Cross-border e-commerce keeps growing, and so does demand for product content and ad creatives. How to raise production efficiency without sacrificing quality, and how to make creativity better serve ad performance, are questions sellers must solve as they scale. In this edition of "Partner · A Playbook for Going Global," we focus on "AI Intelligence & Ad Efficiency," unpacking a 2026 Amazon Ads Partner Awards case: how Amazon Ads partner Saixu ERP teamed up with TecquorA, using AI to connect content production with ad optimization and lift content capacity and creative iteration speed.
As Amazon Ads marketing scenarios grow richer, brands need more diverse content and creative assets for different ad touchpoints. For cross-border sellers with many SKUs and fast marketing cadence, content production has long relied on heavy manual input — from extracting selling points to writing copy, producing visual assets, then checking and iterating. Any bottleneck in the chain can stall the whole marketing rhythm.
How can AI not only speed up content production but also link creative optimization to ad performance?
Working with FMCG seller TecquorA, Amazon Ads partner Saixu ERP built the AI content engine "AI Creative," stringing together product understanding, selling-point extraction, copy and visual asset generation, quality checks and ad optimization. It helps TecquorA raise content production efficiency and keeps creative iteration close to real ad-campaign performance.
Industry challenge: exploding creative demand meets insufficient production capacity
The content-capacity problem isn't unique to any single brand — it's a common challenge across cross-border e-commerce advertisers.
As ads rely more on quality content to drive conversion, demand for creative assets is expanding rapidly: a single campaign often requires copy, images and video prepared at once, with selling points, category traits, language and visual specs all needing to match. During peak promotions, demand multiplies.
Traditional creative production easily bottlenecks: materials are produced item by item manually, capacity can't scale with SKU and campaign volume; too many production steps and long delivery cycles mean content costs accumulate with each revision round; product information is scattered across workflows, making copy, images and video hard to keep consistent; and creative production and ad optimization run separately, so last round's material performance rarely feeds into the next round's production decisions.
As a billion-yuan-level FMCG seller on the US marketplace, brand TecquorA wanted smarter digital and intelligent solutions — raising material generation efficiency and cutting content production costs, while making content itself fit consumer needs better and actually drive conversion.
That means creative production must move from one-off delivery to a sustainable, running pipeline.
Solution: from "one-off asset generation" to a "content-ad" closed loop
Saixu ERP's innovation: rather than positioning AI as a single-point "asset generator," it upgraded AI into an intelligent engine spanning the full content production flow. AI Creative's architecture revolves around three layers: a vertical pre-trained model at the base, multi-agent collaboration in the middle, and an ad-ecosystem closed loop on top.
1. Base layer: vertical pre-training on e-commerce data
Unlike general-purpose LLMs, AI Creative is vertically pre-trained on millions of e-commerce SKU data points, giving the model a native grasp of product attributes, selling-point logic and consumer attention dimensions in e-commerce contexts — content it generates can meet listing requirements without heavy manual correction.
2. Middle layer: a multi-agent production pipeline
Saixu ERP built a multi-agent collaborative system that decomposes content production into specialized agents, each with its own role: a product-understanding agent parses product attributes and category traits — answering "what is this product"; a selling-point agent extracts differentiated points from product data and consumer reviews — answering "why buy"; a copywriting agent generates multilingual, multi-scenario ad copy from selling points and target audiences; a visual-constraint agent ensures generated images and videos fit each ad placement's size, format and visual specs; and a quality-check agent runs compliance checks and quality scoring before output — moving what used to be "manual review" up-front into systematic QA.
This multi-agent system unifies text, image and video generation, connecting with product info, category data, historical assets and multilingual assets stored in the ERP for integrated batch production. Today, one SKU's creative need — from receiving the brief to delivering finished assets — can be completed within 24 hours.
3. Top layer: Amazon Ads API integration for a "content-ad-optimization" closed loop
AI Creative's value isn't just "producing fast" — it's "producing well." Saixu ERP further integrated Amazon Ads API, Amazon Marketing Cloud (AMC), Amazon Marketing Stream, the ad console, campaign manager and campaign reports, with Amazon marketplace as the fulfillment channel. Consumer insights and keyword expansion data are linked to asset generation — ad-side data feeds back to content production to guide the next round's selling-point choices and visual direction, forming a true "content-ad-optimization" closed loop.
Consumer insights + keyword data → AI Creative asset generation → Sponsored Products and Sponsored Product Video delivery → AMC + Stream performance feedback → back to the start.
For example, when AMC insights show a consumer segment paying more attention to a specific product function than expected, AI Creative can automatically adjust copy focus and visual presentation so the next batch of assets fits real consumer preferences. Content production is no longer an "upstream step" separate from ad delivery, but an organic whole iterating in sync with the delivery side.
In the collaboration with TecquorA (campaign period: July 2025 – February 2026), the AI Creative engine exceeded targets on three core KPIs: asset output tripled vs. the old process; review pass rate rose from 50% to over 95%; and delivery time compressed from 7 days to within 24 hours.
The release of asset capacity also translated directly into ad efficiency gains: in an ad mix dominated by Sponsored Products and Sponsored Product Video, average ad creative CTR rose 35% and average ROAS rose 18%.
Meanwhile, AI Creative's customer coverage climbed from 50% to 80% in 12 months, pushing "AI content operations" toward a scalable, replicable industry-level solution.
Brand takeaways and the next phase
The Saixu ERP–TecquorA collaboration offers three transferable lessons for going-global brands facing content-capacity bottlenecks.
1. The value of an AI content engine lies in rebuilding the entire content production flow. A standalone AI copy tool or image generator only solves part of the problem; a full-flow agentic pipeline from product understanding to content generation to quality checks delivers more effective overall efficiency gains.
2. The gap between content production and ad delivery deserves attention. When ad-side data can't flow back to the creative side, each asset iteration is judgment by experience rather than by data. Connecting the Amazon Ads API to close the "content-ad-optimization" loop is the key step for an AI content engine to deliver sustained optimization value.
3. Unlocking asset capacity lets creative teams do more valuable work. When repetitive work — batch asset generation, multilingual adaptation, format checks — is handled by AI, people's energy can shift from "making images" back to "strategy": brand narrative, content direction, test frameworks and other higher-value work.
Looking ahead, Saixu ERP plans to evolve AI Creative from general-category generation toward expert agents with deeper vertical-category experience, and shift from user-on-demand invocation toward proactive system diagnosis and recommendations. As more category knowledge and ad feedback enter the workflow, content teams can gain assistance closer to real product scenarios.
Conclusion
For this case, Saixu ERP won the Technology Innovation Award | 2026 China Winner at the Amazon Ads Partner Awards.
The Amazon Ads Partner Awards recognize partners that apply Amazon Ads solutions, leverage professional strengths and help brands achieve outstanding results. Saixu ERP's winning case showcases AI's potential in cross-border marketing: by connecting product understanding, content generation and ad optimization, it helps brands raise content production efficiency and lets creativity better serve business goals.
From content production efficiency to intelligent ad decision-making, AI is opening more growth space for cross-border sellers.
