incentive. The AI call is not a recovery attempt — it is a qualification and scheduling tool for the human closer.

Healthcare diagnostics cart abandonment — always human

A customer who abandoned a full-body checkup package or a cancer screening test on a diagnostics platform (Healthians, Redcliffelabs, Thyrocare) should never be recovered by a fully automated AI call. The purchase involves health anxiety, possibly a doctor's recommendation, and emotional sensitivity. AI can make a gentle first-touch call with a soft script, but any signal of hesitation should trigger immediate warm transfer to a healthcare counsellor.

The 3-Touchpoint Calling Sequence

Timing matters more than the message. A perfect script delivered 48 hours after abandonment recovers 40% fewer carts than an adequate script delivered 30 minutes after abandonment. The three-call sequence below is based on observed performance across Indian D2C brands in fashion, beauty, electronics, and home categories.

T+30 minutes: First call — the assistance frame

The first call should not lead with an offer or a discount. At 30 minutes, the customer's attention has moved on but the cart is still mentally present. The opening should be framed as assistance, not sales.

"Namaste [Name], main [Brand] ki taraf se bol rahi hoon. Aapne thodi der pehle apni cart mein kuch items daale the — kya koi problem aayi order complete karne mein?"

This opening creates a support frame rather than a sales frame. Customers who say "haan, payment fail ho gayi thi" can be immediately assisted. Customers who say "nahin, bas soch rahi hoon" can be offered a soft incentive or product information. The T+30 call converts at 14-20% when the script is in the customer's preferred language.

T+4 hours: Second call — urgency and value reinforcement

By four hours, the customer has had time to compare prices and talk to themselves out of the purchase. The second call needs to add new information — either urgency (stock signal) or value (price guarantee, free shipping threshold).

"Namaste [Name], [Brand] se call hai. Aapki cart mein jo [product name] tha, woh last 4 piece hi bacha hai. Aur aaj raat tak free delivery offer bhi chal rahi hai. Kya main aapki order abhi confirm kar doon?"

The T+4 hour call converts at 8-12%. Together, the first and second calls recover approximately 60% of ultimately recoverable carts.

T+24 hours: Final call — specific discount, time-limited

The third call offers a concrete, time-limited incentive. Vague offers ("special discount for you") underperform specific offers ("10% off if you order in the next 2 hours") by 35-40%. The incentive should be calculated against the cart value: for a ₹1,200 cart, a 10% discount (₹120 cost to brand) is justified given the ₹1,200 recovery value. For a ₹400 cart, free shipping (₹50-80 cost) is more appropriate than a percentage discount.

"Namaste [Name], last time ke liye bol raha/rahi hoon — aapki [product] ke liye aaj hum 10% discount de rahe hain sirf agle 2 ghante ke liye. Link main SMS karoon aapko?"

The T+24 hour call converts at 5-8%. The three-call sequence combined recovers 22-35% of contactable abandoned carts.

A note on contactability: across Indian D2C brands, 55-65% of abandoned cart customers have a verified phone number in the checkout data. Of those, 45-55% answer at least one call in the three-call sequence. This means the effective recovery pool is approximately 25-35% of total abandoned carts, and the 10-18% overall recovery rate is calculated against the full abandoned cart volume including uncontactable numbers.

Shopify + WooCommerce Integration Setup

The technical integration between a Shopify or WooCommerce store and an AI calling platform is simpler than most brands expect. For most stores, setup takes 1-2 business days.

Shopify webhook integration

Shopify's native

checkouts/create

and

checkouts/update

webhooks fire on cart events, but for reliable abandoned cart detection, use the

carts/update

webhook combined with a 30-minute inactivity window. When a customer adds items to cart and has provided a phone number (typically after reaching checkout step 1 or 2), Shopify can be configured to fire a webhook to the calling platform.

