---
格式版本: 2
标题: "Scaling expertise with Microsoft Foundry"
原文链接: "https://azure.microsoft.com/en-us/blog/inside-microsofts-marketing-team-scaling-expertise-with-ai/"
发布日期: "2026-09-01"
发布时间校准状态: "found"
发布时间需复核: "否"
发布时间来源: "rule:configured_publication_date_rule"
发布时间证据: "azure-blog-publication-date html:original: <time datetime=\"2026-08-31T09:00:00-07:00\""
发布时间校准原因: "信源发布日期识别规则直接确认发布时间"
发布时间校准置信度: "high"
发布时间候选数量: 1
发布时间严格候选数量: 1
发布时间原页读取状态: "source template page reused from URL open"
发布时间未找到原因: ""
发布时间校准时间: "2026-09-07T22:12:42+08:00"
发布时间仲裁状态: "skipped"
发布时间仲裁尝试次数: 0
发布时间仲裁耗时毫秒: 0
发现时间: "2026-09-07T22:06:51+08:00"
入库时间: "2026-09-07T14:31:10.486Z"
来源平台: "固定入口"
搜索渠道: "fixed_url"
搜索词: "https://azure.microsoft.com/en-us/blog/"
匹配关键词:
  - "AI"
相关厂家:
  - "Microsoft"
相关专家:
  []
内容类型: "网页"
抓取工具: "Free Fetch + Defuddle"
清洗工具: "Defuddle Markdown + Defuddle/Readability 正文提取"
原始附件:
  []
AI优质: "否"
AI打分: 38
AI分档: "非优质"
AI质检状态: "不通过"
AI打分理由: "正文主线是微软营销团队利用Microsoft Foundry构建内容审核、消息验证和信息协同代理，属于企业应用与工作流提效，不涉及超节点、AI机架、互连、供电、液冷或机架级部署。来源为微软官方博客，正文完整；相较固定知识库可提取的新事实包括每年审核逾200篇博客、估算每年节省逾2,000小时及连接多类营销运营数据，但均为应用层内部案例。未检出可证明首次出现的机架级新增事实，也无相关产品量产、客户采购或基础设施部署信号，命中应用与模型效率类强否决项。"
AI质检模型: "gpt-5.6-sol"
AI质检时间: "2026-09-08T16:43:41+08:00"
AI主题相关性: 0
AI来源权威性: 14
AI新颖性: 8
AI技术细节: 3
AI商业部署信号: 3
AI完整性: 10
AI评分提示词版本: "v17-精简生产版"
AI评分提示词SHA256: "48fb9777f386026761b4873eaff30807694fb11e9b352d7c69bf2dfde750cc7d"
AI评分知识库版本: "knowledge_base_v1-20260819+runtime.92"
AI评分知识库SHA256: "87cb146754f554920189197de0a266d3b3bbdd834b3853337f49da5d6a0d6c61"
AI评分知识库检索词: "[\"Microsoft\",\"https://azure.microsoft.com/en-us/blog/\",\"Intel\",\"IQ\",\"AMA\",\"ROI\"]"
AI评分知识库命中: "[{\"id\":\"runtime-c9b3ba1144b3cd66b971cde2\",\"title\":\"AMD Instinct GPUs and EPYC CPUs to Power Europe’s Next-Generation LUMI-AI Supercomputer\",\"sourceType\":\"ai_excellent_article\",\"time\":\"2026-08-31\",\"matchedTerms\":[\"IQ\"],\"rank\":-6.30820456124505},{\"id\":\"july-correct-0034\",\"title\":\"AMD Fires Back at Nvidia with Helios AI System, Epyc CPUs\",\"sourceType\":\"labeled_article\",\"time\":\"2026-07\",\"matchedTerms\":[\"Microsoft\",\"Intel\"],\"rank\":-6.107100820417285},{\"id\":\"july-correct-0081\",\"title\":\"AAI 2026: 6th Gen AMD EPYC Server CPUs Power the Agentic Data Center\",\"sourceType\":\"labeled_article\",\"time\":\"2026-07\",\"matchedTerms\":[\"Intel\",\"IQ\",\"AMA\"],\"rank\":-5.335153285260406},{\"id\":\"historical-may-024\",\"title\":\"OpenAI、Microsoft等围绕MRC协议构建更大规模AI以太网训练网络\",\"sourceType\":\"curated_item\",\"time\":\"2026-05\",\"matchedTerms\":[\"Microsoft\",\"Intel\"],\"rank\":-4.889416188093993},{\"id\":\"july-correct-0068\",\"title\":\"Microsoft Azure Expanding AI Infra Choice with AMD Helios™\",\"sourceType\":\"labeled_article\",\"time\":\"2026-07\",\"matchedTerms\":[\"Microsoft\",\"Intel\"],\"rank\":-4.693996400254436}]"
采集批次: "2026年9月7日22点05分18秒"
采集批次ID: "20260907-220517-361"
去重键: "https://azure.microsoft.com/en-us/blog/inside-microsofts-marketing-team-scaling-expertise-with-ai"
---

