E-commerce
July 1, 2026
"How much should I order for a month?" "6-pack or single unit, which is cheaper?" "There are 4 of us at home, which family pack should I choose?" Three pre-purchase questions that the product page often leaves without a clear answer. The customer hesitates, picks the entry-level option, or orders too much and triggers a post-delivery multipack customer support ticket (#390).
An e-commerce quantity choice AI chatbot recommends the right size (single, pack, family size, subscription) based on usage, household, and budget, without aggressive upselling or invented stock promises. It increases useful AOV and reduces pack misunderstanding errors.
This guide #391 covers intents bot_qty_*, flow QSELECT-FLOW, and KPI qty_bot. It differs from the multipack customer support (#390) (post-purchase disputes) and the pack choice guide (#97) (bundle promo merchandising): here, pre-purchase optimal quantity, pack, subscription, and family pack AI use cases.
Summary
Why deploy an AI chatbot for quantity selection?
The pre-purchase quantity choice blocks conversion and AOV on consumables, supplements, and DTC food. Baymard notes that poorly explained variants and ambiguous quantities are among the major PDP frictions (Baymard, PDP UX 2025).
Four challenges without a quantity bot
Under-buying: client takes one unit, low AOV, late repeat purchase
Over-buying: bundle too large, regret, post-delivery ticket #390
Missed subscription: client does a one-shot purchase when a -15% subscription was suitable
Paralysis: 4 quantity variants, cart abandonment
GoPrecision reminds us that bundling and bulk formats increase AOV when they reduce decisions, not when they pile up random SKUs (GoPrecision, bulk psychology 2026).
Angle #391 vs related content
#390 MULTIPACK CS: post-purchase bundle disputes. #391 = pre-purchase prevention via bot.
#97 pack choice: promo bundle merchandising. #391 = unit/bundle/subscription/family bot quantity logic.
#3 product questions: compatibility, size, usage. #391 zooms in on quantity and bulk format.
#309 repeat purchase: REORDER outside of subscription. #391 = 1st purchase optimal format.
DTC Example
Supplement brand, 68% of PDP visitors hesitate between single unit and pack of 3. After QSELECT-FLOW bot: qty_bot_attach_rate bundle/subscription +24 pts, multipack_ticket_rate -31% (correlation #390), bot segment AOV +18%.
Responsible recommendation principle
Always offer "unit is enough if..." before "ideal bundle if...". No forced minimum bundle. Recommended variant add-to-cart link, not a manipulated cart quantity.
How does it differ from multipack #390 and guide pack #97?
Five quantity/volume content pieces, five distinct funnel moments.
Content → moment matrix
#391 QSELECT bot: pre-purchase, recommends qty/format
#97 pack guide: comparative PDP UX, merch quiz
#390 MULTIPACK SAV: post-delivery, comprehension dispute
#309 REORDER: existing customer, repeat purchase cycle
#373 SAVE sub: retention cancel flow, not initial choice
#391 vs #97
#97 = Good/Better/Best table, copy naming, promo bundle attach rate. #391 = 3-5 question conversation → recommended qty variant grounded in metafields.
#391 vs #390
QSELECT bot explains "pack of 6 = 6 units, cart qty 1" before purchase. #390 resolves "I ordered 1, received 6" after the fact. Prevention + resolution pipeline.
Subscription vs one-shot pack
Bot compares: pack of 3 one-shot 90-day usage vs monthly sub 1 unit -15%. SubSummit: Subscribe & Save for predictable profiles, one-shot for testing (SubSummit, Replenishment 2026).
Family size
"Family Size 1 L" variant vs "Standard 250 ml": bot calculates number of people × daily consumption. Distinct from heterogeneous bundle #97.
Complementarity
Merch implements table #97. Bot #391 personalizes recommendation. Support #390 handles remaining confusion.
Which bot_qty_* intents should the chatbot detect?
Twelve bot quantity choice intents cover pre-purchase volume hesitations.
Twelve bot_qty intents
bot_qty_unit_vs_lot: "individual or set, which one to choose?"
bot_qty_lot_savings: "is the set cheaper?" unit price
bot_qty_how_many_month: "how many for 1 month?"
bot_qty_household_size: "there are 4 of us, which size?" family size
bot_qty_subscription_vs_onetime: "subscription or one-time purchase?"
bot_qty_sub_frequency: "every 4 or 8 weeks?"
bot_qty_cart_qty_help: "qty 2 pack 6 = how many?" prevention #390
bot_qty_trial_first: "I'm testing, start small?"
bot_qty_bulk_stock_up: "stock up Black Friday deal"
bot_qty_gift_quantity: "gift, what size to give?"
bot_qty_compare_formats: "difference between 250 ml and 1 L?"
bot_qty_perishable_limit: "pack of 12 before expiration?"