The webhook payload contains:

• customer.phone

  — the customer's mobile number

• line_items

  — array of products (name, quantity, price, variant)

• total_price

  — cart value

• customer.default_address.city

  — for language routing

• token

  — cart recovery URL identifier

The AI calling platform receives this payload, waits the configured delay (30 minutes for the first call), scrubs the number against the NDND/DND registry, and if the number is reachable, initiates the outbound call with a script personalised to the specific products in the cart. Caller Digital's Shopify integration handles all of this natively with no custom development required.

WooCommerce webhook integration

WooCommerce does not have native abandoned cart webhooks. You need either the "Abandoned Cart Lite" plugin (free, 300K+ installs) or "WooCommerce Abandoned Cart Pro" (paid). Both plugins fire a

woocommerce_cart_abandoned

action hook with the cart data when a logged-in or guest customer (with an email/phone captured) abandons after a configurable inactivity window.

Configure the plugin to:

1. Capture guest checkout data at first email/phone field entry (not just on cart page)
2. Fire webhook after 25 minutes of inactivity (giving the AI platform time to initiate the T+30 call)
3. Send payload to the calling platform's inbound webhook endpoint

The WooCommerce payload structure mirrors Shopify: customer phone, cart contents, cart value, and a recovery token that pre-fills the checkout on click. Caller Digital's WooCommerce integration processes this webhook and handles the full three-call sequence with no manual intervention.

What the AI does with cart data

The product names and cart value from the webhook are used to personalise the call script in real time. A customer who abandoned a "Navy Blue Linen Kurta, Size M, ₹1,199" gets a call that mentions the navy blue linen kurta specifically — not "the item in your cart." This level of personalisation cannot be achieved through SMS or push notifications at any meaningful scale, and it is the primary reason voice outperforms other channels on recovery rate.

For brands with complex catalogues (multi-variant fashion, electronics with multiple specifications), the AI uses the product name from the SKU description. For bundles, the AI mentions the bundle name plus the total value.

The Multi-Brand / Multi-SKU Marketplace Challenge

Brands selling on Amazon, Flipkart, or Meesho face a fundamental constraint: marketplace carts are not accessible via webhook. If a customer adds a product to their Amazon cart and abandons, the brand has no visibility into that event, no customer phone number, and no ability to trigger a recovery call. Marketplace infrastructure is designed to prevent exactly this kind of direct brand-to-customer communication.

The practical implication is that abandoned cart recovery via AI voice calls is exclusively a D2C channel play. Brands operating both marketplace and D2C channels should invest their recovery infrastructure entirely in the D2C funnel, and separately invest in pushing marketplace customers to the owned channel through packaging inserts, QR codes, and post-delivery follow-up calls.

The economics of this channel migration justify the investment: a customer who purchases through the D2C channel instead of Amazon generates 15-22% higher margin (no marketplace commission), is recoverable for abandoned carts, and can be reached for upsell and loyalty calls — all of which are impossible for marketplace customers.

For brands that run promotions on both channels, a practical strategy is to make COD available only on the D2C channel and to price the D2C channel at parity or slightly below marketplace after accounting for the marketplace commission savings. This gives price-sensitive customers a genuine reason to purchase direct, and positions the D2C channel as the recovery-ready funnel.

Hindi Script Templates by Vertical

The following scripts are production templates for three major D2C verticals. Each script is designed for the T+30 minute first call and can be adapted for the T+4 hour and T+24 hour calls by adding the urgency or discount layer.

Fashion (kurta, ethnic wear, casual wear)

"Namaste [Name ji]! Main [Brand Name] se bol rahi hoon. Aapne abhi kuch time pehle [Product Name] apni cart mein daali thi — size [size], ₹[amount] ki. Kya order complete karne mein koi takleef aayi? Main abhi help kar sakti hoon."

[Customer says payment failed]: "Koi baat nahi. Aap UPI se try karein ya main aapko direct payment link bhej deti hoon — do second mein payment ho jaayegi. Cart abhi bhi save hai."

