Could a collaboration product introduce itself by telling the story of how its own ad was made?
The interesting part is not adding more computers to the frame or asking the creative team to praise a tool. The work we are watching becomes a goal inside the story. Someone proposes an idea, someone develops it, people discuss a choice, and the result eventually appears before us. The product participates in the process of explaining itself.
This self-referential approach can be called metanarrative, but the label matters less than the creative decision. Does understanding how the work came together require understanding something the product did? If not, extensive behind-the-scenes footage may simply produce another portrait of a team.
An ad can also tell a story of use
On its Slack project page, production company Sandwich describes an early commercial filmed documentary-style in its office, with team members playing themselves. Its account emphasizes features grounded in people’s interactions rather than dense technical detail and interface footage. That is the producer’s written description, not a shot-by-shot observation made for this article.
A later remote production made the self-reference particularly direct. In Sandwich in a Lunchbox: socially-distant ad production, published on Slack’s official blog on 22 May 2020, Sandwich founder Adam Lisagor describes collaborating from home through Slack to make a Slack ad about that experience. The article also recalls their 2014 collaboration. These are historical creative accounts, not a description of Slack’s current capabilities or an independent effectiveness study.
The creative opening is narrow but useful. The result viewers see and the process the brand wants to explain can belong to the same undertaking. The work is both the communication and the thing people inside the story are trying to complete.
That differs from ordinary behind-the-scenes material. A making-of film might explain lighting, set construction or rehearsal. A process-led brand story asks why these people could create this work together. The first examines production; the second turns a product-relevant part of the collaboration into the story viewers need to understand. Both can be valuable.
Give the product a role in one specific creative difficulty
Consider a fictional collaboration-software commercial, not a reconstruction of the Slack campaign or an ONCE client project.
Two illustrators working in different places are making a picture of life by the sea. One draws the people; the other will develop the setting. The difficulty need not involve a dramatic outage or a desk buried in discarded paper. It could be a small uncertainty: is an empty area unfinished, or deliberately reserved for the sea?
If the script jumps straight to a conversation in the software and then a finished picture, the product remains an insert. A more revealing moment is the second illustrator preparing to fill the space, encountering the first person’s explanation, then changing the approach. What matters is not the number of clicks. It is how one person’s intention reaches another.
This concept assumes that the product can carry an explanation associated with the artwork. The actual function and where that information appears must match the real product. Do not invent annotation, versioning or automatic synchronization simply to make the plot work.
Nor should the team become helpless without the tool. The illustrators still contribute judgment, craft and taste. If the product helps preserve and continue an idea, explain that role. It need not also create every idea, repair every relationship, improve efficiency and guarantee the final quality.
The ending need not remove all disagreement. They might still debate the color of the water while now understanding the purpose of the empty space. The product has helped resolve one misunderstanding, not made collaboration permanently frictionless.
Give the work, people and interface different jobs
One useful possibility is to show the finished illustration first. Only later do viewers discover that the sea they admired was once a space someone might have misunderstood. Revisiting the process changes our understanding of the picture’s creation, not its appearance.
Alternatively, begin with the making and let viewers anticipate the finished work. In that version, establish the goal and why it matters early. If the task remains unexplained until the end, messages and discussions can become disconnected operations. There is no mandatory reveal order; choose the one that makes the product’s contribution understandable.
An interface needs a readable relationship, not a complete tutorial
People can convey hesitation, misunderstanding and decisions. The interface explains how information reaches someone. The finished work gives the change a visible destination. With those roles separated, one screen recording need not carry emotion, mechanism and the entire feature catalogue.
In the fictional illustration scene, a detail of the drawing, an explanation of the empty area and a changed brushstroke might establish the relationship. Whether to show the full application depends on what viewers need to identify. Label conceptual interfaces and reenactments rather than allowing them to resemble records of actual use.
Dialogue need not name a feature. “I thought this part wasn’t finished” points to a specific uncertainty in the image; “we need efficient collaboration” does not. If the information that follows changes the next action, narration need not explain all the product’s significance.
Form should also serve the brand. Our Bookbook parody discussion considers how borrowing a familiar launch format can change the way a product is evaluated. Process-led storytelling takes another route: the work’s own formation supplies material for explaining the product. The approaches can coexist, provided a clever wrapper does not become the entire idea.
A polished film does not prove the product did everything
“We made this with our product” is a statement that needs to be true, not a reusable creative flourish. Uploading a finished file once should not imply that the product shaped the whole production. Scripted reconstruction, performed scenes and conceptual images must also be distinguished from an actual account of use.
Even genuine use does not let an attractive film establish time saved, errors reduced or suitability for every team. The creators’ experience, other conversations and post-production may all contribute to the result. A commercial can clarify a relationship of use; numerical effectiveness claims require their own support.
That distinction also matters for AI content tools. A work can show a confirmed sequence of generation, selection and human revision. It should not turn multiple attempts into an unexplained first-try success or assign a person’s final repairs to the model. This is a question of honest presentation, not an assessment of any untested tool.
Some products are better explained another way. Security, reliability or complex deployment may require different demonstrations and supporting material. If making the ad does not engage the value viewers need to understand, self-reference is not a reason to force it into the story.
For a new B2B brand video, bring a real episode of collaboration, the actions the product can genuinely support and an outcome worth seeing to ONCE VISUAL. The first question is not how lively the behind-the-scenes footage should be. Once viewers understand how this work came together, will they understand more accurately what the product contributed?
Sources checked on 16 October 2026. This article uses Sandwich’s project text and the author account on Slack’s official blog; embedded films were not watched and marketing outcomes were not verified. The illustration scenario and original AI cover are fictional concepts, not client work, an actual software interface or a product test.
