E-commerce
June 28, 2026
"Compatible with my washing machine?" "Does it run big or small?" "Is the cable included?" These questions keep coming up in the chat, the bot, and tickets. Your agents answer them well. Your product page, however, remains silent.
Ecommerce Circle observes that a system structure of five PDP blocks (hook, benefits, contextualized specs, objections, reassurance) can increase conversion by 25 to 40% when the "Objection Crusher" block addresses real questions from support (Ecommerce Circle, PDP framework 2026). Gorgias points out that about 1 in 9 tickets is a pre-purchase question, treated as a cost when it is actually a conversion signal (Gorgias, pre-purchase 2025).
This guide #211 covers the detection of customer questions to transform into PDP blocks: signals, scoring, module typology, Shopify implementation. Distinct from product descriptions (#189) (textual copy) and conversations → product page (#75) (global method): here, we focus on which module to add where on the PDP.
Summary
Why should certain customer questions become PDP blocks?
Detecting customer questions to turn into PDP blocks avoids two pitfalls: piling text into the description (unreadable) or leaving the info only in support (invisible to 95% of visitors).
Block vs. paragraph
A PDP block is a scannable module placed in a specific location: quick answers banner below the price, compatibility table, objections accordion, dimensions insert. Lasso points out that fields above the fold reduce uncertainty before the cart click (Lasso, PDP fields 2026).
Why support is a goldmine
Customers express their doubts in their own words. "Does this fit in an EasyJet cabin?" is better than a marketing paragraph saying "compact and practical". Talk Shop recommends separating pre-purchase questions (on the PDP) and post-purchase questions (help page) (Talk Shop, Shopify help page 2026).
DTC Example
Furniture brand: 47 tickets/quarter "does it fit on a 90 cm balcony?". Rich description had no effect. "Dimensions and placement" block with diagram + balcony range: tickets −81%, add-to-cart +16%.
How does it differ from the neighboring guides on the product sheet?
Five neighboring contents, five different outputs.
Conversations → sheet (#75)
Conversations (#75): PDP global loop (copy, visuals, specs). #211 specifies which question becomes which type of block.
Product descriptions (#189)
Descriptions (#189): rewrite descriptive text. #211 covers structured modules (accordion, table, banner).
Support content gaps
Content gaps: cross-functional hub and policy diagnosis. #211: action by SKU on the PDP.
Misunderstood products (#109)
Confusing products (#109): confusion score analytics. #211: content remedy once the SKU is prioritized.
Structure by intent (#165)
Intent (#165): organize a product help page. #211: in-page blocks linked to the moment of purchase.
Promise #211
Detection signals, scoring, block typology, writing from verbatim, Shopify metafields, bot sync, 30-day playbooks.
What signals indicate that a question deserves a block on a product sheet?
Not all questions deserve a block. Target reproducible question → PDP block signals.
Recurrence Threshold
≥ 3 times / 90 days on the same SKU = block candidate (Ecommerce Circle)
≥ 5% of PDP chats on an intent = high priority
≥ 2 reviews of 1-3 stars mentioning the same misunderstanding = reassurance block
Qualitative Signals
Post-chat abandonment: customer asks a question, does not order
"I thought that" formulation: gap between perception vs. product sheet
Mobile rage scroll: long PDP session, no add-to-cart (Hotjar + chat intent)
"Not as described" return: spec missing from the PDP
Out of Scope for PDP Block
WISMO, parcel tracking, refund dispute, account issue: these remain support or post-purchase help page topics. Do not clutter the PDP with post-sales service.
How do you mine chat, tickets, and reviews to detect these questions?
The PDP customer questions mining crosses four sources over a minimum of 90 days.
Source 1: chat and bot
Export intent + product URL + verbatim. Filter /products/* pages. Tag pre_purchase. Qstomy, Gorgias or Zendesk: group by Shopify product_id.
Source 2: helpdesk tickets
Query: tickets with SKU, open status before order or tag product_question. Grep subjects "compatible", "size", "included", "material".
Source 3: reviews and returns
1-3 star reviews: extract negative phrases related to an unmet expectation. Shopify returns reason "incorrect description" or "item not as described".
