Conversational agent prototype for activation — lead qualification and SR collection
Eric San · Updated 16 February 2026 (v7)
This is a prototype WhatsApp bot that qualifies sourcing leads and guides them toward completed SRs. We tested it by replaying real lead messages from the test period — the lead side is real data, the bot side shows how the prototype responds.
Instant Response
24/7
no offline hours
SR in 3 Turns
3
qualified lead → complete SR
Trust-First
T1
answers price/MOQ/delivery before asking
Smart Exit
1-2
turns to filter with specific re-entry door
Context: WhatsApp direct is the strongest channel for quality SRs (4.6% CVR vs 0.5% for forms). But ~76% of WA leads ghost after 1-2 messages — often because the first response doesn't give them enough value to continue. The bot addresses this with category expertise upfront and conversational SR collection that doesn't feel like a form.
I. What We Did
We studied the gold standard conversation, Eugene's requirements document and key findings on lead behavior, and 13 raw WhatsApp conversations from the test period — plus the A/B experiment data (variants 1-6, CVR breakdown, quality SR classification). From these we built a conversational agent and iterated the prompt through 7 versions based on Karl's feedback. The current version (v7) was tested against 8 lead personas drawn from real WA chat data: 8/8 pass, avg score 8.8/10. Key changes in v7: prices-first (when lead names category, first response includes price range), 2-3 line message cap, one-liner answers for catalog/process questions. The iteration methodology is fully reproducible — test personas, prompt versions, and scoring rubric are documented in the GitHub repo.
II. Conversations
We tested v7 against 8 leads drawn from real WhatsApp chat data. Below are the key conversations grouped by outcome: leads that reached a completed SR, leads handed off to the human team, leads correctly exited, and leads that would have been lost under manual handling but were kept alive by the bot. Each conversation includes bounce analysis — what could cause the lead to leave at each turn, and how the bot mitigates it.
Leads → Completed SR
Jesús — Mexico, sportswear for new store, 70K MXN
SR IN 3 TURNS
Real lead from Feb 6, 2026. Lead messages replayed exactly from chat data.
Lead — real message from test dataBot — v7 response from OpenClaw
Hola, quisiera ver el catalogo de sus productos
Bounce risk: he wants a "catálogo" — if just told "we don't have one" he'll leave.
No tenemos catálogo fijo — trabajamos directo con fábricas según lo que necesites. ¿Qué producto estás buscando?
→ One-liner: no catalog, reframed as advantage, straight to "what do you need?" No filler.
tenis deportivos ropa deportiva de hombre y mujer camisas de futbol
¡Buenas categorías! Tenis deportivos con tu logo van de $8-15/par, ropa deportiva $4-12/pieza, y camisas de futbol $5-10/unidad. ¿Cuántas piezas estás pensando para empezar?
→ Prices-first for all 3 categories. One qualifying question. 2 lines.
Apenas voy abrir un local amigo. Quiero algo con mi marca. Mi presupuesto es 70 mil pesos mexicanos de todo surtido
Bounce risk: "apenas voy a abrir" — new store, no company. Could trigger premature exit. But 70K MXN = ~$4,000.
70 mil pesos son ~$4,000 USD, buen presupuesto para arrancar con las 3 categorías. Con eso podrías surtir ~150 pares de tenis + 200 piezas de ropa deportiva + 150 camisas de futbol, todo con tu marca. ¿Ya tienes logo y diseños listos?
→ Did the budget math immediately ($4,000 = viable). → Broke it down by category. → One question to advance.
What the bot did: Converted a lead with no company, asking for a catalog that doesn't exist, in a language other than English — into an advancing conversation in 3 turns. Catalog question got a one-liner (not a paragraph). Budget math ($4,000 = 500 pieces across categories) prevented a premature exit. Prices delivered immediately once categories were named.
femmoraaa — Teenager in Pakistan, Instagram jewelry brand
ADVANCING IN 3 TURNS
Real lead from Feb 3-4, 2026. Originally classified as low-quality by the growth team.
Lead — real message from test dataBot — v7 response from OpenClaw
Hi, I want to source an item. I wanted to source some bulk amount of jewellery and accessories from china, can you give me a breakdown of how does it works and how much of investment do i have to make as a teenager. i would be waiting for the reply
Jewelry from factories runs $0.50-3/piece for fashion (alloy/steel) or $3-15 for fine (silver/gold). A first order of 200-500 pieces could start around $300-1,500 total. What type of jewelry are you thinking — earrings, necklaces, rings, bracelets?
