79 interested replies for Hector AI's Amazon ads platform across three geographies

About Hector AI

Hector AI is an Amazon ad technology platform with two product lines: a self-serve Ad Optimization platform that reverse-calculates the ideal bid per keyword every thirty minutes from live ACoS and CPC, and a self-serve Amazon DSP that removes the traditional entry barriers to programmatic advertising.

Both are built on a native MCP integration layer that connects Amazon Ads data directly to ChatGPT or Claude. Ranked among Amazon's Top 15 Advanced Partners globally. Public reference library: Dalstrong (79% ROAS lift), Noise (3x revenue lift via AMC retargeting), Nua (97% order-volume lift), Campus Shoes (1593% revenue growth on a fresh account), and R for Rabbit (19% ROAS improvement).

The challenge

01Two distinct buyer conversations, one platform. The Ad Optimization angle lands with sellers mismanaged by an average-ACoS approach. The DSP angle lands with brands hitting the Sponsored Ads ceiling. Same platform, different pains, different vocabulary.
02A wide geographic split. US, Rest of World, and India Amazon brands don't share a register. The US brand cares about Prime Video and AMC audience layering. The India brand cares about entry cost and no-minimum DSP access.
03A category flooded with agencies making the same claim. Every Amazon brand has seen an "optimize your ads" pitch. Cutting through meant naming a specific bid or retargeting pain in a performance marketer's vocabulary.
04A round-robin routing challenge. With three regional business heads owning different geographies, positive replies had to route to the right owner automatically.

What we did

Two product angles running in parallel across one book.

Separate active recipes for the Ad Optimization pain (bid recalibration every 30 minutes, ACoS drift on new SKUs) and the DSP pain (AMC audience build, DSP at $100/mo, no minimums, live in four days).
Every prospect got the angle that fit their observable: wide SKU portfolio with bid drift got Ad Optimization; a brand hitting the Sponsored Ads ceiling got DSP.
Reply-rate attribution split by product angle to see which pain did the work per tier.

Company-observable openers with strict verb discipline.

Every email opened with a peer-voice reaction to one specific observable: a recent SKU launch, a rebrand, a collab, a retail expansion.
Lead-in verbs rotated across "saw," "noticed," "caught," so no single verb dominated a batch. The words "loved" and "impressive" were banned.
The observable picked the pain: a new-SKU launch surfaced bid-drift-on-new-SKU; a retail expansion surfaced DSP-for-off-Amazon-audience.

Geographic recipe splits calibrated to market economics.

US and Rest of World brands ran one recipe track; India brands ran a separate track.
Each adapted the DSP economics: US brands heard Prime Video reach and AMC layering; India brands heard DSP starting at $100/mo with no upfront commitment.
Client-shared Amazon Advanced Partner data enriched the lists, so the audience was pre-qualified for scale before a single send.

MCP integration as the differentiated wedge.

Every campaign referenced the native MCP layer as the differentiator: analyze campaigns and take real-time actions directly from Claude or ChatGPT.
The MCP hook doubled as a signal to the technical buyer, who reads MCP integration as a real engineering choice, not a marketing phrase.

Two closing styles tested in parallel.

The Retail Revamp recipe tested a book-a-meeting demo close, phrased fresh per email ("worth a quick 15-minute demo to walk through how it plugs into your account?").
The Ad Optimization + DSP + MCP recipe tested a fixed peer-nudge close, verbatim every time: "Worth parking 10 minutes to explore?" No calendar link, no artifact promise.
Both ran concurrently so the sales team could compare demo-book rate by closing style, not just by product angle.

Round-robin routing across three regional owners.

Positive replies routed automatically across three regional business heads: US and Rest-of-World, ROW-only, and India.
The autoresponder looped in the right owner with their calendar link, CC'd the right team members per geography, and never mixed owners on one thread.
The routing logic became part of the campaign infrastructure, not a manual coordination task.

Results

Interested replies across US, ROW, and India Amazon brands

164,086

Emails sent across 37 active campaign recipes

Named demos booked with ad-platform prospects, spanning both product angles

Structure and volume are confirmed from 37 active recipes. Reply-rate and persona breakdowns aren't surfaced here by choice. Named demo prospects are anonymized per site rule.

Key performance highlights

37 active campaign recipes in parallel across Ad Optimization (Agencies + Retail) and DSP (Retail Revamp + Ad Optimization & DSP + MCP).
Two closing styles tested in parallel, a book-a-meeting demo close and a fixed peer-nudge close.
Client-enriched Amazon Advanced Partner data kept the audience pre-qualified on both US/ROW and India tracks.
MCP integration as a differentiated wedge, one of the only Amazon ad platforms with a native ChatGPT and Claude connection.
Round-robin autoresponder routing across three regional business heads handled positive replies without manual coordination.

Interested replies

Emails sent