[Customer says still thinking]: "Bilkul, sochiye. Ek baat bataaun — yeh [Product Name] fast move kar rahi hai, aaj kal ke orders mein zyada chal rahi hai. Agar aap chahein toh main ₹50 off apply kar deti hoon abhi."

Electronics (phone accessories, wearables, small appliances)

"Hello [Name], [Brand] ki taraf se call hai. Aapne [Product Name] cart mein add kiya tha — ₹[amount] wala. Kya aapko koi specific question tha product ke baare mein? Warranty ya compatibility ke baare mein bata sakta hoon."

[Customer asks about warranty]: "Haan, isko 1 saal manufacturer warranty hai aur hamare paas 15-day replacement policy hai agar koi issue aaye. Aaj ka order kal tak deliver ho jayega."

[Customer mentions Amazon comparison]: "Amazon pe jo listings hain wo mostly third-party sellers hain bina warranty ke. Hamaare paas official brand warranty hai aur COD bhi available hai."

Beauty and skincare

"Namaste [Name], main [Brand] se hoon. Aapki cart mein [Product 1] aur [Product 2] tha — kya koi confusion tha shade ya skin type ke baare mein? Main guide kar sakti hoon."

[Customer unsure about shade]: "Aapka skin tone kaise hai — fair, medium ya dusky? [Answer] ke liye [specific shade] best rahega. Bahut customers ka favourite hai yeh."

[Customer says too expensive]: "Samjhi. Agar aap [Product 1] sirf laana chahein toh [Product 2] baad mein bhi add kar sakte hain. [Product 1] pehle try karein — 7-day return hai agar suit nahi kiya."

The pattern across all three verticals is the same: the AI leads with an assistance frame, uses the specific product name from the cart data, handles the most common objections with prepared responses, and offers a concrete next step. Generic scripts without product personalisation recover at 4-6%; personalised scripts with product names recover at 10-18%.

A/B Testing Framework for Recovery Call Optimisation

Running an abandoned cart recovery programme without a structured A/B testing framework means leaving 30-40% of recoverable revenue on the table. The following variables should be tested in sequence, not simultaneously, over 4-6 week windows with minimum 500 calls per variant.

Call timing: T+30 minutes vs T+2 hours (first call)

This is the highest-impact variable. In testing across fashion and beauty D2C brands, T+30 minute calls outperform T+2 hour calls by 28-35% on conversion rate. The mechanism is recency: at 30 minutes, the cart is still the most recent shopping activity in the customer's memory. At 2 hours, it is competing with everything else that has happened.

Benchmark: T+30min achieves 14-20% conversion on answered calls; T+2hr achieves 10-14%.

Opening line: question frame vs offer frame

Two variants:

• Question: "Kya order complete karne mein koi issue aayi?"
• Offer: "Aapki cart ke liye special 10% off hai abhi."

The question frame outperforms the offer frame on conversion rate (16% vs 12%) because it positions the brand as helpful rather than promotional. However, the offer frame performs better with customers who have previously purchased and are price-sensitive. Segment by new vs returning customer for optimal results.

Discount framing: percentage vs absolute amount vs free shipping

For cart values under ₹1,000:

• "10% off" vs "₹100 off" vs "Free shipping" — in testing, "₹100 off" outperforms "10% off" even though they are identical amounts. Absolute values anchor more strongly for lower-income segments.

For cart values ₹1,000-₹3,000:

• Free shipping (₹70-100 value) performs on par with 5% discount, and costs the brand less when the cart value is high.

Single call vs 3-call sequence

Brands new to recovery calling often run a single call and measure results. The uplift from adding a second and third call is significant: second call adds 6-9 percentage points to total recovery rate; third call adds a further 3-5 percentage points. The marginal cost of calls 2 and 3 is low (customer is already in the system), and the ROI is strongly positive.

