What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
Generative AI creates the most marketing value when it improves an end-to-end workflow—not simply when it increases the volume of content. The strongest opportunities are creative development and campaign work, followed by personalization, customer insight, reporting, repetitive service tasks, and market-opportunity discovery. The test for any use case is whether it improves a defined outcome enough to justify the cost of integration, review, and change.
Where does generative AI create the most value in marketing?
Creative development and campaign work are the most established starting points in the available survey evidence. In Gartner’s survey of 418 marketing leaders, fielded July–September 2024 and published February 18, 2025, 77% of organizations adopting generative AI said they used it for creative development; the figure was 84% among Gartner’s high-performing group. Strategy development adoption was 48% overall and 52% among that high-performing group. Gartner also reported that 47% saw a large benefit from AI in campaign evaluation and reporting. These are survey findings, not proof that AI itself caused stronger performance.
Other workflows can offer value when they connect generation or analysis to a real decision or customer interaction. McKinsey’s examples include personalization, interpreting customer feedback, service response, and identifying product or campaign opportunities. They illustrate possible applications, not guaranteed results.
Free tools Windows power users keep installed
One-click scans. No signup required.
| Workflow | Where AI can help | Outcome to evaluate |
|---|---|---|
| Creative and campaign development | Draft and adapt campaign copy or concepts; support planning and strategy. | Campaign quality, engagement, conversion, or production time—not raw asset volume. |
| Personalization | Generate or adapt messages for segments and use response data to refine them. | Engagement, conversion, and relevance, balanced against the cost and risk of using customer data. |
| Customer insight | Help organize feedback, comments, and preferences for human interpretation. | More useful recommendations or faster identification of customer needs. |
| Campaign evaluation and reporting | Assist with summarizing performance and surfacing patterns for review. | Timeliness and usefulness of decisions; verify calculations and underlying data. |
| Service and repetitive process steps | Retrieve information, draft replies, or support routine actions within a defined workflow. | Response and resolution time, service quality, and escalation rates. |
| Opportunity discovery | Scan trends and customer commentary for concepts, product needs, or campaign ideas. | Relevant opportunities identified and validated—not ideas generated alone. |
Examples from McKinsey’s marketing analysis show the range. Stitch Fix uses generative AI to help stylists interpret customer feedback and make product recommendations; Instacart uses it for recipes, meal planning, and shopping lists. Mattel has used AI to generate Hot Wheels concept images, Kellogg’s to scan recipe trends for social campaign ideas, and L’Oréal to analyze online comments and media for product opportunities. These are company examples, not controlled comparisons.
#1 Best Overall
What personalization results can—and cannot—show
McKinsey reports that Michaels Stores used a content-generation and decision platform to increase the share of email campaigns personalized from 20% to 95%. The same article relays case figures of a 41% increase in SMS click-through rate and a 25% increase in email click-through rate, attributed to a Persado case article dated March 30, 2022. These are company case claims reported by McKinsey, not independently audited causal estimates or a forecast for other marketers.
Salesforce’s tenth State of Marketing report says it surveyed 4,500 marketing leaders worldwide and found that 83% recognized a shift toward personalized, two-way messaging, while only one in four were satisfied with how their organizations use data to support it. The page’s methodology describes the survey period as October–November 2025. This vendor-sponsored survey describes reported priorities and satisfaction, not the effectiveness of a particular AI system.
Where service automation may help
McKinsey describes an unnamed direct-to-consumer retailer using generative AI to retrieve back-end information, take actions, and draft customer replies in the brand voice. The article reports an over-80% decrease in time to first response and a four-minute reduction in average resolution time. Those are case-study figures for that retailer; they should not be generalized to other organizations or treated as results attributable to AI alone.
Rank #2
Why more adoption does not automatically mean more value
McKinsey’s State of AI survey, fielded July 16–31, 2024 and published March 12, 2025, included 1,491 respondents in 101 nations. Among respondents, 71% said their organizations regularly used generative AI in at least one business function, up from 65% in early 2024. Marketing and sales were among the functions most often reporting use. This is a survey of respondents, not a census of organizations or a measure of marketing effectiveness.
Reported financial impact was less widespread: more than 80% of respondents said their organizations were not seeing a tangible enterprise-level EBIT impact from generative AI, while 17% said at least 5% of organizational EBIT in the prior 12 months was attributable to it. These are respondent reports, not audited financial results, and do not isolate marketing’s contribution. In the same survey, 21% of respondents at organizations using generative AI said at least some workflows had been fundamentally redesigned.
