Survey Stimulus Management in India: Concept Cards, Shelves, Video, Timers, Sort, Drag and Swipe on SuperJ

SD-036 Concept Card / Pack Shot Display Is Controlled Image Exposure by Design: A pack or concept is not “an image in a question”. It is a stimulus object with a stable ID, a minimum view time, zoom, mobile-responsive rendering at 1080, and rotation tracking that survives reporting. On Hercules Works, an image_multiple_choice or image_question with media ref 1080 travels inside questions_json as a typed question, its URL injected later via POST /api/chats/{id}/inject-media and tagged to a stimulus ID that every follow-up question references for rotation and recall. That ID is why SD-008 Latin Square rotation can balance each pack in each position the same number of times across respondents, eliminating order and position bias in sequential monadic designs, and why SD-010 Random Rotation can shuffle for large samples. Built by Jupiter Meta Labs in Hyderabad, this stimulus-ID discipline turns a pretty picture into measurable exposure data that Poseidon later models as a Knowledge Graph edge with confidence EXTRACTED. See rotation design at market segmentation survey India and deployment of media at survey deployment superj India.

Where Pack Shots Win or Fail in India: FMCG packs win on shelf standout, claim comprehension and value cues. A pack shot that renders at 420 pixels on a 6-inch phone, cropped by an email router, tells you nothing about shelf pop. On the SuperJ app, 1080 images render full-bleed with pinch-zoom and a forced exposure timer from SD-039 that holds the stimulus for a mandatory minimum duration before the respondent can proceed. That timer is a flag in the SuperJ payload, not a polite request, so the Parquet that lands for Poseidon has exposure duration per respondent, not just a self-reported “I saw it”. Pair this with image cards for sequential monadic or proto-monadic designs where each stimulus is evaluated individually before a final preference screen, and you get both individual reaction and comparative preference without contamination. Explore shopper context at fmcg consumer research India and panel delivery at consumer panel India.

From Exposure to Verifiable Recall: After forced exposure, recall and comprehension questions are automatically gated by showIf that waits for exposure completion, not a self-report. That guarantees every claimed recall came from someone who actually saw the pack for the mandated seconds, so your “62% recall among Tier 2 mothers” is defensible with correct base size via routing-pair denominator correction. Poseidon’s Data Analysis node then merges exposure flags with rating grids and ranking Borda counts without manual VLOOKUPs, and the chart advisor picks grouped bars or ranking charts per data shape using an Okabe-Ito inspired palette. On Hercules Works this is painless because the SuperJ panel and Poseidon handle the mechanics while you keep control over stimuli, order and duration, with rewards sustaining 60-90%+ completion even on image-heavy packs.

SD-037 Shelf Display / Planogram Testing Brings the Kirana Aisle Into the Survey: A virtual shelf shows multiple pack variants in a retail shelf context — the planogram — so respondents can pick, examine, zoom or rank products as they would in-store, not as isolated images. On Hercules Works, the shelf is a stimulus composite built from pack shots tagged to IDs, rendered as a responsive planogram on the SuperJ app, with interactions like tap-to-examine and rank that map to matrix rows downstream. That context matters because shelf distinctiveness is not about liking one pack; it is about being found and chosen among five competitors in four seconds. For packaging distinctiveness, shopper conversion and shelf-standout studies, SD-037 is the missing bridge between lab ratings and in-aisle behaviour. See shopper methods at product concept testing platform India and retail consumer research India.

How Shelf Maps to Data Without Losing the Context: Behind the shelf is the same Parquet columnar model that makes Poseidon fast. Each pack position on the shelf maps to a column with known type, each interaction to a valid formula — frequency for picks, Borda for rankings, grid heatmap for attribute batteries — and Poseidon’s column-to-formula mapping validates every analysis automatically so percentages and ranks are never invented. Denominator correction respects that only shelf viewers answered the shelf task, so cross-tab bases are correct by city tier or NCCS without you writing WHERE clauses. The 18-node report then narrates the shelf story around the research brief — “Pack C wins on shelf pop among Tier 2 value-seekers, but loses on trust among NCCS A metros” — with the “So What” rule that leads with business implication, then evidence, never the reverse. Built by Jupiter Meta Labs in Hyderabad, SD-037 turns shelf art into analysable data on a phone.

