I turn complex, data-heavy SaaS workflows into interfaces that feel simple.

Bluumly began as a tool to unify social media marketing, but a solid foundation wasn't enough. We noticed a recurring challenge for our users: social media success wasn't translating into business success. We set out to solve this by asking: how can we bridge the gap between creative output and revenue? This question became the catalyst for our innovation, transforming Bluumly from a scheduler into a growth-focused ecosystem.

"Creating content is easier said than done, and for small shops, the value just wasn't there yet."

While Bluumly succeeded in making content creation easy, our users were left with a critical question: how is this contributing to my bottom line? We realized that ease of use wasn't enough to sustain a subscription; we needed to provide proof of value. The attribution framework was conceptualized as a direct response to this user pain point, providing the transparency needed to link creative efforts to revenue.

After iterating on the lo-fi designs, I fed the wireframes directly into Figma Make's AI to accelerate the transition to high fidelity. This saved significant time in determining layout and visual hierarchy — areas that often slow the early stages of high-fidelity work. Once I was satisfied with the AI-generated output, I imported it into the main Figma file and refined each screen to align with Bluumly's branding and design guidelines, ensuring the final designs felt native to the product.

Building a bridge between social content and commerce required more than just UI — it required a robust logic layer. I worked closely with our data scientist to define a Confidence Scoring System that could accurately attribute revenue to creative assets. We developed a multi-factor algorithm that analyzed clicks, orders, and conversion rates to assign a specific score to each post. By calculating the time-to-purchase window, we were able to filter out noise and provide small business owners with a clear, honest view of their marketing impact.

I then transitioned this into a high-fidelity interface focused on three areas: an at-a-glance dashboard displaying attributed orders and revenue metrics; a view-switching system for both post-level and campaign-level analysis; and a weighted UI hierarchy that highlights Confidence Scores to distinguish direct conversions from assisted sales. By utilizing shadcn/ui and reusing established components, I ensured the new attribution screens felt native to the existing Bluumly ecosystem — delivering not just screens, but a functional roadmap that respected our developer's timeline.

Bluumly was still early-stage, so this project wasn't about proving lift against existing usage data — it was about making a genuinely hard problem legible. Attribution modeling involves noise, multi-factor scoring, and a time-decay window that's difficult to explain even to other designers.

The real measure of success was whether a small business owner with no data background could look at a Confidence Score and immediately know what to trust — turning a backend algorithm into something a non-technical user could act on with confidence.