Source 4: internal search and unmatched bot
Site search bar + bot logs no_answer on product page. Nexu recommends using unresolved chatbot logs as PDP backlog (Nexu, logs → PDP 2026).
Notion collection format
Columns: customer verbatim | SKU | source | count 90 d | intent tag | status (candidate / in progress / live). Goal: 200 raw rows → 30 clusters.
What type of PDP blocks for which types of questions?
The PDP client questions block typology maps each cluster to a specific UX module.
Correspondence table
Device / part compatibility → checkmarks table (Yes/No/Partial) + tested models
Size / fit → size guide + "the model is X tall, wears M" + returns link
Pack contents → "Included / Not included" bulleted list with flat lay photo
Dimensions / weight → annotated diagram + comparison with a common object (2-seater sofa)
Care / usage → washing icons + 2 lines per scenario (machine, oven, outdoor)
Competitor comparison → honest "vs [Brand X]" block (Surfient: AI comparison Q = strong lift)
Price / value objection → "Why this price" insert (materials, warranty, manufacturing)
Quick answers banner → 4-6 top questions below the price (Idukki, conversational PDP)
Placement rule
Ecom Design Pro: Q&A near the buy box, not in the footer (Ecom Design Pro, Q&A UX 2026). Compatibility and size: above the CTA. Care and comparison: under the gallery or in the "Product FAQ" accordion.
How to score and prioritize questions to convert into blocks?
The PDP block prioritization scoring avoids redesigning 400 SKUs in parallel.
Impact × Volume Formula
Score = (questions 90 d × 2) + (PDP SKU sessions / 1000) + (PDP bounce rate × 10) + (unit margin / 50). Go threshold: score ≥ 25.
Effort / Impact Matrix
Quick win: 3-Q&A accordion from existing tickets, 2 hrs/SKU
Medium: 10-line compatibility table, 4 hrs + product validation
Heavy: custom dimension diagram, photoshoot, 1-2 days
Batch Prioritization
Sprint 1: top 5 SKUs by score (often high-traffic best-sellers). Sprint 2: SKUs with bounce > 70% and a clear cluster. Sprint 3: new collection arrivals (shadow mode bot for 2 weeks → real questions before block launch).
Kill Criteria
Question < 2 occurrences over 90 d after description enrichment: no dedicated block, macro bot answer is sufficient. Question resolved by variant selector (visible color): UX fix, not text block.
How to write a PDP block based on customer verbatim?
Writing a PDP block from customer verbatim requires three copywriting rules.
Rule 1: question = customer title
Keep the actual wording. "Compatible with 7 kg washing machine?" beats "Appliance compatibility". Long-tail SEO + instant recognition.
Rule 2: response in 2 steps
Sentence 1: direct yes/no/range answer. Sentence 2: nuance or condition. Ecom Design Pro: an opening sentence in plain language, followed by a spec detail (Ecom Design Pro, Q&A UX).
Rule 3: max length
Accordion: 40-80 words per response. Quick answers banner: 25 words max. Compatibility table: status column + 10-word note.
Example before / after
Verbatim: "I have a MacBook Air M2 2022, does it charge via USB-C PD?" Block: Q "Compatible with MacBook Air M2 (2022)?" A "Yes, USB-C Power Delivery charging up to 65W. USB-C cable included. Not compatible with MagSafe-only."
Internal validation
Merchandising validates claims. Ops validates measured specs. Support validates that the answer replaces the most frequently used macro.
How to implement blocks on Shopify without a complete redesign?
Implement Shopify question blocks without a complete theme redesign.
Option A: JSON metafields (recommended for OS 2.0)
Settings → Custom data → Products → metafield custom.pdp_blocks type JSON. Structure: array of {type, question, answer, order}. Liquid snippet pdp-question-blocks.liquid renders accordion + QAPage schema from the same source (Surfient: one source, visible + JSON-LD) (Surfient, Shopify 2026 schema).