When to Choose Shelf Over Single Pack: Choose SD-037 when distinctiveness and findability are the question: new pack architecture, private-label defence, value-tier launch. Stick to SD-036 single card when pure concept liking is the question: claim, benefit, or price communication. On Hercules Works you can even combine them in a proto-monadic flow — evaluate one pack monadically with its battery, then see all packs together for a final preference — with rotation, timers and exposure flags consistent across both stages. That composite stimulus history travels into Poseidon as routing edges and stimulus IDs for longitudinal reuse. Explore rotation discipline at quantitative research methods India and visual reporting at survey data visualization India.

SD-038 Video Stimulus Insertion Is Non-Skippable Playback With Comprehension Gates: A video ad is not a link to YouTube. On Hercules Works, SD-038 embeds video as a typed video question with forced exposure flags in the SuperJ payload, so playback is non-skippable on the SuperJ app and completion is enforced before the next question renders. Comprehension and recall questions are then triggered automatically upon completion via showIf that references the video qId, not a respondent’s claim. That enforcement is why a TVC tested in Hindi and Hinglish across Mumbai and Hyderabad can report recall that you can defend: every “yes, I recall” came from someone the app held on the film. Treat this as the ad pre-testing baseline for TVC, branded content and influencer cuts; without it, recall is inflated by skimmers. See ad diagnostics at ad effectiveness research India and multilingual delivery at multilingual survey tool India.

SD-039 Forced Exposure Timer — Minimum View Time Before Proceed: SD-039 holds any stimulus — image, shelf or video — for a mandatory minimum duration before the respondent can proceed, configurable per stimulus. It prevents the cursory swipe where a respondent flicks past a pack in 0.4 seconds and still answers “likeability 9”, a common source of inflated concept scores on email panels. On Hercules, the timer is a runtime flag on the question that SuperJ enforces natively; it pairs with video non-skippable logic and with SD-036 pack zoom, so exposure is both timed and verifiable. Poseidon later treats timer satisfaction as part of the routing pair for denominator correction, so your liking metric is calculated only among genuine viewers. Built by Jupiter Meta Labs in Hyderabad, this timer-plus-video combo is why ad and pack metrics on Hercules Works survive board scrutiny while link-out methods do not.

From Enforced View to Verifiable Insight: When video and timer events land as flags in Parquet, Poseidon’s planners map them as routing pairs and the Survey Knowledge Graph carries exposure edges with EXTRACTED confidence. Every query then applies the correct WHERE automatically, chart advisor selects horizontal bars for recall percentages or heatmaps for cross-tab recall by city tier, and narrative synthesis frames causality correctly: “Recall follows guaranteed exposure, not self-report, so lift is comparable across Hindi and Hinglish cells.” DuckDB at 10-20× over row-based tools and semantic cache at 78% keep those cuts instant, and SSE streaming with persistence means the 18-node report never loses work on refresh. On Hercules Works this is painless because media enforcement and verification are infrastructure, not follow-up tasks.

SD-040 Card Sort Organises Items Into Meaning on a Phone: Card sort asks respondents to organise items — attributes, concepts, brands or claims — into self-defined or predefined categories via drag-and-drop, producing category membership and similarity data that powers brand positioning and benefit segmentation. On Hercules Works, card sort is a vertical_ranking grid or ranking variant optimised for touch, where categories are stimuli IDs and moves are logged as typed interactions in questions_json, not as loose text. That structure matters for Poseidon: each placement maps to a column with known type, each category to a valid formula for co-occurrence and grid heatmaps, so similarity is measured correctly without manual coding. Use card sort when taxonomy is the question — how Indian consumers group “authentic” versus “convenient” in ready-to-eat, or “trust” clusters in BFSI. Context at market segmentation survey India and qual context at qualitative research India.

