Amazon India +0.98% Share of Voice in 6 Weeks: The Complete Playbook Behind the Number

Six categories. Six weeks. +0.98% Share of Voice on Amazon Rufus. +2.4% Top-5 presence. This is the complete methodology - what we ran, why it worked, and what a US brand should copy versus adapt.

The context

Amazon India is a top-5 ecommerce market by GMV and a top-3 market for Amazon by shopper count. Rufus rolled out to Amazon India in late 2024 / early 2025 as part of the APAC expansion. By early 2026 Rufus was already the primary discovery surface for a meaningful share of Amazon India shoppers - particularly in mid-market urban categories where shoppers wanted opinionated recommendations rather than sifting through a search results page.

Ankita Bajaj Shankar's team at Amazon India ran a targeted six-week Rufus optimization sprint across six categories in Q1 2026. We supported the sprint with the Citation Rank methodology and the parallel-query testing infrastructure. What follows is the exact sequence and the numbers.

The categories

Six categories were chosen because they had (1) meaningful Amazon India GMV, (2) demonstrable Rufus-driven discovery already visible in early testing, and (3) a mix of established and emerging brands so that lift could be measured across competitive intensity levels. The six were:

• Home appliances - kettles, blenders, air fryers.
• Personal care - hair care, skincare, oral care.
• Kitchenware - cookware sets, non-toxic pans, storage.
• Baby care - feeding, sleep, hygiene.
• Fashion accessories - bags, watches, sunglasses.
• Health & wellness - supplements, fitness, wellness devices.

Each category had ~30-50 anchor SKUs across the participating brand set, with a total corpus of ~220 SKUs going through the optimization sprint.

Week 1: baseline

What we did: For each of the 220 SKUs, we submitted 15-20 shopping queries to Rufus manually via the Amazon India app. Each query was designed to plausibly surface at least one SKU in the corpus. We logged Rufus's response - which SKUs it mentioned, in what order, at Top-1 / Top-3 / mentioned / not mentioned resolution.

Baseline numbers (aggregate):

• Share of Voice (the share of relevant Rufus responses in the corpus queries that mentioned any of the 220 SKUs): 4.32%
• Top-5 presence (share of relevant queries where at least one of the 220 SKUs appeared in Rufus's Top-5 recommendations): 18.4%
• Top-1 presence: 4.1%

Distribution of baseline health across SKUs:

• Roughly 35% of SKUs had zero Rufus presence for any of the relevant queries - these were the "invisible" SKUs to optimize first.
• Roughly 40% of SKUs had scattered mid-range presence - these were the "recoverable" SKUs where movement was possible.
• Roughly 25% of SKUs already had strong presence - these were the "defense" SKUs where we needed to hold rank, not chase it.

Immediate diagnostic: Q&A depth was the shared weakness across almost every underperforming SKU. Median Q&A count per SKU was 6. Top-quartile (already-winning) SKUs had 20+. This gap was the target.

Week 2: prioritization

What we did: With the baseline map in hand, we prioritized the 220 SKUs into three tiers:

• Tier A - attackable (~60 SKUs): SKUs where the current Top-3 Rufus recommendation was thin (short listings, low Q&A depth, few reviews). These were the highest-ROI targets - beating a thin incumbent with focused work.
• Tier B - competitive (~110 SKUs): SKUs where the current Top-3 was moderately strong. Movement possible but slower and required more effort per SKU.
• Tier C - defense (~50 SKUs): SKUs already in Top-3 for their key queries. Focus was on holding position (maintaining Q&A depth, sustaining review velocity), not aggressive rewrite.

We concentrated 60% of the sprint's effort on Tier A, 30% on Tier B, 10% on Tier C. This was the load-bearing prioritization decision. Chasing across all 220 SKUs would have diluted the effort to invisibility.