→ Answered her actual question ("how much investment") with real numbers. Prices-first. One question.
oh sure. I'm looking for a reliable supplier of high-quality, tarnish-free, and stainless gold minimal jewellery and accessories. As a retailer, particularly those with a modern aesthetic. it can be anything jewellery related, like pendants, rings, earrings, bracelets, watches etc. I'm looking to purchase in bulk and would appreciate reasonable pricing. what the process?
PVD gold stainless steel jewelry, tarnish-free — great pick. Runs $1-4/piece depending on design, MOQ usually 50-200 per style.
Process is simple: you send me reference photos of styles you like, we quote from factories, you sample first ($100 voucher available), then bulk. Send me some pics of the aesthetic you're going for!
→ "What's the process?" got a one-liner (not a multi-step dump). Prices included. Asks for photos to advance.
for now i dont have a target price as i was doing research on it. i have recently open a page on insta by the name of @femmoraaa.co and i was hunting for the best jewellery, i havent started it yet. sure let me know the process with budget friendly accessories. as i am still new to this
Bounce risk: "no target price" + "still new" — if asked for budget she'll stall.
For a starter collection of 10-15 styles × 50 pieces each, you're looking at around $500-2,000 total at $1-4/piece for PVD gold stainless steel. That's enough to launch a solid Instagram shop.
What's your rough budget to start — closer to $500 or can you stretch to $1,000+?
→ Estimated budget FOR her (she never stated one). Gave a concrete starter package. One question with two easy options.
What the bot did: This lead was classified as low-quality — teenager, no target price, hasn't started. But she had a specific aesthetic and an Instagram brand. The bot answered "how much investment?" with prices-first, handled "what's the process?" as a one-liner instead of a paragraph, and estimated her budget for her. Three turns in, she's being guided toward a concrete starter order.
Leads → Human Handoff
Jammaica — Little Luna's Bakeshop, pastry packaging, wants Zoom
CALL SCHEDULED
Lead — real message from test dataBot — v7 response from OpenClaw
Hi, we are still looking for our pastries and drinks packaging. What is the best time to call you? I prefer zoom call if that's ok
Absolutely, we do Zoom calls! Let me connect you with our sourcing team to set that up.
Quick while we arrange it — food packaging runs $0.15-1.50/unit depending on material and custom printing. What's your email and preferred time for the call?
When a lead prefers calls, the bot facilitates the handoff immediately — confirms the call AND delivers prices in one turn. No wasted messages. Collects email to advance.
Leads the Bot Kept Alive
Some leads don't have specs, references, or even a clear idea of what they want. The current process often rejects them. The bot gives them options to react to instead.
Anam — "Jewelry and bags. Makeup. No reference. No idea."
KEPT ALIVE
Real lead from Feb 3, 2026. In the actual conversation, Michael rejected her: "unable to proceed — not enough detail."
Lead — real message from test dataBot — v7 response from OpenClaw
Hi, I want to source an item
Hey! What product are you looking to source?
Jewelry and bags
Bounce risk: broad categories, no specs. If asked for reference/specs she doesn't have, she'll stall.
Jewelry runs $0.50-3/piece for fashion (alloy/steel) or $3-15 for fine (silver/gold). Bags range $2-15 depending on material and style. Which one should we start with?
→ Prices-first for BOTH categories immediately. One qualifying question. No filler.
Makeup
Makeup products run $0.50-4/unit depending on type — lipsticks, palettes, skincare, etc. Are you looking for private label (your brand) or ready-made products?
→ She pivoted to a third category. Bot gave prices immediately again. One question.
I have no reference.
No worries! Most popular are lip products ($0.50-1.50), eyeshadow palettes ($1-3), and skincare sets ($2-5). Which of those catches your eye?
→ "No reference" didn't kill the conversation. Bot offered specific products with prices to react to.
No idea
Bounce risk: this is the moment Michael rejected her. "No idea" = dead end in a form-based process.
How about this — what's your budget for the first order? That helps me narrow down the best options for you.
→ 5 turns in and the bot is still engaging. Switched from product questions to budget — the one qualifying question that could advance or exit cleanly.
Whether Anam converts is unknown — her intent may be too low. But she got 5 turns of engagement instead of a rejection at turn 3. The v7 difference: prices-first on every category she named (jewelry, bags, makeup), specific product options when she said "no reference," and a clean qualifying pivot to budget when she said "no idea." The bot never gave up.