Rank #3
McKinsey found workflow redesign had the biggest effect among 25 tested organizational attributes on the ability to see EBIT impact, while tracking well-defined KPIs had the most impact on the bottom line among 12 adoption and scaling practices examined. These are survey analyses and associations, not proof that a particular redesign or KPI will cause a fixed return. The practical implication is to improve the workflow and measure its intended result, rather than treating access to a model or generated volume as evidence of value.
Which marketing tasks should teams automate or augment first?
Start with bounded tasks that already have an owner, a clear outcome, and a reviewable result. Use AI to assist with a step before handing it authority to make customer-facing decisions or take consequential actions. McKinsey recommends concentrating first on two or three use cases where off-the-shelf tools can make an immediate difference in priority areas, then learning before pursuing customized systems or broader process transformation.
Rank #4
- Choose a specific workflow. Name the task, its current handoffs, the people responsible, and the point at which AI would contribute. “Improve campaign performance” is too broad; “draft segment-specific email variants for review” is testable.
- Set a baseline and KPI. Record current performance before changing the workflow. Choose a measure tied to the intended result, such as review time, conversion, response time, or customer-service quality. Assign someone to own the measure.
- Count the full cost. Include tool and integration work, data preparation, human review, training, and ongoing maintenance. Time saved at one step may be offset by extra checking or correction elsewhere.
- Run a contained pilot. Limit the audience, data, and decisions involved. Compare results with the baseline or an appropriate control where feasible, and account for other changes that could affect the outcome.
- Review errors and customer effects. Track inaccurate, biased, inappropriate, or unusable output and any resulting escalations. Define who can stop the workflow and how staff correct or report problems.
- Scale only when the evidence supports it. Keep, adjust, or stop the use case based on measured outcome, total effort, and risk. Extend it only when the workflow, controls, and accountable ownership can support wider use.
How should marketing teams measure generative AI’s ROI?
Measure the result the workflow is meant to produce, not the amount of AI activity. A content workflow might target faster production without losing quality; a service workflow might target faster resolution while preserving customer satisfaction. For each use case, define a baseline, a KPI, an owner, and the period over which results will be assessed.
- Separate output from outcome. Count drafts or assets only as process measures. Pair them with quality, engagement, conversion, speed, or another business result that matters for the workflow.
- Include review and rework. Record time spent checking, correcting, escalating, or replacing AI output. A faster first draft is not a net productivity gain if downstream work grows by more.
- Be careful with attribution. Campaign results can change because of audience, offer, timing, channel, or other adjustments. Where possible, use a comparison that helps distinguish the AI-supported workflow from those factors.
- Track quality and risk alongside efficiency. Monitor accuracy, customer complaints, escalation rates, and policy or privacy incidents so a speed gain does not conceal degraded outcomes.
- Review value at the workflow level. Compare measured benefits with integration, oversight, training, and operating costs. Do not infer enterprise financial impact from one successful task or a higher volume of generated content.
What risks should marketing organizations control?
McKinsey identifies hallucinations, bias, data-privacy violations, and copyright infringement among generative AI risks. Its marketing analysis also says the technology is typically poorly suited to high-stakes decisions, regulated environments, heavy-volume requests, or numerical reasoning. Marketing teams should match controls to the consequence and exposure of each use case rather than assume a generic review step removes risk.
Best Value
- Accuracy: Verify factual claims, product details, prices, offers, and numerical statements against authoritative sources before use.
- Privacy: Limit personal or confidential data to what the workflow permits, and ensure data handling follows organizational requirements and applicable obligations.
- Bias and appropriateness: Check whether targeting, language, or recommendations produce unfair or unsuitable treatment across customer groups.
- Intellectual property: Establish review and usage rules for generated material and source inputs; do not presume that generated output is free of rights concerns.
- Accountability: Name the business owner, technology oversight, escalation path, and person authorized to approve customer-facing output or actions.
Human review should be proportionate to exposure: a low-risk internal brainstorm may need a different control than a promotion sent to customers or an automated service action. In McKinsey’s 2025 survey, among respondents at organizations using generative AI, 27% said employees reviewed all AI-created content before use, while a similar share said 20% or less was checked. The variation highlights that “human review” is not a consistent control unless its scope, responsibility, and escalation rules are explicit.
Quick Recap
Sources and scope
- McKinsey, “The state of AI: How organizations are rewiring to capture value” — published March 12, 2025; survey fielded July 16–31, 2024.
- Gartner, survey of generative AI adoption in marketing campaigns — published February 18, 2025; survey fielded July–September 2024.
- McKinsey, “The power of generative AI for marketing” — marketing use-case analysis and company examples.
- Salesforce, State of Marketing Report: Tenth Edition — vendor-sponsored survey page; survey period stated as October–November 2025.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