Trigger signals
Words: how many, quantity, set, pack, size, family, subscription, month
Idle variant selector 45s without add-to-cart
Toggle individual ↔ set 2+ times same session
PDP Widget "Which quantity?" click
Conversation tags
qty_bot_detected, qty_bot_recommended_lot, qty_bot_recommended_sub, qty_bot_recommended_unit, qty_bot_cart_explain, qty_bot_handoff. Sync analytics attach rate per intent.
Intent priority
bot_qty_cart_qty_help (prevention #390) > bot_qty_unit_vs_lot > bot_qty_household_size > bot_qty_subscription_vs_onetime. Active cart: read cart before recommending.
Which QSELECT-FLOW conversational flow should be configured?
The QSELECT-FLOW flow guides the quantity choice in 7 to 8 steps, QB-1 to QB-8.
Base sequence (bot_qty_unit_vs_lot or bot_qty_how_many_month)
QB-1 Home: "I'll help you choose the right format. 3 quick questions."
QB-2 Usage: solo, couple, family of N people, professional
QB-3 Duration: 2-week trial, 1 month, 3 months stock
QB-4 Constraint: tight budget, storage space, short expiry date
QB-5 Recommendation calculation: read metafields consumption_days, units_per_pack
QB-6 Present options: single unit | pack of N | subscription | family pack with unit price
QB-7 Add-to-cart link: recommended variant + alternative "smaller if..."
QB-8 Close: "Need checkout help?" handoff if hesitated for 3+ messages
bot_qty_trial_first branch
Recommend single unit or starter kit. "Start with 1 unit for 30 days. If it works, a pack of 3 saves 12%." Link to starter kit #308 if applicable.
bot_qty_subscription_vs_onetime branch
If regular usage > 60 days and predictable profile: present subscription -X% + suggested frequency. Otherwise, one-shot pack. Never subscription by default without QB-3 duration confirmed.
bot_qty_cart_qty_help branch
Table: variant | cart qty | units/pack | total units | price. Aligned with MULTIPACK-EDU #390 as a preventative measure. "You have 2× pack of 6 = 12 units. Is this your intention?"
bot_qty_perishable_limit branch
Read shelf_life_days metafield. Refuse to recommend pack of 12 if consumption < shelf life. See perishables.
Flow restrictions
Do not modify cart qty API without consent. Do not say "only the pack of 12 is left" if single unit stock is OK.
What product data should be connected to make recommendations?
The quantity choice bot database combines the Shopify catalog and usage metafields.
Required metafields per variant qty
units_per_pack: 1, 3, 6, 12 (sync #390)
unit_price_display: calculated pack ÷ units
consumption_days_per_unit: usage duration of 1 unit for a standard profile
format_label: unit | pack | family | sub
shelf_life_days: if perishable
subscription_eligible: bool + discount_pct Recharge
Shopify Storefront MCP lookup_catalog and get_product provide structured variants to the bot (Shopify Dev, Storefront MCP).
Recommendation formula QB-5
units_needed = (household_size × consumption_days_per_unit × duration_days) / consumption_days_per_unit. Round up to the next variant (pack of 3 if 2.5 is needed). Display transparent calculation to the customer.
Numerical example
Protein: 1 scoop/day, 30-day tub. Couple (×2) for 1 month = 2 tubs. Recommend pack of 2 or sub 2/month based on QB-4 budget.
Logged-in customer history
If 2+ orders of the same SKU: suggest usual qty + pack option if not yet tried. See purchase history bot.
Source of truth rule
No invented consumption_days. If metafield is missing: QB-4 questions only, conservative unit recommendation.
How to present bulk packages, subscriptions, and family sizes without pressure?
The quantity recommendation bot balances AOV and trust, without dark patterns.
QB-6 message structure
Recommended: variant + reason 1 sentence ("2 people, 1 month")
Savings: "€4.99/unit vs €5.49 single unit" if bundle
Prudent alternative: "Single unit is enough if you are testing"
CTA: add-to-cart link for recommended variant_id
Bundle vs subscription
Bot comparative table:
Bundle of 3 one-shot: €89.97, €29.99/u, stock at your place.
