What happens when every team gets faster, but campaign performance doesn’t improve?
In my previous blog, I explored how to make individual AI wins useful across an organization. The next challenge is harder: keeping the strategy, evidence, and reasoning behind a campaign connected as the work moves between teams. Inside Adobe’s Growth Marketing and Insights (GMI) organization, we’re building a connected operating model to preserve the context around the work.
The fragmentation trap
What your teams can do on their own is evolving rapidly, and as a marketing leader, you need to pay as much attention to siloed decision-making as to the time saved.
AI can reinforce the assumptions a marketer brings to a prompt. It can also decrease collaboration which enhanced ideas through diverse perspectives and pressure testing. What results are teams creating conflicting approaches based on different assumptions without recognizing where those plans diverge or conflict. Multiply that across teams and a campaign can launch on time while every group chases a different definition of success. Faster execution does not solve the problem if each team must reconstruct the strategy, evidence, and decisions behind the work before it can act.
We learned this in my own GMI organization. With broad access to AI tools and encouragement to experiment, insights became scattered across the organization. As work moves from the brief through planning, content development, activation, measurement, and optimization, different people and systems take over. Throughout this process, information is repeatedly re-shared, checked, and rebuilt.
The workflow can resemble a relay race, with each team completing its leg and passing a document or asset onto the next team of specialists. At every pass, meaning can dwindle or degrade. For customers, this can translate to an inconsistent experience. If an offer changes while the lifecycle team is preparing a message, the activation team needs the update before launching. Otherwise, the message can reach audiences with outdated information.
As AI speeds up execution, it becomes increasingly important for teams to work more intentionally toward the same business outcome and to proactively seek alignment, gaining better understanding of plans. Just as AI can increase execution speed, it can help connect and maintain the shared context teams need to align around a campaign’s strategy, evidence, and underlying reasoning.
The challenge is not only that information can fragment as it moves through the workflow. AI can also make it easier for teams to move ahead independently, before colleagues have had a chance to test the assumptions shaping the work. That is how fragmented context can become fragmented decision-making.
The island effect
When teams work from incomplete or different context, they can make plausible decisions that do not add up to a coherent campaign. AI can intensify this risk by making each function more self-sufficient and reducing the cross-functional interactions that historically helped teams challenge assumptions, refine ideas, and build institutional learning.
If interactive meetings give way to long prompts and offline documents without input from colleagues, important questions may never surface. As we incorporate AI tools into our marketing processes, it’s important to ask, “Who has had a chance to question this work?” Or “What might we be missing?” There’s no substitute for the lived experience, institutional knowledge, and insights of trusted colleagues.
A regional marketer knows why an audience assumption will not hold in a particular market. A creative colleague sees a mismatch between the message and the customer you intend to reach. You risk missing those insights when a recommendation takes shape with AI, possibly without those colleagues’ input.
Here are some ways in which we design rigor into how we work at Adobe GMI:
- Practice paired analytics. A second analyst examines the data and method before a recommendation moves forward. A finding from one market may justify a local change without supporting the same decision across every region.
- Shared performance measures. Your quarterly business review (QBR) provides an opportunity to examine recommendations across functions. In preparing one marketing QBR, teams in my organization created a shared source for customer group performance measures so everyone could work from the same information before debating what the results mean.
- Documented outcomes. Our Global Media Experimentation team connected its experiment library to a chat tool so colleagues could retrieve results from previous tests. A shared source of truth and access to earlier research provides teams with documented outcomes to examine before starting down a new path.
- Collective decisions. Decide which decisions require input from analytics, media, creative, brand, and regional teams, then build that review into your process. Ask your teams to record why a recommendation changes and the evidence behind the change. The next person working on the problem should be able to follow the reasoning.
The shared context advantage
Last quarter’s results can show what happened. They cannot, on their own, tell you whether the same audience is still relevant, whether the offer still fits the moment, or whether the conditions behind the result still hold.
A data foundation makes information available. A shared context foundation makes it usable. It brings together the institutional knowledge behind a decision: the strategic objectives, customer insights, briefs, assumptions, and business constraints that explain what the data means and how teams should act on it.
It gives people and AI a common basis for interpreting performance and making decisions within a specific campaign workflow — not simply retrieving results, but applying them with the right business and brand context.
Here’s how we’re bringing it together:
- Trusted facts. Approved business, customer, product, market and brand information, including what the brand can credibly promise.
- Strategic context. Jointly developed briefs, audience definitions, offers, guardrails, and decision records.
- Evidence. Analytics, research, experiments and performance signals, with the method and limits visible.
- Operational context. Consistent campaign and audience names, asset metadata linked to the correct offer and market, approval paths and workflow status. At GMI, a homegrown agent checks campaign assets for incorrect metadata before activation. It handles one task within a broader effort to connect the media workflow.
- Human contribution. An analyst qualifies a finding. A media buyer flags a placement requirement that affects which creative can run. Their updates should become part of the shared context so other teams can act on them.
A systems approach for marketing
Marketing campaigns span creative, media, agency, analytics, and activation teams. Yet the brief, asset, metadata, audience, budget, and performance history often do not move across systems together. The result is delays, handoff errors, duplicate work, and slower campaign optimization. Solving those breakdowns takes more than improving one task or function. Organizations that realize the value of AI will redesign how work, context, and judgment connect across the enterprise.
A systems approach organizes your marketing around a shared business outcome. It considers how decisions about audience, creative, media spend and activation affect one another, from the first brief through measurement and planning for the next campaign.
Your media planner needs to understand the audience for which your creative lead is developing content. Your campaign operations manager needs the approved offer, assets, and market requirements before launching. Your marketing analyst’s findings need to reach the teams making the next creative and spending decisions. Improving one task has limited value if the campaign stalls at the next handoff.
The human edge
Agents can surface signals and recommend actions. People stay accountable for making the calls that matter, like budget shifts, brand expression, market strategy, and customer experience tradeoffs. The return on your AI investment depends, in part, on how your teams use the time they save. As leaders, we need to reinvent our workflows but keep people’s judgement, skepticism and knowledge at the heart of the process.
The advantage in AI comes from the shared context, workflows, governance and human judgment that guides your agents. As AI helps us execute faster, our responsibility as leaders is to keep that context current and give people room to question it. Build a system where people lead, agents accelerate execution and every campaign builds on what the last one taught us.
I’ll be at Adobe MAX 2026 in Miami this week, and I’d love to connect and hear how your teams are tackling this.