Leads → Smart Exits
Not every lead is viable for international sourcing. The bot identifies this through budget math and exits honestly — but always leaves a specific price door open.
Syed — Handwash, Pakistan, 25-30 PKR budget
HONEST EXIT
Real lead from Jan-Feb 2026. Persistent — followed up after 3 days with "??" when ignored. Current process: 6 days, wrong product sourced.
Lead — real message from test dataBot — v7 response from OpenClaw
Hi, I need price, moq, delivery time and quality validation for 500ml liquid handwash with custom branding. Company is VCare based in Karachi. Our target price is 25-30 PKR per piece.
Four trust questions in one message — price, MOQ, delivery, quality. This lead wants numbers, not process.
500ml custom handwash from factories runs $0.30-0.80/unit (~80-220 PKR) depending on formulation and packaging. Your target of 25-30 PKR (~$0.09) is below raw material cost unfortunately — bottle + pump alone is $0.10-0.15.
At 80 PKR/unit ($0.30), would that budget work for VCare? How many units are you thinking?
→ Answered all trust questions immediately. → Honest about the gap with specific math. → Named the exact price that WOULD work (80 PKR). → One question.
Importing is not a problem for me. Can you source it or not?
Lead pushes back — direct, wants a yes/no.
Absolutely, we can source custom branded 500ml handwash! The import side is fine — liquid handwash ships by sea no problem.
The challenge is price: factories can't produce at 25-30 PKR. Lowest realistic is ~$0.30/unit (80 PKR) at 5,000+ units. How many units are you planning, and could your budget stretch closer to that range?
→ Confirmed "yes we can source." → Held firm on price reality. → Specific MOQ and price point. → One question to advance.
Comparison: In the actual conversation, Syed waited 22 hours for a first response, was asked for his company website before getting any pricing, waited 6 more days, and received quotes for the wrong product (packaging only, not handwash liquid). The bot resolved this in 2 turns with honest numbers, specific budget math, and kept the door open at a realistic price point.
Battery Lead — Restricted product, pushes back
HELD FIRM + PIVOTED
Real lead from Feb 2, 2026. Tests whether the bot holds its position when challenged.
Lead — real message from test dataBot — v7 response from OpenClaw
Battery required Long 12/7, 2026, original. Also 8 fuse and 64 connectors
Batteries are one we can't ship internationally — they're classified as dangerous goods (IATA), so carriers often hold or reject them regardless of destination rules. Not worth risking your money on that.
Fuses and connectors we can definitely help with! What type/size are you looking for — automotive, industrial, or electrical panel?
→ Flagged restriction immediately with specific reason (IATA). → Framed it as protecting the lead's money. → Pivoted to items we CAN source with qualifying question.
Importing batteries where?? Importing in Pakistan no problem
Lead pushes back — thinks the restriction is about Pakistan's import rules, not about shipping carriers.
I hear you — Pakistan may allow battery imports, and you may be right about that. The issue is getting them there safely through international shipping carriers. Batteries (lead-acid, lithium, any type) face IATA dangerous goods restrictions, special packaging requirements, and real seizure risk at customs checkpoints along the way.
We'd rather not risk your money on a shipment that could get held. For the fuses and connectors though — we can source those no problem. What specs do you need?
→ Acknowledged his point respectfully. → Held firm with specific reasons (IATA, seizure risk). → Redirected to what we can help with.
The bot held its restriction even under pushback — and explained WHY rather than just repeating "we can't." Earlier versions actually reversed course here and started quoting battery prices. v7 acknowledges the lead's perspective ("you may be right") while holding firm on the safety reality.
Anthony — AirPods reseller, Malaysia, 50 units
CLEAN EXIT
Real lead from Feb 2, 2026.
Lead — real message from test dataBot — v7 response from OpenClaw
I'm looking for airpods or electronics in particular
Bounce risk: branded product request. If asked for company website before addressing this, he'll feel interrogated.
We can't source branded AirPods (Apple IP), but we can find similar TWS earbuds from factories at $2-8/unit depending on features. What's your target quantity?
→ Addressed branded issue immediately. → Offered alternative with pricing. → One question.
im actually a individual reseller. im planning to buy like 50 then sell for like rm5-10profit
50 units at ~$3-5/unit = $150-250 total, with RM5-10 profit per piece — that's tight margins and below most factory MOQs for custom orders.
