While the news feeds of many chief marketing officers and chief revenue officers are dominated ... [+] byAI headlines, there's still a lot of confusion on how to get the best ROI from their GenAI efforts. (Photo by Jaque Silva/NurPhoto via Getty Images)NurPhoto via Getty ImagesMost marketing leaders have explore generative AI and personalization on their annual plans. Thats not mere conjecture, but an assertion rooted in evidence so much that AI has now become the latest buzzword in product marketing, as I wrote in this article.For context, a recent McKinsey survey revealed that the use of generative AI across organizations jumped from 55% in 2023 to 78% in 2024, with respondents most often report using the technology in the IT and marketing and sales functions.So, without a question, the news feeds of many chief marketing officers and chief revenue officers are dominated by AI headlines. Their inboxes are full of best practices and all or most of their vendors have integrated some form of productivity-boosting AI tools.Some of these tools are extremely helpful, while others seem to have taken inspiration from Microsofts Clippy, a well-intentioned but highly-obtrusive assistant from the 2000s that actively interfered with office work.The GenAI revolution holds both incredible promise and fear, but if you look back a few decades, you can see this pattern repeated in the desktop publishing, internet and mobile phone revolutions.Thirty years ago, when desktop publishing solutions first emerged, they took print production away from the professional designer and put tools into the hands of anyone with a PC. What resulted was a flooding of the market with fast, inexpensively produced, but amateurish content that often cheapened brand values.GenAI is no different. As Bill Anderson, global lead of revenue orchestration at Slalom, told me in an interview, the rapid emergence of GenAI now requires revenue growth and marketing leadership to take a step back and ask themselves not just what are the potential benefits of the technology, but also to take a sharp look at its costs and limits.This means starting with a careful strategic assessment before adding yet another platform to their existing tech stacks that takes a holistic look at the entire revenue pipeline.Holistic Revenue PipelinesMarketing tech providers have traditionally offered solutions that only support specific functions like content personalization, lead management, or customer analytics. At many organizations, whether small or large, the situation has created a confounding mess of disconnected systems, siloed data and ultimately a poor customer experience all of which negatively impact a companys revenue.Marketers built their pipeline from the perspective of the cloud software they were integrating. They forgot to look at the holistic revenue orchestration pipeline, creating a customer journey like a squirrel walking a picket fence. The platforms are optimized for their task, but the journey is difficult and uncomfortable.Even when theyre aware of the potential backlash that GenAI can produce, many chief revenue and marketing officers today dont fully grasp that success in the current business climate awash with tsunamis of data wont be achieved by simply adding an AI platform to their already bloated tech stacks, Anderson noted.Theres often a substantial, unrecognized cost involved in generating a mass amount of AI content; processing, storage and review fees often overlooked by brand managers. Press releases touting hundreds of thousands of images dont address the structural impact: humans still need to review the images and so companies have to hire third-party agencies to support the load.Simply adopting more and more AI tools in hopes that they will magically boost ROI is the fastest way to do less with more. As Jaspers 2025 State of AI in Marketing Report revealed, 56% of marketers are using AI in isolated, ad-hoc ways, and 51% cannot track ROI or see the true business impact of their AI investments.Instead, achieving the holy grail of an AI strategy that produces high-quality, brand-compliant marketing content on a massive scale, using key data points and performance indicators that support a positive customer experience, requires a fundamental rethinking by company leadership.So, before adding another AI-driven tool to an already bloated tech stack, marketing leaders must take a holistic view of the revenue pipeline and ask themselves if that AI tool will solve a real bottleneck, or just create a new one?Minimum Viable DataPerhaps counterintuitively, one of the biggest drivers of unnecessary costs associated with deploying GenAI into an organizations revenue chain results from an attempt to mine too much data with the system more than required to yield useful outputs.For Anderson, not only can an overambitious, data-greedy approach generate inaccurate conclusions about likely customer preferences and more, it inevitably results in excessive data storage and processing costs. Most large companies, for example, keep vast quantities of data stored across various departments, including previously abandoned pet projects, which are irrelevant and should be excluded from a new GenAI model.Based on my recent experience consulting with many CROs and CMOs at numerous Fortune 500 companies, I have observed that the most effective leaders recognize the importance of taking a more minimalist approach when deciding which data sources to tap with a GenAI model, he said.In this regard, Anderson offers a simple formula to keep in mind when planning and deploying a new GenAI solution into an organizations revenue chain: The minimum viable data required equals the most valuable data available, or MVD = MVD, for short. This simple formula describes the absolute minimum data points that your sales teams use to estimate if someone will buy or not.The MVD = MVD approach strips away all the cruft, historical sources and pet analytics to its core, allowing product owners and revenue marketers to focus on a few key points. Your most valuable data is the bare minimum an AI model needs to produce useful sales intelligence, like melting down gold to remove its impurities.Starting With StakeholdersEffectively applying the principle of MVD = MVD to GenAI entails leadership first bringing all the revenue chains stakeholders together in a planning-stage work group, with the goal of answering an initial question, What is the minimum viable data and therefore also the most valuable data for this proposed solution?By defining the minimum data needed to indicate a customers likelihood to buy, demand generation managers can develop a catalog of use-case experiments that define audiences, desired outcomes, and the data fields needed to support the experiment. This catalog helps refine the prompts and safe boundaries of generative output.In addition to tightly-defined experiments, marketers can also use image-to-text AI tools to generate keywords and metadata of an image. This can then be analyzed and leveraged to find hidden patterns that boost response rates.For example, Adobes Gen Studio may report that certain colors or lighting effects bring more clicks than others. Leveraging this automatic experimentation using metadata allows the marketer to combine disciplined experimentation with generative, AI-powered insights.So, rather than expect GenAI tools to create content to replace creatives, creatives are tightly defining the experiments and parameters for the GenAI tools to operate in. Using GenAI to perform the grunt work of generating hundreds of backgrounds for various-sized banners frees the creatives up to do more strategic work like creating behavior-based, personalized engagement programs.Road To SuccessInstead of pushing out that Explore GenAI task, marketing leaders need to plan for people first, starting now. What does success look like? By working backwards from success, CROs and CMOs alike can build a team of sales, creatives and marketing ops to define exactly what the GenAI tool will create. Now the CRO has an effective measurement framework in place before buying any new tools.