Wouldn’t it be easier if improving an ad didn’t mean starting from scratch?

With Visual Recommendations, you don’t have to start all over again with a blank page. Neurons now generates visual examples and new directions to help you optimize ads fast so you can move forward with confidence.

From Insight to Action — Instantly

Neurons has always helped you understand what works in an ad.

Now, with Visual Recommendations, Neurons actually shows you what stronger alternatives could look like.

Visual Recommendations are designed to help you go beyond testing ads and get concrete visual examples and new directions to see what stronger alternatives could look like.

‍

‍

Meet the Wildcard

Sometimes, optimization isn’t about incremental improvement.

Sometimes it’s about rethinking the direction.

That’s what the Wildcard is for.

The Wildcard is not tied to a specific recommendation. Instead, it maintains your core product and branding elements while exploring a new creative interpretation aimed at improving overall impact.

Customers consistently tell us this is one of their favorite parts — because it sparks discussions that wouldn’t have happened otherwise.

‍

‍

What Visual Recommendations Are Used For

Based on early feedback and beta conversations, teams are using Visual Recommendations for three main things:

1. Seeing What “Better” Versions Could Look Like

Instead of written insights and recommendations, you see concrete visual alternatives. Click Generate and get them in just 2-3 minutes.

2. Exploring New Creative Directions — Fast

Explore different alternatives in minutes, and see new creative directions.

3. Aligning Teams Faster

Rather than long decks or subjective debates, you can say: “Here. This is what we mean.”

And suddenly the conversation moves forward.

‍

‍

What Makes This Different from “Just Using AI”?

Visual Recommendations are not just an image generation tool. Behind the scenes, Neurons runs a structured, multi-step process:

1. Text Recommendations: We start with predictive insights about attention, engagement, memory, and clarity.
2. Translation Layer: Those recommendations are translated into structured prompts designed specifically for visual interpretation.
3. Variant Generation: The system explores multiple interpretations — combining recommendations individually and in groups.
4. Quality Control Agents
   
   • A variation checker ensures real, visible change.
   • A judge verifies the recommendation is meaningfully implemented.
   • Predictive models evaluate expected impact.
5. Curation Based on Impact: Only examples expected to increase the Neurons Impact Score move forward. The rest are filtered out.

With each image generation, you’ll get up to 5 visual recommendations that show optimized alternatives, and sometimes a Wildcard that deliberately breaks from the given creative direction to spark new thinking.

Each time you run a generation, the system explores dozens of visual variations, checks them against predicted responses, and filters out the noise in just 2-3 minutes.

Try Visual Recommendations Today

You no longer have to choose between speed and effectiveness. Now, you can have both.

Go beyond testing ads. Start improving them with Neurons, today!

‍

‍