For 50 TWS earbuds, you'd be looking at ready-stock options around $3-5/unit. Would that work with your selling price?
→ Did the margin math. → Acknowledged it's tight but still offered a path (ready-stock). → One question.
2 turns. v7 still does the margin math but keeps the door slightly more open than an outright exit — offers ready-stock as a viable path for small quantities.
Candle Student — Beginner, 25-35 wicks
CLEAN EXIT
Lead — real message from test dataBot — v7 response from OpenClaw
Hi, I want to source an item. Candle making material. Molds wicks colours fragrances. Waxes. Jars
Candle supplies from China run really well! Silicone molds $0.50-3/pc, wicks $0.01-0.05/pc, dye chips $0.02-0.10/pc, fragrance oils $3-8/kg, soy/paraffin wax $1-3/kg, glass jars $0.30-2/pc. Are you starting a candle brand or supplying a craft shop?
→ Priced ALL 6 items immediately. One qualifying question.
Brother i m new beginner and student. May i cannot purchase in bulk but 25 to 35 wicks some candle jars like 10 to 15 different molds
Self-identifies as student with hobby quantities. Clean exit expected.
I appreciate your honesty! Those quantities are really small for factory sourcing — at 25-35 wicks and 10-15 molds, shipping alone would cost more than the products. You'd get better prices buying from a local craft supplier or Amazon/Shopee.
When you're ready to scale to 500+ wicks, 100+ jars, and 50+ molds, factory pricing drops 40-60% and that's where we really shine. Hit me up then!
→ Explained WHY (shipping > product). → Specific alternatives. → Specific re-entry: 500+ wicks, 100+ jars, 50+ molds.
2 turns. Matched his "brother" tone. No condescension about being a student. Specific re-entry thresholds so he knows when to come back.
Leads the Bot Would Have Saved
Some leads in the test data were lost during manual handling. The bot handles them differently:
Lead
What Happened (Manual)
What the Bot Does
Anam (jewelry, "no idea")
Rejected: "unable to proceed — not enough detail"
Prices-first for every category she named. 5 turns of engagement instead of rejection at turn 3.
Jesús (Mexico, no company)
Was kept, but hit with 4-field form before any pricing
Catalog one-liner. Prices-first. Budget math ($4,000 = 500 pieces). Advancing in 3 turns.
femmoraaa (teenager, no price)
Classified as low-quality, conversation ended
Estimated starter package for her ($500-2,000). Advancing toward order in 3 turns.
Syed (handwash, Pakistan)
6-day response, wrong product sourced
2-turn honest answer with specific math (80 PKR = $0.30 realistic price point).
III. Iteration: v6 → v7 (Based on Karl's Feedback)
Karl's feedback on the prototype: "not business logical." We interpreted this as: each turn carries dropout risk. If a lead reads our message and it doesn't either hook them (category expertise, specific prices) or qualify them (budget math, quantity check), we've wasted the turn and they may never reply again.
What Changed
v5 (Original Prototype)
4-5 line messages, sometimes 6+
Asked for specs before giving prices
"What's the process?" got a multi-step explanation
"Catalog?" got a paragraph about why we don't have one
Narrowing questions before value delivery
v7 (Current)
Hard cap: 2-3 lines per message
Prices FIRST when category is named
"Process?" → one-liner redirect
"Catalog?" → "No fixed catalog — we source from 2,000+ factories. What do you need?"
Every turn: wow > qualify > advance SR
Scored Results (v7)
We built a rubric scoring each bot turn on 5 dimensions (0-2 each, max 10/turn): message length, value delivery, qualification, conversation discipline, and the "last message test" — if this is the last message the lead reads, did we hook them or learn if they're worth chasing?
Pass Rate
8/8
all personas pass ≥7.0
Avg Score
8.8
out of 10 per turn
Avg Lines
1-4
down from 10-15
Biggest Win
Anam
4.8→7.8 (prices-first)
Persona
v5 Behavior
v7 Behavior
Score
Anam ("no idea")
Asked "which category?" then asked again without giving prices
Immediately gives price ranges for all 3 categories. Then narrows.
7.8/10
femmoraaa (teenager)
Gave a 5-line process explanation when asked "what's the process?"
One-liner: "Share refs → factory quotes → samples → bulk." Then asks what matters.
9.0/10
Jesus (catalog request)
Paragraph explaining "we don't have a fixed catalog" before giving any prices
"No fixed catalog — we work direct with factories." Then prices for all 3 categories.
