On a recent call, the head of marketing at a mid-sized B2B SaaS company walked me through everything his team was doing to fix the workflow problem.
They built a shared library in Notion. They published a voice guide for the product. They will use AI training to read and write twice. They hold monthly office hours where the team’s most successful AI users answer questions. The CMO personally wrote how to use the thoughtful AI for the memo and reminded everyone that the goal was volume.
Still, the line of work kept coming. Partially completed briefs that read like drafts of something better. The slide decks felt great on top and fell apart on the third shot. Brochure copy that briefly touched and missed the audience.
Workslop is an easy brand to name. Diagnosing where it occurs is difficult and resides in a different category of organization.
That’s where most of the workflow discussion stops
BetterUp Labs and Stanford research — the original September 2025 HBR study and a follow-up in January 2026 — put the numbers in perspective. 40 percent of employees experienced a work slop last month. Each event costs just two hours to clean. For a company with 10,000 people, that figures out to about $9 million a year, gone, to fix AI-powered work that was supposed to save time.
The number that hits me hardest comes from Asana’s State of AI at Work study. Only 19% of knowledge workers say they have clarity on what types of work AI should perform in their role. That figure explains all the data.
The biggest fix in the conversation right now is the one I’ve been on the phone with for the past six months. Leaders must demonstrate meaningful use of AI. Teams should set clear guard lines. Individuals must develop what BetterUp calls a pilot mindset.
Human AI output should reach the same level as human output alone. Greg Kihlstrom’s latest piece for MarTech extends the argument into a more marketing-specific space, telling marketing leaders to step up and define the handoff lines with IT, legal, and procurement.
All of this is fine. Nothing is wrong. And every part of it puts the burden on the same place: each promoter, each leader, each person’s attitude. That’s the layer many teams have pulled over in the last 18 months, and the real fix lies elsewhere.
Your customers are searching everywhere. Make sure it’s your product he appears.
The SEO toolkit you know, and the AI visibility data you need.
Start a Free Trial
Start with
When the system is broken
Workslop is what happens when individuals produce AI-powered work and there is no connective tissue between them.
In an active sales team, learning must move quickly. The content specialist calculates in the first week that this model requires a short and strong person to do anything useful. The designer finds out in the second week that this image tool is looking for product colors in hex, not in plain English. An email marketer discovers in the third week that the subject lines of the AI sound normal unless you feed them the last three subject lines that worked.
Each of those lessons is real. Each one was earned. In most sales teams I see, none of that information goes very far. The content expert doesn’t know what the designer was thinking. An email marketer doesn’t know what a content expert has learned. There is no place assigned where someone says, “This is what worked, this is how I got there, try it your way and tell me what works.”
What you end up with is a team of talented people, each running their own little R&D project in parallel. Each one gets better at his piece. Combined group output still produces workflow because none of one person’s learning gets passed on to the next person. When someone moves groups, that learning goes out the door too.
This is the part of the workflow problem that no one clearly states. It’s a learning and communication problem, and it’s not going to be solved with another training session or a sharp voice cue for the brand. It is solved by building an infrastructure that carries learning between people.
What AI communication optimization looks like
In my book “Hyperadaptive,” I call this connection layer the AI activation hub. A hub is a small group of people within an organization, virtual, physical, or both, whose job it is to keep AI capability flowing throughout the team in both directions.
It’s something different from a help desk, library, or AI ticket inbox. Those are the static collections you go to when you need to look something up. The hub is people whose job it is to actively transfer learning to the group.
A practical idea, in a marketing context. An active center does a few specific things.
- Stay current and improve learning: Hubs translate what’s new and what’s working into bite-sized, exclusive content that arrives in the workflow, more like a two-minute Loom in a Slack channel than a wiki page that no one opens.
- Holds office hours and brings people together: Hubs facilitate a live, active experience. A hub member pairs a marketer with AI fluency and a marketer with business acumen, and the resulting work is better than either of them could build alone.
- Maintains a usable information engine: When engineering firm iMBrace built theirs, it cut through its time searching for information. That’s a number to pay attention to. The repository is live, queried in natural language, and continuously updated by the Hub itself.
- Dimensions where AI does and does not benefit from its maintenance: Hubs track what’s working and report that pattern back to leadership. This is an AI marketing center for good job descriptions that are completely absent.
When was the last time your team built something that flows AI learning between members on purpose, rather than hoping it happens at a coffee machine?
New job market data suggests that marketing is starting to figure this out. According to Carilu Dietrich’s latest piece, senior AI marketing roles are growing rapidly under titles such as head of marketing AI, center AI marketing lead, and senior director of AI projects.
Postings of related GTM engineers on LinkedIn more than doubled in six months, from about 1,400 in mid-2025 to more than 3,000 in early 2026. Marketing creates a hub in real time and gives it a unique name.
The teams I see doing it the right way play the role well. They describe the hub leader’s job as facilitating learning, pairing people, and making the group smarter on purpose. That’s a different job description from “implementing AI standards and policing data quality,” which is where most of these roles come in now.
Where this is headed
The sales teams that will solve the workflow in the next 12 months will be the ones that build the connective layer that carries the learning between people, so if one salesperson calculates something, the whole team uses it by the end of the week. That’s the real fix. Efficiency follows. It always is.