TRAI Compliance for Abandoned Cart Recovery Calls

This is the section most Indian D2C brands get wrong, and getting it wrong exposes them to TRAI penalties of up to ₹25,000 per call.

Abandoned cart calls are promotional, not transactional

The most common compliance mistake is classifying cart abandonment calls as "transactional" — because the customer started a transaction. This is incorrect. Under TRAI's Telecom Commercial Communications Customer Preference Regulations (TCCCPR), a communication is transactional only if it relates to a transaction that has been completed. A cart that was not purchased is not a completed transaction. Abandoned cart recovery calls are promotional communications.

This means all of the following apply:

DND scrubbing is mandatory. Before placing any recovery call, the customer's mobile number must be checked against the National Do Not Disturb (NDND) registry. Numbers registered on DND cannot receive promotional calls. If your AI calling platform does not perform automatic NDND scrubbing before every outbound call, you are non-compliant.

Promotional calling hours apply: 9am-9pm only. Recovery calls cannot be placed outside this window, even if the abandonment occurred at 11pm. Queue calls triggered outside hours for the next morning's 9am window.

DLT template registration is required. All call scripts must be registered as promotional templates on the TRAI Distributed Ledger Technology (DLT) platform before use. Template registration takes 2-5 business days. Changes to the script require re-registration.

Commercial number series (140x) must be used. Outbound promotional calls must originate from a 140x number series, not a regular mobile or landline number. Using a personal mobile number for promotional calls is a TRAI violation. Caller Digital's platform uses a compliant 140x number series by default.

What is NOT required: additional explicit consent for the call

If the customer accepted the brand's terms and conditions during checkout — which virtually all Indian checkout flows include — and those T&Cs include language about marketing communication consent (which almost all standard T&Cs do), no separate opt-in for the voice call is required. The existing checkout T&C acceptance covers promotional voice calls to the checkout phone number.

Practically: integrate NDND scrubbing as a pre-call step in your workflow, register your call scripts on DLT before launching, ensure your calling number is on the 140x series, and respect calling hours. These four steps make your cart recovery calling programme fully TRAI-compliant.

ROI Calculation: The Math for a D2C Brand

The following calculation is for a D2C brand in the fashion and lifestyle category with 5,000 abandoned carts per month and an average cart value of ₹1,200.

Input assumptions:

• Abandoned carts per month: 5,000
• Average cart value: ₹1,200
• Contactable (phone number available): 60% = 3,000 carts
• Answer rate across 3-call sequence: 50% = 1,500 unique answers
• Recovery conversion rate (of answered calls): 12% = 180 orders recovered
• Average order value at recovery (including upsell): ₹1,350

Revenue recovered per month: 180 × ₹1,350 = ₹2,43,000

Cost of calling (3 calls per cart, average ₹15/call):

• 3,000 carts × 3 calls = 9,000 calls
• At ₹15/call: ₹1,35,000

(See voice AI pricing in India for a full breakdown of per-call costs by platform and volume tier.)

Net recovery after calling cost: ₹2,43,000 − ₹1,35,000 = ₹1,08,000 net per month

Comparison: email recovery

Email conversion rate: 4%, cost negligible (₹0.10/email for bulk)

• 5,000 emails × 4% = 200 recoveries at ₹1,200 = ₹2,40,000
• Cost: 5,000 × ₹0.10 = ₹500
• Net: ₹2,39,500

At first glance, email appears more profitable. But this ignores three factors: (1) email requires a valid email address, collected in only 40-50% of Indian D2C checkouts vs 85-90% for phone numbers; (2) email recovery rates of 3-5% assume an active inbox — Indian email open rates for promotional messages are 8-12% (vs 18-22% for well-run B2C email lists globally); (3) email cannot handle objections, so the 4% who convert are only the easiest recoveries. Voice calling recovers the 8-14% who have objections that can be resolved.