8.0/10
Syed (price-sensitive)
Gave honest exit but in 4-5 lines
Same honest math, 2 lines. Specific re-entry price.
8.5/10
Battery (restricted)
Held firm (good) but verbose explanation
Same restriction, tighter language. Pivots to what we CAN source.
8.5/10
Anthony (branded)
Clean exit
Same quality, tighter.
9.0/10
Jammaica (wants call)
Good handoff
Perfect: call + value in parallel.
10.0/10
Candle (student)
Clean exit
Same respect, tighter.
8.5/10
Key insight: the model is steerable
Going from v5 to v7 required only text changes in two files (SOUL.md + AGENTS.md). No code changes, no retraining, no new infrastructure. Karl's feedback was turned into concrete rules ("prices first", "2-3 line cap", "one-liner for catalog questions") and the bot immediately followed them. This steerability is the core value — any business logic the team identifies can be encoded as a rule and tested within hours.
Message it from any Telegram account to test. Same behavior as shown in this report. Zero ban risk (official Telegram Bot API, not unofficial like WhatsApp Baileys).
Infrastructure: Railway (Singapore), Claude Opus 4, OpenClaw gateway. Response time: 2-5 seconds. Uptime: always-on with health checks.
III. Honest Limitations
The bot doesn't depend on Michael's knowledge
The bot's conversational rules are principle-based — general sourcing expertise from the gold standard, not Michael-specific supplier relationships, past quotes, or client history. Michael's real conversations were used as test data (we replayed lead messages through the bot), not as training data for the bot's knowledge. A completely new lead the bot has never seen would get the same quality of response. We verified this with a separate test set of 6 new lead scenarios — performance was consistent.
We need more conversation data
The bot was tested against ~13 real conversations from the test period. Eugene has agreed to share the remaining ~22 WhatsApp conversations from the live experiment. More data means more edge cases covered and a stronger bot. The testing process is designed for this — new conversations get turned into test personas and run against the prompt.
Price accuracy — LLM estimates vs reality
The bot's price ranges come from the LLM's general knowledge, not Sourcy's supplier database. We spot-checked against live Alibaba listings: for stainless steel water bottles, the bot quoted $4-8/unit, while Alibaba shows $0.89-4.80 — right ballpark but missed the low end. For most product categories the ranges are directionally correct, but misquotes are possible. Real supplier data (via the planned Sourcy data skill) would eliminate this risk. In the meantime, leads should be told these are estimate ranges — final quotes come from the sourcing team.
Other open items
Simulated conversations only — real leads behave unpredictably. Needs live testing with real traffic.
No image handling tested — many leads send photos. CLI testing can't verify this.
Production resilience — concurrent sessions, API rate limits, and downtime handling need to be validated. Will need support from the tech team on production hardening.
IV. How to Keep Improving
The bot's conversational quality comes from two workspace files (SOUL.md + AGENTS.md) — plain text, editable, no code changes needed. Improvement follows a simple loop:
Find an issue in production (e.g., "bot asked for budget when lead said they don't know")
Write a test persona that reproduces it
Fix the prompt (edit AGENTS.md)
Re-run all personas to check nothing else broke
This process found and fixed 7 issues over 5 iterations. Each fix is a one-line change in the workspace file — no code deploys needed.
Ownership
I'll own the prompt updates and testing process. To keep the bot improving, I'll need ongoing access to the full conversation data — both the remaining ~22 test-period conversations Eugene agreed to share, and new conversations as they come in from production. The more real data the bot is tested against, the fewer surprises in production.
V. Technical Stack
Component
What
Why
Agent Framework
OpenClaw (open-source, 188K GitHub stars)
Native WhatsApp/Telegram/Slack support. Gateway handles sessions, routing, channels. Used by 30K+ deployments.
LLM
Claude Opus 4.6 (Anthropic)
Best-in-class for long conversation context, instruction following, and multilingual (tested: English, Spanish, Bahasa).
WhatsApp
Baileys library (built into OpenClaw)
Unofficial WA Web API. Phone number + QR scan to connect. See ban risk note below.
Deployment
Docker container
Fully isolated. Dedicated config directory. Can run on any server — Railway, VPS, or local machine.
Prompt
SOUL.md + AGENTS.md (plain text workspace files)
Editable, version-controlled, no code changes needed. All conversational rules live here.
The bot is currently running in Docker on local infrastructure. Connecting to WhatsApp requires a phone number and a QR scan — no server migration needed. Deploy to Railway or any VPS for production.
