A Worked Example from Liganova MaSH!: One Brief, Five Spud Touchpoints
By Polly Barnfield, OBE, CEO of Maybe*
A real client social brief, from inbox to delivery, showing where Spud picks up the work and where the human team picks it back up. Theory is cheap. Here is the iceberg in motion.
Every edition of Scaling Up With Agents includes a worked example, because architecture diagrams and capability lists are not the same thing as a system doing the job. This example follows a single creative brief through the Liganova MaSH! workflow, and shows where Spud touches the work and where the human team takes back over.
The scenario is a typical social brief for a luxury global brand. New launch content for a regional market, multi-asset, two-week turnaround, with the brand standards that come with the account.
Step 1: Resource planning. Three phone calls become one message.
The brief lands. The producer needs to know who is available, what the skill mix is, and how to staff the team for the next two weeks.
In the old workflow, that is a round of Teams messages and phone calls. Is this designer free Friday? Can I borrow that motion person? Do we have the regional language coverage we need? Half a day, easily, before the work itself can start.
In the new workflow, the producer asks Spud in Teams. Spud queries Float, returns the available designers and motion artists with the right skill profile, books their time, and confirms the booking back into Float. One message. Done.
“When we start this client we want X amount of designers, this skill set, this time period. And, yes, you can have the time please. Three phone calls or Teams messages become one.”
Step 2: File finding. Natural language, not folder navigation.
The team needs reference assets. Previous regional launches, brand guideline updates from the last quarter, the latest CI documents.
In the old workflow, that means navigating Box, hoping the file naming is consistent, asking colleagues if they remember where something lives. The cost compounds across the agency every day.
In the new workflow, the team asks Spud. Natural language. Pull the regional launch deck from Q2 and the latest CI update for the European market. Spud surfaces the files in Teams. Work continues.
Step 3: Synthetic audience validation. Stress-testing before the expensive part begins.
The creative team has an early concept. In the old workflow, validating that concept means waiting for client feedback, or running expensive research, or going to the audience after launch and hoping for the best.
In the new workflow, the synthetic audience capability lets the team test the concept against a synthetic version of the target audience before any production money is spent. They can stress-test the creative idea, segment the content into specific niche audiences for relevance, and refine before the brief gets expensive to change.
This is the capability that does not just save time on the existing process. It changes what the agency can do at this stage of a brief at all.
Step 4: Brand compliance. Real time, in the design tool.
Designers begin production. In the old workflow, brand compliance checks happen at review stages, with senior designers and brand stewards spending significant time on CI checking before anything went out. Multiple eyes, multiple loops, all of them billable hours that the client did not value because they were grunt work, not creative work.
In the new workflow, the brand compliance agent checks work in real time against the client’s CI as it is created. It flags issues early. The senior designers and brand stewards are still in the loop, but they are checking the agent's output rather than doing the line-by-line work themselves.
The result is faster turnaround, lower review burden, and CI standards that hold consistently across 2,500 pieces of content a year.
Step 5: News on demand. Context without context-switching.
Throughout the brief, the team needs to stay current on what is happening in the regional market, with the client, and with the wider category. In the old workflow, that is browser tabs, email digests, and the slow erosion of focus that comes with hopping between aggregators.
In the new workflow, any team member can ask Spud for relevant news, sliced by topic, client, or brief, delivered directly into Teams. The information arrives. The work does not break.
What gets handed back to the human team
Notice what Spud does not do in this example.
It does not write the creative. It does not decide the visual concept. It does not pitch the strategy. It does not have the conversation with the client. It does not make the judgement calls about brand voice, cultural sensitivity in a regional market, or whether a campaign idea is brave enough.
Those are the parts of the job the team was hired to do. And they are now the parts of the job the team has time to do well, because Spud has taken the resourcing calls, the file hunting, the early validation, the line-by-line CI checks, and the news monitoring off their plates.
“All the administrative tasks that used to be billable time are billable again. The same-sized team that can now service more clients and even better.”
That is the iceberg principle from the previous post made operational. One Teams chat. Five touchpoints across one brief. A team that ends the two weeks having done the work they were trained for, not the work the system kept asking them to do instead.
This is part 3 of 4 in the Liganova MaSH! edition of Scaling Up With Agents. The final post in this edition is about the leadership question that runs underneath all of this. How do you measure ROI when the return is mostly qualitative, and what does it take to lead an AI deployment without falling into the trap of cost-cutting framing?
Hear James walk through the Liganova MaSH! build in his own words on the FutureWeek podcast. Listen here.
Want to map your own workflow to find the Spud touchpoints? Book a call with the Maybe* team.