Sub 1/month -15%: €25.49/u, auto delivery, pause possible.
Recommend subscription if QB-3 duration > 90 days and regular use.
Family format
"Family 1 L = 4×250 ml. For 4 people 2 weeks: 1 family size vs 2 standard sizes." Price/unit/ml always visible. Fudge: absolute savings amount anchors better than % alone (Fudge, pricing volume 2026).
Anti-patterns
Pre-select bundle of 12 by default on checkout
Bot says "best deal" without single unit alternative
Sub pre-checked without explicit QB
Ignore expiration of bulk bundle
Credibility before upsell
Oscar Chat: short conversion messages, direct link to variant (Oscar Chat, conversion 2026). Max 4 bubbles before CTA.
How to route subscription, batch and unit according to the profile?
The bot quantity routing matrix maps customer profile → optimal format.
Four QB-2 profiles
Tester: unit or starter → bot_qty_trial_first
Solo regular: 1-3 month pack or monthly subscription
3+ household: family size or 6+ pack depending on consumption
Promo stock: max pack if shelf_life OK + QB-4 confirmed
Subscription decision tree
Usage > 60 days + same SKU 2× in history → offer subscription
First purchase → pack or unit, no subscription push
Tight budget QB-4 → unit or long frequency subscription (low cost/month)
Gift → unit/medium pack, no subscription (gift order)
Recharge integration
Bot read plans: 4 weeks, 8 weeks, 12 weeks. QB-6 subscribe & save variant link. Handoff if existing subscription modification → subscription support.
Wholesale redirect
bot_qty_bulk_stock_up qty > retail threshold (e.g., 24 units): B2B qualification, no consumer pack.
Agent handoff
Cart > €200, complex allergy diet, or 4 hesitation messages: pre-filled ticket profile QB-2 to QB-5.
Which triggers and widget placements for the quantity bot?
The quantity choice bot deployment targets multi-variant PDPs, volume, and cart.
Five QSELECT Triggers
T1 Incoming message: keywords qty, pack, how many, subscription
T2 Variant toggle idle: 2+ unit/pack changes in 30 s
T3 PDP widget: "Which quantity for me?" under selector
T4 Multipack cart: pack line item detected, qty × units hint
T5 Consumables collection: replenish category, discreet bot hint
Priority Pages
SKU 3+ qty variants (unit, pack of 3, pack of 6, family pack)
Supplements, food, hygiene, coffee, pets
PDP with visible Recharge subscription toggle
Proactivity Cap
Max 1 T3 popup/session. T4 only if cart contains a pack without interaction for 20 s. No payment checkout popup (#386).
Mobile
QB-2 quick replies: Solo / Couple / Family 3-4 / Family 5+. Short responsive qty table.
A/B Test
T3 widget vs T1 passive. Measure qty_bot_attach_rate and multipack_ticket_rate over 60 days.
Which qty_bot KPIs should be measured each month?
The bot quantity selection KPIs link recommendation, AOV, and #390 tickets.
Eight key metrics
qty_bot_session_rate: qty bot sessions / multi-variant PDP sessions
qty_bot_attach_rate: post-recommendation add-to-cart / bot sessions
qty_bot_lot_uplift: % pack+family vs PDP baseline without bot
qty_bot_sub_conversion: sub activated post-bot / bot_qty_sub intents
qty_bot_resolution_rate: resolved without handoff / interactions
qty_bot_cart_explain_saves: pre-checkout cart qty corrections
multipack_ticket_correlation: decrease in #390 post-QSELECT
qty_bot_csat: satisfaction post-recommendation
DTC Benchmark
qty_bot_attach_rate 35-48 %, lot_uplift +15-25 pts bot segment, resolution 70-82 %, sub_conversion 20-35 % sub intents, multipack_ticket -25-35 % at 90 days.
Weekly CRO + Support Dashboard
Top bot_qty_* intents, attach by QB-2 profile, bot vs non-bot AOV, correlation of multipack tickets.
Audit Transcripts: 20/month
Score: single unit alternative proposed? transparent calculation? no fake stock? correct variant link?
ROI
8 % PDP sessions × 40 % attach × delta AOV pack +12 € × 45 % margin. Minus tickets #390 × agent cost 8 min.
What edge cases and safeguards should be planned for?
Ten bot quantity edge cases and playbook responses.
