Aura - AI Portfolio Manager
Aura empowers media managers to simulate strategic "what-if" scenarios, enabling informed, high-impact decisions for proactive campaign planning.
Problem Statement
A delightful user experience is to be designed for marketing leads in a Food and Beverage organization so that they can simulate and understand their budget utilization across multiple marketing channels (say, TV advertisements, radio, digital, email, magazine, and many others). The marketing lead is expected to create multiple scenarios and compare them regularly with the base scenario to derive insights and plan the expenses accordingly.
Food and Beverage organization: PepsiCo
A delightful user experience is to be designed for marketing leads in a Food and Beverage organization so that they can simulate and understand their budget utilization across multiple marketing channels (say, TV advertisements, radio, digital, email, magazine, and many others). The marketing lead is expected to create multiple scenarios and compare them regularly with the base scenario to derive insights and plan the expenses accordingly.
Channel
TV advertisements
Radio
Digital Media like Instagram, Youtube, Facebook
Magazine
Instore promotion
Influencer marketing
Basically
The marketing lead is expected to create multiple scenarios and compare them regularly with the base scenario to derive insights and plan the expenses accordingly.
What do we know...
We are in Food & Beverages organization.
The marketing leads needs to create base scenario.
Compare base scenario with Alternate scenarios created.
Based on the comparison we need to make business decisions.
What do we don’t know
What are the primary metrics important for the F&B marketing?
Since F&B is different from others what are the most imp factor for them. what they want to achieve?
How is the base scenario created?
How do we decide which alternate scenarios have to be created?
How is the dashboard to be designed?
Persona - Ananya Mehta
Meet Ananya Mehta, she is a Senior Portfolio Marketing Lead at a large food and beverage company in Delhi NCR. She manages a portfolio of beverages, including soft drinks, juices, bottled water, energy drinks, and ready-to-drink coffee.
Redefined Problem Statement
Managers lack an intuitive way to simulate and compare different budget scenarios, leading to decision paralysis during planning and a lack of defensible evidence when justifying final decisions to stakeholders.
Needs
Media managers need a unified command center to track live campaign performance, model and compare dynamic "what-if" budget scenarios against a baseline plan, and continuously track planned versus actual outcomes to validate strategic decisions.
Current User journey and Problems
Trying to answer the unknown
What are the primary metrics important for the F&B marketing?
Marketing in the Food and Beverage (F&B) industry differs fundamentally from other sectors because food is a sensory, emotional, perishable, and high-frequency purchase driven by immediate cravings rather than long-term evaluation. Unlike tech or automotive brands that sell features or ROI, F&B marketing must capture "share of stomach" through craving-driven visuals, experiential packaging, and in-person sampling or micro-influencer taste tests to overcome consumer hesitation.
Additionally, because food products have strict expiration dates and low switching costs, marketing strategies must be hyper-localized, tightly synchronized with supply chains to move inventory quickly, and closely aligned with stringent health and labeling regulations.
How is the base scenario created?
A base scenario in F&B marketing is the organic sales level achieved without promotions or paid ads. To build it:
Clean Historical Data: Strip out past discounts, ad campaigns, and holiday surges to find true organic sales.
Account for Repeat Habits: Measure natural reorder cycles (e.g., daily coffee, weekly grocery restocks).
Adjust for Seasonality & Time: Factor in weather (hot vs. cold seasons), day-parts (lunch vs. dinner), and weekday vs. weekend patterns.
Normalize Shelf Life & Stockouts: Correct for lost sales caused by expired batches or out-of-stock items.
Control for Distribution: Set baseline numbers per store shelf, delivery radius, or channel.
Note: Assuming the current market is the base scenario.
How do we decide which alternate scenarios have to be created? or which parameters have to be simulated?
To choose alternate scenarios and simulation parameters in F&B marketing, focus on the factors with the highest uncertainty and business impact.
1. How to Decide Alternate Scenarios
Select 2–3 plausible future states anchored around key business questions:
Best-Case (Demand Surge): High campaign virality, favorable seasonal weather, and full supply availability.
Worst-Case (Supply/Cost Shock): Key ingredient shortage, spoilage/shelf-life bottlenecks, or price increases reducing demand.