Weeks 3-4: listing rewrites (the Q&A depth push)

What we did: For every Tier A and Tier B SKU, we ran a listing overhaul focused on four things:

1. Q&A depth to 20 per SKU. Mined the SKU's existing reviews for real customer concerns. Drafted questions in genuine shopper language. Answered factually, prose format. Half via Manufacturer answered (the seller-side Q&A tool), half seeded organically through review-request follow-up.
2. Backend structured attributes to 100% completion. Every backend field - Item Type, Material, Ingredient List (for consumables), Compatibility, Capacity, Certifications - filled honestly. Many SKUs had 40-60% completion; we brought them to 95-100%.
3. Bullet copy rewrite emphasizing structured outcomes. Not marketing prose. Each bullet a distinct benefit or spec, stated cleanly.
4. Fresh review velocity campaign. Email follow-ups on recent orders. Insert cards in shipments. Vine enrollment for eligible SKUs. Goal: 3-5 new reviews per week per top SKU for the sprint duration.

What we did NOT do: major A+ content overhauls. A+ was left alone unless it was outright missing. Our field data (see How Rufus Ranks Products) had already shown A+ was low-weight for Rufus rank; the sprint focused effort where it moved the needle.

Weekly rescan: By end of week 4, we ran the full parallel-query test again. Results:

• Share of Voice moved from 4.32% → 4.71% (+0.39 points at the halfway mark).
• Top-5 presence moved from 18.4% → 19.7% (+1.3 points).
• Top-1 presence moved from 4.1% → 4.4%.

The half-way check confirmed the trajectory but also revealed a specific insight: Tier A movements were dramatic (~2× share); Tier B movements were more modest. The sprint's back half doubled down on Tier A tactics.

Weeks 5-6: iteration

What we did: With four weeks of movement data, we reallocated effort weekly toward the queries and SKUs showing the fastest movement. Specifically:

• Queries where Tier A movements were fastest got additional Q&A depth (specifically Q&As targeting that query's exact intent phrasing).
• SKUs still not moving after 3 weeks of investment were re-examined. Usually the issue was one of two things: (1) the incumbent's advantage wasn't in Q&A but in something we hadn't yet touched (usually Prime eligibility or a strong review recency signal), or (2) the query itself was too competitive and needed a different SKU angle.
• Sponsored Products reallocation. For Tier A SKUs moving well, we tested a 25% reallocation of Sponsored Products budget to Rufus-adjacent placements. Early payback was fast - the Rufus-adjacent slots converted materially better than legacy Sponsored placements in the same categories.

Weekly rescan (end of week 5): Share of Voice at 4.96% (+0.64 total), Top-5 presence at 20.3% (+1.9). Final rescan (end of week 6): Share of Voice at 5.30% (+0.98 total), Top-5 presence at 20.8% (+2.4 total), Top-1 at 4.7% (+0.6).

The numbers held on a second parallel-query test three weeks later, confirming they weren't measurement noise.

Ankita's takeaway

From her post-sprint debrief:

"The single biggest surprise for us was how much movement came from customer Q&A. Our teams had treated Q&A as customer service overhead. It turns out it's the single most important listing lever we have on Rufus. The second surprise was how quickly Rufus responded to Q&A additions - most of the movement we saw was in weeks 3-5, which is fast for any Amazon signal to reweigh."

Amazon India's teams have since made Q&A depth a standing metric in their Rufus dashboard.

Why this worked so fast in India (and what US brands should adapt)

Six weeks is fast for Amazon ranking movement. Two reasons the timeline was compressed for Amazon India that don't apply as directly to US brands.

1. Baseline listing quality was lower. Q&A depth in Amazon India categories averaged 6 per SKU. In US categories the average is 14. That means the marginal return per Q&A added is higher in India - every Q&A moves the SKU relatively more against a lower baseline. In the US, you're adding Q&As to a category where competitors already have 15-20; your delta per Q&A added is smaller.

2. Category competition was moderate. Amazon India's mid-market categories had fewer dominant brands than the US equivalents in Q1 2026. There was room to jump from mid-market SKU to Top-3 recommendation with focused work. US categories have deeper incumbency defense.