Option B: Reusable Q&A metaobjects
Metaobject "Product Q&A Entry" linked to products. Common questions (30-day return) as global entries. SKU-specific questions as local entries. Talk Shop details the metaobjects flow (Talk Shop, Shopify product Q&A).
Option C: Theme editor sections
Collapsible content blocks on product template. Fast for 5-10 hero SKUs. Not scalable to 500 SKUs without metafields.
QAPage Markup
5-8 Q&As per top-traffic SKU. Questions = exact match visible text. Only one QAPage block per URL. Include honest comparison if recurring cluster (Surfient: AI conversion lift +48% among panel of 120 merchants).
How to synchronize PDP blocks, bots, and support macros?
A PDP block without sync support recreates inconsistency within 30 days.
Weekly sync flow
Block published in Shopify → metafield export → updated RAG bot chunk
Gorgias macro aligned with the same wording (PDP- prefix)
Bot intent
product_[sku]_[topic]points to the block, not to LLM improvisation
Proactive widget
If "size" cluster > 10% of PDP chats: bot nudge "Size guide available on this page" instead of opening a ticket. See proactive messages.
Post-block measurement
D+14: compare SKU pre-purchase tickets, PDP bounce, add-to-cart rate. D+30: CSAT on pre_purchase segment. Goal: −40% repeated questions on covered intent.
Versioning
Date block_reviewed_at in metafield. Alert if > 90 days without review or if product policy changes (new collection, reformulation).
Which placement or formatting mistakes kill the impact?
Five PDP block anti-patterns that undo the mining effort.
Error 1: footer-only block
Nobody scrolls. Fix: quick answers under the price, accordion before reviews.
Error 2: copying the generic help page
"3-5 day delivery" on every SKU. Fix: product spec content only.
Error 3: 15 accordion questions
Wall of text. Fix: max 5-6 visible Q&A, rest on the help page.
Error 4: schema without visible content
Google penalizes a hidden Q&A accordion. Fix: same text rendered on screen and in JSON-LD.
Error 5: mining once a year
New collection = new questions. Fix: bi-weekly support + e-commerce ritual (see #189 playbook 5).
How does Qstomy detect questions to transform into blocks?
Qstomy automates the detection of questions that should live on the PDP, not just in the chat.
Detection → block features
SKU intent clustering: groups PDP verbatims, exports top clusters
pre_purchase Flag: filters WISMO and litigation
Block type suggestion: compat, sizing, pack_content according to intent
Priority score: volume + product URL bounce (via analytics integration)
Metafield-ready export: Q&A JSON for Shopify import
Post-block tracking: reduction in repeated questions on covered intent
Quantified DTC Scenario
Electronics brand, 340 active SKUs, 18% PDP chats without self-service response (unmatched bot). Qstomy 90-day mining: 42 clusters, top 8 SKUs = 63% of question volume. Deployment of 8 blocks (5 accordions + 3 compat tables) via metafields. After 10 weeks: pre-purchase tickets for top 8 SKUs −54%, average add-to-cart +19%, top SKU PDP bounce −12 pts (71 → 59), unmatched PDP bot −47%.
Explore AI customer support, Shopify integration, request a demo.
Which operational playbooks should be launched in 30 days?
Playbook 1: mining 90 d (4 h)
Export chat, tickets, 1-3 star reviews. Filter /products/. Notion 200 lines min. Top 10 clusters by SKU.
Playbook 2: scoring + batch 1 (2 h)
Calculate section 6 score. Select 5 SKUs. Assign section 5 block type by cluster.
Playbook 3: Q&A writing (1 d)
5 SKUs × 4 Q&As max. Verbatim → title. Merchandising + support validation. JSON metafield export.
Playbook 4: Shopify implementation (1 d)
Metafields + Liquid snippet + QAPage schema. Mobile QA: accordion, readability, visible CTA.
Playbook 5: bot sync + KPI W+4
Re-index RAG. Align macros. Dashboard: pre-purchase tickets, add-to-cart, bounce, unmatched. Bi-weekly 30-min ritual.
Useful linking
Your customers are already designing the ideal PDP in their questions. All that's left is to choose the right block, in the right place, using their words.
Enzo
June 28, 2026