The complete recovery stack: email + voice + WhatsApp

The highest-performing D2C recovery programmes run all three channels in a coordinated sequence: email at T+1 hour (low cost, catches easy recoveries), WhatsApp at T+3 hours (8-12% recovery on those who open), AI voice call at T+30 minutes and T+4 hours (catches the objection-driven abandonment that email and WhatsApp cannot address). Total programme cost: ₹2-4L/month for a 5,000-cart-per-month brand; total recovery: ₹5-8L/month.

This is also the funnel that supports the post-purchase confirmation and upsell playbook — once a cart is recovered via voice, the brand has a confirmed customer who can be enrolled in the post-purchase call sequence for upsell and retention.

Case Study: Fashion D2C Brand, 8,000 Carts/Month, 78% Abandonment

A direct-to-consumer ethnic wear brand based in Surat was running 35,000-40,000 monthly sessions on their Shopify store with an 80% abandonment rate and approximately 8,000 abandoned carts per month at an average cart value of ₹1,480. Their existing recovery programme consisted of a 3-email sequence that was recovering approximately 220 carts per month (2.75%).

Month 1: AI voice calling programme launched. Three-call sequence configured with Shopify webhook integration (1.5 days setup). NDND scrubbing integrated, 140x number series activated, DLT templates registered. First 2,500 calls made in week 3 of the month (delayed by DLT template approval). Recovery rate: 8.2% of answered calls, 312 additional carts recovered.

Month 2: Script optimised based on Month 1 call disposition data. The most common abandonment reason identified: COD unavailability for first-time customers. The brand added COD as an option for first-time customers with cart value under ₹1,500, and the AI was scripted to mention COD availability as a payment option during the recovery call. Recovery rate jumped to 11.7%, 687 carts recovered.

Month 3: Full three-call sequence deployed with correct timing. Upsell script added to the T+30 minute call for confirmed buyers. Recovery rate: 12.4%, 992 carts recovered at average ₹1,480 + ₹210 upsell = ₹1,690 per recovered order.

Month 3 recovered revenue: ₹16,76,480

Less calling costs (approx ₹3.8L for 24,000 calls across 8,000 carts × 3 calls):

Net recovered revenue, Month 3: ₹12,96,480 — approximately ₹11.9L net after programme overhead

This was from ₹0 in Month 0 (no voice recovery programme at all). The Shopify integration cost was a one-time ₹0 (Caller Digital's native Shopify integration). The DLT registration cost was ₹5,000 one-time. The ongoing programme cost is fully variable — pay per call, per outcome.

The brand has since extended the programme to cover their post-purchase confirmation sequence (reducing COD RTO by 31%) and a quarterly customer reactivation sequence — a pattern consistent with how high-growth Indian D2C brands build a complete voice AI stack one use case at a time.

Building the Right Abandoned Cart Recovery Stack

The decision to run AI voice calls for cart recovery is not a technology decision — it is a revenue arithmetic decision. For any D2C brand with more than 1,500 abandoned carts per month and an average cart value above ₹800, the ROI is positive even at conservative recovery rates.

The setup checklist:

1. Shopify or WooCommerce webhook configured to fire on cart abandonment with phone number and cart data (1-2 days)
2. NDND scrubbing integrated as a pre-call step in the workflow (automated with Caller Digital)
3. DLT templates registered for all three call scripts in all planned languages (2-5 days)
4. 140x number series active (provisioned by calling platform)
5. Call timing configured: T+30min, T+4hr, T+24hr
6. Cart value segmentation logic set: under ₹1,500 = full automation; ₹1,500-₹8,000 = warm escalation; over ₹8,000 = human closer
7. Language routing configured based on customer city or checkout language preference
8. Reporting dashboard live with per-call disposition, recovery rate, revenue recovered, and cost per recovered cart

The full programme, from webhook to first call, can be live in under a week for most Shopify and WooCommerce stores. The question is not whether it works — the data is unambiguous. The question is how many carts are recoverable right now that no channel is currently reaching.

---