Competitor/Macro Shift: Aggressive competitor discounting or a drop in delivery radius/traffic.
2. Key Parameters to Simulate
For a global FMCG/beverage enterprise like PepsiCo, simulations rely on Revenue Growth Management (RGM) and Marketing Mix Modeling (MMM).
Key parameters to simulate across alternate scenarios:
Price-Pack Architecture (PPA) & Elasticity: Simulating price changes across package sizes (single-serve cans vs. multi-pack PET bottles) and testing consumer price elasticity.
Trade Promotions & Retail Co-Investment: Depth and frequency of temporary price reductions (TPRs), end-cap displays, and modern-trade supermarket bundle deals.
Cross-Brand Portfolio Cannibalization: How pushing one product (e.g., Pepsi Black/Zero) cannibalizes or lifts sister brands (Regular Pepsi, 7UP, Mountain Dew, Lay's combos).
Media Mix Reallocation (Above-the-Line vs. Below-the-Line): Budget shifts between massive cultural sponsorships (IPL, Super Bowl, music festivals), TV GRPs, digital/quick-commerce ads (Blinkit, Instamart), and in-store POP displays.
Route-to-Market & Channel Execution: Volume distribution splits between General Trade (local kiranas), Modern Trade (hypermarkets), Quick Commerce, and Out-of-Home/Foodservice (QSR fountain partnerships).
Competitor Counter-Actions: Aggressive price cuts, retailer slotting wars, or marketing blitzes by direct competitors (e.g., Coca-Cola).
Weather & Regional Shocks: Temperature anomalies (e.g., unseasonal summer rain reducing cold beverage pull) and agricultural commodity cost spikes (e.g., sugar, aluminum, potatoes).
Note: While the primary objective is optimizing channel-level media budget allocation, the model also incorporates key external and market variables, such as competitive pricing, weather patterns, regional demand shocks, and route-to-market dynamics, to deliver robust, real-world simulations.
User Journey
Lets try creating wireframe with AI.
Since we have a clearer picture now, I will consolidate all the information we've gathered and share it with Gemini. Even though our conversation has been in bits and pieces, here is a single prompt summarizing everything:
Prompt
Act as a Principal UX/UI Designer and Design Systems Architect.
Generate a complete, modern, presentation-ready dashboard wireframe as a single, self-contained HTML file styled with modern Tailwind CSS (via CDN) and Lucide Icons (via CDN/Vanilla JS) that can be imported directly into Figma using the "HTML to Design" plugin or rendered as a 1920x1080 slide.
### Theme & Aesthetics:
- Enterprise F&B / FMCG Brand Analytics (Clean, executive, high-contrast, premium aesthetic).
- Background: Slate-50 (#F8FAFC), Cards: White (#FFFFFF) with subtle Slate-200 borders, rounded-2xl corners, and soft elevation shadows.
- Typography: Inter / Plus Jakarta Sans. Clear visual hierarchy (24px headers, 32px bold KPIs, 12-14px microcopy and data labels).
- Palette: Primary Slate-900, Indigo-600 (ATL/Brand), Cyan-600 (Digital/QC), Amber-500 (Trade/In-Store), Purple-600 (Simulations), Emerald-600 (Positive Growth), Rose-600 (Risk/OOS).
### Layout & Sections (1920px Canvas):
1. TOP NAVIGATION / HEADER:
- Left: Brand icon + Title "F&B Commercial Intelligence & Scenario Sandbox" | "Summer Hydration Q3 Campaign".
- Right: Filters (Region: National, Period: Q3 Rolling 12Wk) + Action Buttons: "[+ New Scenario]", "[Run Simulation]", "[Export to Figma/Deck]".
2. ROW 1: EXECUTIVE KPI SUMMARY (4-Column Grid):
- Card 1: Total Market Volume Share (34.2% | ▲ +1.8% vs YA | Sparkline trend + Target line).
- Card 2: Incremental Cases Lift (4.8M Units | ▲ +12% vs Target | Planned vs Actual progress bar).
- Card 3: Blended MMM ROAS (3.42x | ▲ +0.32x vs Baseline | Marginal efficiency tag: "Optimal").