Business leaders are facing a familiar challenge at an unfamiliar scale.

Every organization is being asked to move faster as markets change quickly, customer expectations continue to rise, and technology advances at a pace that can feel overwhelming. Teams are expected to deliver greater results, often with the same resources they had before.

AI is helping organizations meet those expectations. Some of the strongest examples I’ve seen revolve around scaling the judgment, strategy, and success measures that strong performers already set for themselves and their teams. AI agents apply that expertise consistently across a growing volume of work, helping them deliver more without sacrificing quality.

We’ve seen it firsthand on my team. As innovation cycles have accelerated, product launches have increased from a quarterly cadence to weekly—and sometimes even daily—events. Our teams are now supporting a growing volume of launches, up to 150% year over year.

To relieve the pressure, we’ve looked for places where AI can help teams at Microsoft find the right information faster, reduce repetitive coordination, and bring more consistency to work that depends on shared context. To do that, we used [Microsoft Foundry](https://ai.azure.com/home), Microsoft’s platform for building and managing enterprise AI applications, to create agents grounded in business knowledge and embedded in the flow of work, helping our teams operate at greater scale while staying focused on the work where their expertise matters most.

## Why context matters

One lesson became clear very quickly: AI is only as good as the data it has access to. General-purpose AI can generate content, but enterprise decisions depend on information spread across documents, workflows, business systems, communications, and institutional knowledge.

For us, [Microsoft IQ](https://www.microsoft.com/en-us/ai/microsoft-iq) helped connect that business context to our AI capabilities. Rather than asking employees to assemble information from multiple sources, agents could draw from the same knowledge people rely on every day to surface relevant information and support better decisions.

But [IQ](https://www.microsoft.com/en-us/ai/microsoft-iq) does more than ground AI in the right data. It helps connect the knowledge and workflows that shape how the business actually operates.

That shift changed the role AI could play. Instead of simply helping people find information, it could help teams work from a shared understanding of what’s happening across the business.

[Learn how Microsoft IQ helps bring together people, data, knowledge, and workflows](https://www.microsoft.com/en-us/ai/microsoft-iq)

Context alone wasn’t enough. The breakthrough wasn’t a single agent. It was creating a way for teams to build on what was already working.

As people shared successful agents and AI skills, expertise started becoming easier to reuse and scale. Ideas that began with one team could quickly create value for many others.

[Microsoft Foundry](https://ai.azure.com/home) became important because it allowed us to ground agents in organizational knowledge, connect them to existing workflows, and operationalize them beyond a single team.

In many ways, this reflects a broader lesson across AI adoption. As Jay Parikh recently wrote, “ [AI alone doesn’t transform a business. The system around it does](https://blogs.microsoft.com/blog/2026/06/02/ai-alone-wont-change-your-business-the-system-running-it-will/).” The following examples show what that looked like inside our marketing organization:

## Raising the quality bar at scale

As our [Microsoft Foundry](https://ai.azure.com/home) business grew, so did the volume of content we needed to create. Our team now reviews and publishes more than 200 blog posts each year, maintaining a consistent quality bar increasingly dependent on a small number of subject matter experts. Much of their time was spent applying the same review criteria over and over again.