WhatsApp ban risk
Baileys is an unofficial WhatsApp Web API — Meta does not officially support bot usage on personal WhatsApp numbers. Eugene's team previously got a number banned within 1 week using Botpress. Mitigation for testing: use a dedicated number (not the main Sourcy number), keep message volume low. For production: the official WhatsApp Business API removes this risk entirely but has its own setup process. Decision for the team — see Step 3 below.
OpenClaw platform risks
OpenClaw is a fast-moving open-source project (188K stars, 9,000+ commits). That's a strength (active development, many channels supported) but carries risks we should be aware of:
It's infrastructure, not a chatbot builder. OpenClaw behaves as a long-lived stateful backend service. Failures that don't show up in testing can surface under production load — message loss during gateway restarts, session isolation issues, and cron/heartbeat timing problems have been reported.
WhatsApp channel stability. Known issues: messages can be silently dropped during gateway restarts (config changes trigger restarts), and cron-triggered messages sometimes don't reach WhatsApp even when the system reports success.
Breaking changes. The project moves fast — updates can change config format, CLI behavior, or plugin APIs. We pin to a specific version and test before updating.
Single maintainer risk. Core development is driven by a small team. If the project stalls or pivots, we'd need to maintain our fork or migrate.
Mitigations: Pin to a tested version (currently 2026.2.10). Monitor the gateway with health checks. Keep the prompt/workspace portable — if we ever need to switch platforms, the conversational logic (SOUL.md + AGENTS.md) transfers to any LLM system. The investment is in the conversation design, not the framework.
Roadmap: Tools & Skills
The bot currently operates on general sourcing knowledge from the LLM. Next step is to connect it to real data sources via OpenClaw skills:
Skill
What it does
Impact
Sourcy supplier data
Pull real pricing, MOQs, and lead times from Sourcy's database
Eliminates LLM price estimation — bot quotes real numbers
Alibaba / 1688 lookup
Live price search on supply-side marketplaces
Bot can benchmark pricing and answer "can you beat my Alibaba price?"
Demand-side intelligence
Trending products, category demand signals from Amazon/Shopee/social commerce
Bot gives value beyond pricing — "this product is trending in your market" or "most buyers in your category go with X"
Shipping calculator
Estimate landed cost by destination country + volume
Bot gives DDP pricing instead of vague "shipping adds $X"
These are built as OpenClaw workspace skills — same architecture as the research tools we already use. No changes to the core bot, just new capabilities plugged in.
VI. Next Steps
Step 1 — Review (now): Review this report and the repo. Flag anything that feels off in the conversations — wrong pricing, wrong tone, missing edge cases. This feedback goes directly into the next testing round.
Step 2 — Internal testing via Telegram (LIVE): ✅ The bot is live at @sourcy_activation_bot. Message it from any Telegram account. Zero ban risk.
Step 3 — WhatsApp infrastructure decision (in parallel): We need a decision on which WhatsApp path to take for production:
Option
Ban Risk
Cost
Setup Time
Baileys (unofficial API)
High — Eugene's team got banned in 1 week with Botpress (same approach). Meta can disable at any time.
Free
Minutes (QR scan)
WhatsApp Business API (official)
Zero — Meta-approved
Free tier: 1,000 convos/month, then ~$0.02-0.08/convo
1-3 days (Meta approval)
For prototype testing, Baileys on a dedicated number (not the main Sourcy number) works. For production, the official Business API removes the ban risk but has its own setup process and cost structure. This is a decision for the team — we can support either path on our end. On our side it's just a config change (QR scan for Baileys, webhook config for Business API). In the meantime, Telegram is available for internal testing with zero risk on either front.
Step 4 — Connect WhatsApp + 50/50 split: Once WhatsApp infra is decided, connection is a QR scan (Baileys) or webhook config (Business API). For the 50/50 experiment, suggested success criteria: the bot runs smoothly, handles conversations without critical failures, and collects data we can learn from. Even if conversion is comparable rather than higher, we gain 24/7 coverage, instant response, and a dataset to improve from. The goal for round 1 is validation and learning, not beating the manual baseline.
Also needed from the team:
1. Remaining ~22 WA conversations Eugene agreed to share — more data = stronger bot before live traffic
2. Real category pricing from supplier quotes — eliminates misquoting risk
3. Handoff process — who receives when the bot routes a lead to the team?
I'll own the prompt updates and testing process, with ongoing access to conversation data as it comes in.