An AI tool that writes an ad gets plenty of attention. A tool that takes finished ads and turns them into a client presentation may solve a more immediate problem for a busy marketing team.
At BlueStar’s VARTECH 2026 event, James Korte, director of marketing at BlueStar US, described an internal application that does exactly that. Placing a series of social ads into realistic layouts and assembling them in PowerPoint used to take his team 30 to 45 minutes. BlueStar’s homegrown replacement for a paid tool called Ad Mockup Studio now puts the ads in context and exports the presentation with one click.
“I hated doing that,” Korte said of the manual version.
That example captured a recurring theme of “Blending Creative Content with Modern Marketing Tools,” a BlueStar Academy session hosted by John Martin, senior content marketing manager at BlueStar US, and featuring Korte and Bre Malott, a digital marketing specialist at BlueStar Europe: Let AI handle the repetitive work, and keep people in charge of the decisions that make marketing effective. The panel summed it up with a familiar line from AI’s early days – people want AI to do the laundry, not write the poetry.
For MSPs working with limited marketing resources, the session offered several practical ways to apply that approach.
Start with a task that eats time
Korte said BlueStar has built roughly 10 internal tools in 2026 using vibe coding – AI-assisted software development in which users describe an application in plain language, then work with an AI tool to build and refine it. His platform of choice is Lovable; Replit is another popular option.
One example is Brief Builder. It takes information from a standard project intake form, runs it through BlueStar’s guidance on what makes good marketing and produces creative-brief-style deliverables with starting points for headlines and copy. Korte’s favorite layer: The team instructed the system to channel a handpicked group of great copywriters and marketers from history. He called the result “a creative brief on steroids.”
Another tool pulls data from HubSpot, LinkedIn, ClickUp and another platform into a single dashboard. It answers a request BlueStar CMO Dean Reverman had made for years: one view of campaigns and email subscriber activity to bring into board meetings. Korte built it in a matter of hours, largely by plugging in API keys.
He also uses a Claude project to generate quarterly bonus documentation for the seven or eight people he manages. Team members submit raw notes on what they accomplished, and the tool selects the most significant items, weights them and produces a formatted, branded document. “Of course, I check it,” Korte said. “It saves me hours.”
The common thread is a defined problem. Each tool addresses work the team already understands: preparing a brief, packaging ads, gathering information from separate systems or documenting results.
An MSP can use the same starting point. Identify a recurring marketing task, document the inputs and desired output, and ask whether a small tool could remove the manual steps. Monthly campaign reports, QBR slide preparation and case study outlines built from project notes are likely candidates. Count the time spent reviewing and correcting the output as part of the cost – when the panel asked which marketing task takes too much time, one attendee’s answer was “correcting AI.”
Take content strategy out of the slide deck
A content strategy can lose momentum after the meeting where it’s presented. Korte described the familiar pattern: He builds a strategy deck defining pillars, themes, formats and audiences, walks the team through it, “and then life goes on.” Before long, people forget what the pillars were.
“One of the reasons that I’m so infatuated with vibe coding right now … is because I feel like it’s useful for taking strategy out of a PowerPoint slide and putting it into a tool,” he said.
BlueStar’s tool uses what the company has told it about its audiences – how many readers care about data capture versus point of sale and payments, for example – to recommend article topics and score their relevance. Once a writer accepts a topic, the tool suggests people to interview, questions to explore and a visual blueprint for the finished piece, from the hero image and bulleted agenda to scannable section formatting.
Korte cited an article in progress by BlueStar Academy’s lead trainer about scaling a point-of-sale business from deployments of a handful of locations to 50 or 100. The tool organizes the assignment, but the trainer still conducts the interviews, distills what he learns and writes the article. A designer then uses the blueprint to format it.
For an MSP, the value is consistency. Define priority customers, the business problems the company solves and the services it wants to explain, and those decisions can drive an editorial workflow that keeps topic selection on track. A topic score is only as useful as the guidance behind it, though. People still need to decide whether a proposed article answers a real customer question and supports the company’s goals. As the panelists noted, AI can’t decide who you’re selling to or which direction your company is rowing the boat.
Give AI your voice and brand standards
Malott described giving ChatGPT her speaking style, words she likes and words she wants to avoid. “I want it to sound like things are truly from me,” she said.
Martin, a writer at heart, described a research routine built on two tools. He runs the same prompt in Copilot and Gemini, asking for ideas, statistics, sources and how current the information is. The responses often differ significantly, so he merges the useful material and rewrites it in his own voice or BlueStar’s. “At the end of the day, I want it to sound like it’s from a human being,” he said.
For MSPs, that suggests a practical role for AI in organizing research and producing a first pass. Checking the underlying sources and adding the company’s own expertise remain part of the job. Agreement between two AI responses isn’t verification.
The same principle applies to visuals. “I don’t think the designs are bad. I think they are cluttered,” said Korte, who has spent more than 20 years designing by hand. AI-generated graphics tend to cram in copy, icons and the same script-style fonts, producing a default look audiences now recognize. The fix is to give the tool references, such as a screenshot of a design, color scheme or typography you like. Malott suggested prompting the AI to treat an ad like a billboard: a big, bold message, less text and a clean layout. As the panel pointed out, much of a designer’s job is removing things from the page.
Korte said Claude gets BlueStar’s slides about 80% of the way there when he points it at a folder containing the company’s fonts, colors and graphic elements. When it misuses an element, he uploads a screenshot showing correct usage, and the next result comes closer. An MSP can do the same by gathering approved logos, colors, fonts and examples of effective work in one place, along with guidance on audience and tone.
Settle governance before tools spread
BlueStar’s official policy, the panel noted, is that Copilot is the only AI tool approved for sharing company data – even as the marketing team experiments with ChatGPT, Claude, Gemini and Lovable for other work.
That mix creates its own problem. Korte’s 15-person U.S. marketing team builds in different tools, so a useful assistant created in one person’s account can’t easily be shared. BlueStar plans to pursue a team account, and Korte described another option under consideration: an internal app built on an AI provider’s API, so the whole team logs in to one tool and the company pays for usage instead of individual subscriptions.
Ownership matters, too. Korte acknowledged the security questions vibe-coded apps raise and said he is working out how to move one of his projects off Lovable, using its GitHub integration and a copy of the database. “I don’t plan on not paying for Lovable, but I don’t want to be tied to it either,” he said.
MSPs will recognize those issues. The same customers experimenting with vibe-coded tools and personal AI accounts will need help with access controls, data handling, API key management and ongoing maintenance. An MSP that has worked through those questions for its own marketing team is better positioned to answer them for clients.
Keep review in the workflow
AI also helps with troubleshooting. When a colleague reports a broken HubSpot workflow, Malott now shares a screenshot with AI to pinpoint the missing step, instead of tracing the problem from the beginning.
Finished work still needs inspection, though. Korte recalled an AI-generated hero image for a healthcare research article showing a patient in a hospital bed and a nurse nearby. The nurse’s ID badge carried the patient’s photo – a mistake BlueStar didn’t catch until a friend at another company emailed a screenshot. “We’ve got to look more closely at those,” Korte said.
For MSPs, review should cover facts, brand details and the small visual elements that can undermine credibility.
The session’s strongest lesson was practical: Choose a familiar task, use AI to remove the work around it, and keep people accountable for the result. Skepticism has a place in that process. When an attendee admitted to still being skeptical near the end of the session, the panel’s reply was simple: Good. BlueStar’s examples give MSPs a place to begin, especially when repetitive production work crowds out time to understand customers and create content worth reading.