Rather than reviewing every draft from scratch, one of our content leaders documented the rubric she uses to evaluate a strong blog and refined it until it reflected the standards our team expected.

Using [Microsoft Foundry](https://ai.azure.com/home), we translated that expert-defined rubric into a repeatable workflow that could identify gaps and opportunities before content reached a human reviewer. The capability was integrated directly into the content creation process, bringing instant feedback to every drafted post and making expert-defined standards available to every content creator.

Review cycles that once required substantial manual effort can now be completed in minutes, resulting in higher satisfaction and over 2,000 estimated hours saved annually across the team. More importantly, the approach demonstrates a broader pattern organizations can apply in many domains: use AI to apply established criteria at scale so experts can focus their time where judgment, coaching, and experience create the most value.

What we automated is consistency, not judgment. Our team set the bar based on our expertise; the AI agent reviews every post against that bar.

## Validating messaging before it reaches customers

As the pace of innovation accelerated, one question kept coming up: would our messaging resonate with the customers we were trying to reach?

At Microsoft, we aim to keep the customer at the center of everything we do. That led us to look for ways to evaluate messaging before it reached customers, using more than internal opinions alone.

We applied that approach through AI Messaging Assistant (AMA), which helps evaluate messaging and positioning against different audience perspectives before going to market. Instead of relying only on internal opinions, teams can pressure-test whether a message is clear, relevant, and actionable for the stakeholders they are trying to reach.

Using [Microsoft Foundry](https://ai.azure.com/home), we grounded AMA with a virtual congress of personas based on real customer conversations and extended it with the expertise, product knowledge, messaging guidance, and business context our teams rely on every day. That made it possible to move from a one-off AI experiment to a repeatable workflow where teams could evaluate messaging against a shared understanding of audience needs rather than rebuilding that understanding for every review.

The broader pattern is using AI to pressure-test important decisions before they reach customers, partners, or employees.

[**Microsoft scales customer intelligence with AMA to speed decisions, deliver ROI**](https://www.microsoft.com/en/customers/story/26270-microsoft-microsoft-foundry)

## Keeping teams aligned as the pace accelerates

As launch activity accelerated across our business, keeping teams aligned became harder than creating the work itself. New announcements arrived daily. Priorities shifted quickly. Information was spread across planning backlogs, documentation, meetings, and operational systems. Our marketers were spending too much time assembling context and not enough time acting on it.

In response, we started by writing down how our marketing work actually gets done, turning an unwritten process into a clear specification. With that in hand, we could sort the work: which parts required a marketer’s judgment, which could be automated, and which could be delegated to AI.

Using agents built on [Microsoft Foundry](https://ai.azure.com/home), we connected the systems our teams already rely on, including planning backlogs, documentation, meeting signals, and other operational sources. Rather than manually gathering updates from dozens of places, teams can work from a real-time view of key developments, upcoming launches, and changes that affect go-to-market plans.

This transformed alignment from a manual effort into a repeatable workflow. Instead of spending time assembling information, teams spend more time understanding what changed, why it matters, and what actions to take next.

The challenge was never a lack of expertise. It was coordinating that expertise across a rapidly changing environment.

The outcome is not merely faster communication. It is better organizational alignment. When teams operate from the same base, decisions happen faster, handoffs become smoother, and organizations can respond more quickly to change.

## Scaling expertise, reducing friction

Across each of these examples, the goal wasn’t automation for its own sake. The goal was making expertise available wherever it could create value. Looking back, the lesson wasn’t that a single AI capability changed how we worked. It was that building the right system around those capabilities allowed expertise, context, and judgment to scale across the team.

Technology will continue to evolve. The pace of business will continue to accelerate. But the differentiator remains the same: People provide the judgment. People set the strategy. People define success. AI helps them scale it.

## Microsoft Foundry

Foundry helps teams create agents grounded in enterprise knowledge, connected to the tools people use every day, and designed for production workflows.

[Ready to build with Foundry?](https://ai.azure.com/home)

![Person working at computer.](https://azure.microsoft.com/en-us/blog/wp-content/uploads/2026/06/CLO20b_Evan_office_001-scaled.jpg)