- Card 4: Retail On-Shelf Availability (OSA) (95.8% | ▼ 4.2% Out-of-Stock Risk | Status Pill: "Healthy <5%").
3. ROW 2: CORE ANALYTICS & ECONOMETRICS (2 Columns: 8-Col Main + 4-Col Health):
- Left (8 Col): "Incremental Volume vs Media Spend (Media Mix Modeling)" -> Visual representation of stacked weekly spend bars (TV/ATL, Digital Video, Quick-Commerce Ads, Trade Sampling) overlaid with a smooth volume trendline and flight milestone markers.
- Right (4 Col): "Channel Saturation & Marginal ROI Curves (Hill Function)" -> Visual S-curves showing current spend position relative to diminishing return thresholds for TV (68%), Digital (84% - Near Saturation), and Trade Promo (45%).
4. ROW 3: VELOCITY, SHELF & COMPETITIVE SOV (3-Column Equal Grid):
- Card 1: "Quick-Commerce Velocity & Share" -> Platform breakdown (Blinkit 42%, Instamart 36%, Zepto 22%) with organic vs paid conversion and Cost/Incremental Unit ($0.44).
- Card 2: "Regional Shelf Health (OSA Heatmap)" -> 4-Zone Matrix (North: 96.8% Green, West: 94.2% Green, South: 89.5% Amber Risk, East: 95.1% Green).
- Card 3: "Share of Voice (SOV) vs Share of Market (SOM)" -> Clustered visual comparison (Brand: 38% SOV / 34% SOM, Competitor A: 31% SOV / 29% SOM, Competitor B: 19% SOV / 24% SOM).
5. ROW 4: INTERACTIVE SCENARIO SIMULATION & TRADE-OFF ENGINE (Hero Feature Panel):
- Header: "Scenario Modeler: Baseline Plan vs Alternate Allocation" with comparison toggle.
- Left Half (Levers/Sliders):
* Above-the-Line (TV, Radio, OOH): $3.2M [Slider at 55%]
* Digital Media & Social Ads: $1.8M [Slider at 70%]
* Quick-Commerce Retail Media (RMN): $1.4M [Slider at 85%]
* In-Store POS & Sampling: $0.8M [Slider at 40%]
- Right Half (Real-Time Impact Matrix):
* Projected Cases: 5.12M (+6.5% vs Base)
* Projected Net Revenue & Blended ROI: 3.58x (+0.16x)
* Risk Index: Low OOS impact
* Delta Summary: "Reallocating $200k from Linear TV to Quick-Commerce yields +140k cases before hitting saturation."
- Footer: "[Save Scenario B]", "[Compare Side-by-Side]", "[Export Defensible Deck]".
Deliver the entire working HTML/Tailwind wireframe code inside a single copyable block with semantic structure and modern layout styling.
What does AI design look like
I'm pretty comfortable with Google AI Studio, so I dropped the prompt in to build out the full dashboard. The output turned out great, all the key metrics and visuals are spot on. The Scenario Marketing Engine works just like I wanted, making it super easy to test different channel mixes and see how the numbers shift.
Note: Though I will really loved the result, I wanted to make a little adjustments. Simplify it a little
Figma Make Design
Login screen
The login screen is simple and follows the modern design, I used some reference from Pinterest and Dribbble, to get the UI right.
Dashboard/Landing page
I simplified the dashboard in Google AI Studio with a few quick adjustments:
Cleaned up the layout to make it less busy.
Added info icons with easy-to-understand explanations for each metric.
Moved "Run Alternate Scenario" into a popup modal to keep the main dashboard focused.
Simulator
The simulator popup features intuitive sliders to adjust marketing spend across channels, instantly updating the forecasted variance metrics in real time.
Figma make link
Password - Aura2026
Takeaway
As a UX designer, my job is to solve real problems for users, and the challenges faced by marketing media managers felt like the real deal. Even though I've been a UX designer for years, I didn't know much about marketing, so this assignment was definitely a challenge. I dove into learning marketing concepts, watched YouTube videos on different dashboards, and studied how managers forecast campaigns over and over. All that research helped me narrow things down to the solution we landed on.