Business strategy

Halo AI: How Dillion Verma Repeats One Video Format

Dillion Verma explains Halo AI's repeatable video format and a payment choice: 29 of 30 creators picked a base fee plus view bonuses.

Listen to the audio analysis

Create your own

Written for builders and creators.

Most consumer app founders spend months testing dozens of different marketing "vibes" to see what sticks. But one founder just hit 300,000 dollars in monthly recurring revenue in only 45 days by doing the exact opposite.

Editor's note

Why this matters now

Dillion Verma says Halo AI, an AI image app, reached $300,000 in monthly recurring revenue 45 days after launch. His Superwall Podcast account focuses on repeating one format across a network of 85 creators: an absurd image, a quick app demo and a prank played out through text messages.

The structure stayed fixed while the creators, images and prank stories changed. Verma describes how successful videos became material for new scripts, the payment options he offered and what his creators selected. Those decisions explain how he organised the program behind the reported growth.

The source

What it says

Distilled from the original. The notes above and below are the editor's own.

The Demo Is Part of the Story

The opening image supplies the premise. Verma gives a prank about pretending to have cooked a parent's koi fish as an example. The video then shows the image being made in Halo AI before a text-message exchange supplies the reaction and payoff.

That sequence gives the app a necessary role. Without the generated image, there is no prank to carry into the conversation. The loading screen also repeats the product name while the viewer waits for the result.

The format can stay recognisable while the image and story change. Verma credits that repetition with 1.2 billion views over 120 days. The figure does not establish unique viewers or how many viewers became retained subscribers.

The Payment Plan 29 of 30 Creators Picked

Verma describes offering 30 creators a choice of compensation. 29 chose a $20 base per video plus bonuses at view milestones, including thresholds such as 100,000 and 1 million views. One chose payment per thousand views.

He says the milestone option remained more popular even when creators were told that the per-view arrangement could pay more. The combination offered a guaranteed base and visible bonus targets. The choice shows what this group preferred, without establishing why every creator chose it.

The operational difference is clear. A milestone plan asks whether a post crossed agreed thresholds. A per-view plan calculates the payment from its view count. Both still need tracking, a measurement window and agreed treatment of views after the initial payment.

This is a choice within Verma's creator group, not evidence that milestone pay is universally preferred or cheaper. Jay Raavi's Nomadtable network provides a contrasting account built around per-view payments.

Turning Successful Videos Into Briefs

Verma describes collecting videos that passed 1 million views, extracting their content and keeping it in local text files for an AI-assisted scripting workflow. New ideas could then draw on examples of the format the team wanted.

The valuable constraint is the input. A creator receives suggestions grounded in the team's own successful videos, rather than a generic request to write a viral script. The described workflow supplies relevant examples. It provides no performance comparison between generated and human-written scripts.

That matters because the workflow supports the existing format. It does not explain how Halo found the first successful format or guarantee that the audience will keep responding to it.

The Loan Followed an Early Return

Verma says an early return of roughly 5 to 6 times creator spending led him to take a personal loan of about $100,000 to expand the program. He also describes margins around 50% during the period discussed.

Those are his financial estimates. Assessing the return on a larger campaign would also require subscriber retention and a full cost breakdown. More views produce more revenue only while the relevant conversion and payment behavior hold.

The same limit applies to future paid advertising. Creator economics do not show what it would cost to reach a new audience through an ad auction.

Editor's note

What to do with this

Halo's useful contribution is a concrete production system: the product demonstration enables the prank, successful examples guide the next scripts, and the payment offer sets visible targets for creators.

Those details are transferable questions for a creator program. The reported revenue and loan decision belong to Verma's case. Repeating the format can make production more consistent while leaving audience response, subscriber retention and the return on the next dollar uncertain.

Explore the Superwall Podcast creator marketing playlist.

The original

0 to $300k/mo in 45 days with my ai app (just copy me)

The Superwall Podcast · 29 March 2026

All posts