SD-041 Drag & Drop Ranking Makes Priority Thumb-Natural: SD-041 replaces dropdown ranking with touch-optimised drag-to-order lists — respondents rank claims from most to least compelling by dragging large cards into priority order, with min and max selections enforced so incomplete rankings are not counted. On mobile-first India, this is not a cosmetic choice; dropdowns punish Tier 2 respondents with tiny hit targets and cause systematic drop-off that concentrates responses in metros and among desktop-literate users. On the SuperJ app, drag-rank renders as large cards with haptic-friendly targets and smooth snapping, and routing can enforce that ranking is completed before proceeding, so Borda counts and median ranks reflect genuine trade-offs rather than ability to tap a 12-pixel row. Poseidon’s column model stores the ordered semicolon list with the Borda formula pre-bound, so “which claim is most compelling among Tier 2 mothers?” returns a verified ranking with correct base, not a guessed chart. See ranking methods at consumer segmentation analysis India and maxdiff survey tool.

Why Mobile-Optimised Sorting Changes Who You Hear From: When sorting is thumb-friendly, completion among Tier 2 and Tier 3 respondents rises from the 30% that desktop patterns yield to the 60-90%+ that app-native delivery ensures on the SuperJ app, with rewards and ZK-verified uniqueness keeping quality high and zero bots. That reach is why card sort and drag-rank on Hercules Works do not just improve UX; they change inference by removing the literacy and device bias that would otherwise concentrate responses in metros and NCCS A. Built by Jupiter Meta Labs in Hyderabad, SD-040 and SD-041 are how you hear Bharat’s taxonomy and priorities clearly — authentic versus convenient, trust versus price — then analyse them deterministically via DuckDB on Parquet, not via a transcriber’s guess or a spreadsheet patch.

SD-042 Swipe-Based Rating Turns Screening Into a Thumb Game: Swipe-based rating shows concepts, images or statements one at a time and lets respondents swipe left to reject or right to accept, with an optional rating overlay or like/dislike toggle. On Hercules Works, swipe is a mobile-optimised stimulus type where each swipe maps to a binary or ordinal column with frequency and overlap formulas pre-bound, and the sequence is fully rotatable so order does not decide fate. This highly engaging format is ideal for rapid concept screening, claim evaluation and creative pre-screening on the SuperJ app, where swipe keeps 18-24 attention and suppresses speed-through because each stimulus demands a deliberate gesture. For Gen Z reads see best online survey tools for gen z market research 2026 and top survey platforms for market research on teenagers 2026.

When Swipe Beats Grids: Choose swipe when you have many stimuli and need quick like or dislike with breadth: 30 pack claims, 40 ad frames, 50 benefit words. Use grid or drag ranking when you need depth: prioritise the top 5 with Borda and median rank with explicit confidence. On Hercules Works, you can chain both in a funnel — swipe 30 claims to find the top 12 by accept rate, then drag-rank those 12 for priority with Borda scoring — with stimulus IDs consistent across both stages so Poseidon can trace the funnel from swipe accept rate to Borda rank in one cohesive narrative with goal traceability. Exposure flags from SD-039 still apply per stimulus, so a swipe without the minimum view time is not counted, keeping rapid screening fast without being sloppy or recall-inflated. See method fit at advanced survey analytics and nps analytics platform.

From Thumb Gestures to Verifiable Strategy: Swipe data lands in Parquet as per-stimulus columns where multi-select normalisation via UNNEST of semicolon-delimited storage is handled automatically, so percentages are selections not respondents where the formula demands it. Poseidon’s Data Analysis node aggregates swipe accept rates, Open-Ended Intelligence codes optional swipe reasons in 8+ languages including Hinglish, Visualisation picks horizontal bars for accept overlap and heatmaps for cross-tabs, and Narrative Synthesis writes the funnel around the research brief with goal traceability and the “So What” rule before the report is assembled in the 18-node graph. DuckDB at 10-20×, semantic cache at 78% and three-layer verification at 99.1% keep swipe-at-scale both fast and auditable, with streaming that survives refresh. Built by Jupiter Meta Labs in Hyderabad, SD-042 is how Indian brands screen 50 ideas in an afternoon on verified humans via the SuperJ app at subscription pricing from ₹0/month.