1. Recommend pack of 12 perishable product
Guardrail shelf_life. bot_qty_perishable_limit cap calculated max pack.
2. Child / medical dosage
No bot dosage calculation. Handoff + link to leaflet. Regulated products.
3. Family multi-product allergy
Handoff. Bot does not recommend a pack without a grounded ingredients list.
4. Customer wants qty 10 pack of 6 (= 60 units)
bot_qty_cart_qty_help table + confirm. Wholesale if > threshold.
5. Existing sub, wants to change qty
Redirect to Recharge portal, no new sub via PDP bot.
6. "Buy 3 get 2 free" promo
Distinct QSELECT. Bot explains merch promo, no standard QB-5 recalculation.
7. LLM hallucinates consumption_days
QB-6 templates with metafields only. Unit fallback if data is missing.
8. Gift, unknown recipient
Recommend standard medium size, not family XL.
9. Pack stock depleted, unit OK
Recommend available unit. Do not push OOS pack.
10. Customer regret post-bot (ticket #390)
Log conversation_id order note. Support MULTIPACK-FLOW with QB-6 bot context.
Link #392 future
Bulk-size-savings article (#392) will elaborate on the size savings argument. #391 bot applies conversational logic.
How does Qstomy recommend the right quantity?
Qstomy on Shopify: QSELECT-FLOW with intents bot_qty_*, metafields units_per_pack and consumption_days, batch/sub/family comparison, direct variant link.
QSELECT Qstomy Capabilities
qty_intent_detect: 12 intents section 3
qty_variant_read: variants, unit_price, stock API
qty_calculate_qb5: household × duration → units_needed
qty_sub_compare: Subscription plans vs one-shot batch
qty_cart_explain: qty × units_per_pack pre-checkout (#390)
qty_recommend_cta: add-to-cart variant + unit alternative
qty_handoff_context: profile QB-2 to QB-5 to agent
Complement #390 + #97
#97 PDP table merch. Qstomy QSELECT customizes. #390 resolves residual disputes with bot log if present.
Quantified DTC Scenario
Food brand, 4 SKUs batch of 3/6/family, 72% hesitation mobile variant selector.
After Qstomy QSELECT: qty_bot_attach_rate 43%, qty_bot_lot_uplift +21 pts, multipack_ticket_rate -33%, qty_bot_sub_conversion 28%, AOV bot segment +16 €.
Explore AI sales agent, customer support and request a demo.
What is the checklist for deploying QSELECT-FLOW?
QSELECT Checklist (12 steps)
Audit variant quantities: metafields units_per_pack, consumption_days, format_label
Align PDP #97 unit/pack/family/subscription table
Configure 12 bot_qty_* intents in section 3
Draft flows QB-1 to QB-8 + QB-6 templates approved by CRO
Connect Recharge plans if subscription is active
Activate bot_qty_cart_qty_help cart (prevention #390)
Expiration and regulated products guardrails
Triggers T1-T5 + mobile proactivity limit
Staging tests for 8 profiles (tester, family of 4, subscription, gift, bulk)
Weekly qty_bot KPI Dashboard
Audit 20 transcripts/month: unit alternative, no dark patterns
Correlate multipack_ticket_rate #390 quarterly
At a glance
#391 = pre-purchase quantity choice bot, #390 = post-purchase support
QSELECT-FLOW: profile → calculation → pack/subscription/family + unit alternative
Responsible recommendation: credibility before pack upsell
Grounded Metafields: no invented consumption_days
KPI qty_bot_attach_rate: target 35-48%
FAQ
Difference with #390 multipack?
#391 helps choose before purchasing. #390 resolves "I misunderstood the pack" disputes after delivery.
Difference with #97 pack guide?
#97 = PDP merchandising UX. #391 = personalized conversational bot for qty/subscription/family.
Can the bot force pack 12?
No. Recommendation + unit alternative. Customer chooses variant via CTA link.
Relation to subscription #373?
#373 = save cancel flow. #391 = initial subscription vs. one-shot choice pre-purchase.
Cart Qty 2 pack 6?
bot_qty_cart_qty_help: 12 units. Confirm intent before checkout (#390 prevention).
Go further
Test staging: "couple, 1 month" profile on SKU pack 3/6, verify calculated recommendation, correct variant link, unit alternative mentioned.
Share this guide #391 with CRO and support: the right quantity bot sells the suited format, not the largest pack by default.
Enzo
July 1, 2026