What US brands should copy:

• The prioritization framework (Tier A / B / C).
• The 60/30/10 effort allocation.
• The Q&A depth focus.
• The weekly-rescan iteration cadence.
• The Sponsored reallocation experiment in week 5-6.

What US brands should adapt:

• Expect a 10-12 week timeline, not 6.
• Set Q&A targets at 25 per top SKU (higher than India's 20) to meaningfully separate.
• Invest more in review velocity - US categories have deeper review moats.
• Don't cut A+ content investment; the US shopper is more likely to visit the PDP directly after Rufus recommendation, and A+ affects on-page conversion.

The generalizable playbook (US-adapted, 12 weeks)

Weeks 1-2 - Baseline. Same as India: parallel-query test, tier the SKUs, snapshot the 9 ranking signals per SKU. Full methodology in How Rufus Ranks Products.

Weeks 3-6 - Listing overhaul. Q&A depth to 25 per top SKU. Backend attributes to 100%. Bullet rewrite emphasizing structured outcomes. Review velocity campaign kicks off. Do not stop at the halfway mark; the compounding value is in the sustained work through week 6.

Weeks 7-10 - Iterate + amplify. Weekly rescans. Reallocate effort weekly to fastest-moving Tier A queries. Start the Sponsored Products budget reallocation (20-30% to Rufus-adjacent) as a 4-week test.

Weeks 11-12 - Compound + expand. Extend the methodology to the next 20 SKUs. Establish the ongoing monthly rhythm - Q&A additions, review velocity, monthly parallel-query rescan.

By end of quarter, US brands running this sequence typically see:

• Top-3 Rufus visibility grow 30-60% on optimized SKUs.
• Aggregate category Share of Voice grow +0.4 to +0.8 points on Rufus (versus Amazon India's +0.98 in 6 weeks - the delta is timeline more than magnitude).
• Rufus-adjacent Sponsored payback within 4-6 weeks.

What this doesn't do

Worth being honest about ceiling.

• This does not immediately translate to off-Amazon visibility. ChatGPT, Gemini, and Perplexity rank Amazon-listed SKUs based on different signals (external authority, press coverage, review depth on Amazon and elsewhere). Rufus optimization does not automatically win ChatGPT rank. See our surface-by-surface guide for the full picture.
• This does not close the checkout inside the agent. Amazon Rufus keeps the transaction inside Amazon; the merchant relationship is Amazon's. Winning Rufus rank is winning Amazon visibility; it doesn't rebuild your DTC customer relationship.
• This does not solve Alexa+ voice-agent commerce. Alexa+'s top-1-winner dynamic requires different work (outcome-focused claims, safe-default positioning). Rufus optimization is a prerequisite for Alexa+ but not sufficient.

If you want to win agentic commerce across all six surfaces - Rufus and the five external agents - you need the multi-surface strategy laid out in The Six AI Agents Every Brand Needs to Show Up In. Amazon Rufus is one of six and the tactics don't transfer cleanly.

CTA

If you want to run the equivalent 12-week US sprint on your top-20 SKUs, start with a baseline. Drop your URL at free-tools/citation-rank - we'll return a 24-hour report with your current Rufus visibility, the Tier A / B / C SKU classification, and the three highest-leverage moves for the first 4 weeks.

If you're ready to run the sprint with our parallel-query testing infrastructure and weekly reporting, book a demo. The Growth tier covers most DTC brands; enterprise pricing kicks in at higher SKU counts.

The +0.98% Share of Voice number is not magic. It is the outcome of six weeks of focused, prioritized, honest listing work. Any brand willing to run the same sequence - US, EU, or elsewhere - can produce a comparable move.

- The Tru Commerce team (formerly Asva AI)

The same measurement-first loop drove our Amazon India engagement, where Top-5 presence rose +2.4% in 6 weeks - see the Amazon case study. The usual first step is Share of Voice.