We’re a performance marketing agency managing cross-channel paid media for clients across Meta, Google, TikTok, Snapchat, Apple Search Ads, Reddit, Remerge, and RTBHouse. We need to build a fast-loading, client-facing reporting dashboard that pulls data from 8 ad platforms, normalizes it, and presents it in a structured weekly format with initiative-level and regional breakdowns.
The current reporting setup uses live API queries and is too slow (~20 seconds) with the level of blending required. We’re moving to a pre-aggregated data warehouse architecture to achieve sub-5-second load times.
Tech Stack
• Funnel.io (Agency Medium plan) - data extraction, normalization, and transformation
• Google BigQuery - data warehouse for pre-aggregated views
• Looker Studio with BI Engine - dashboard/visualization layer
• Cloudflare Access - client-facing vanity URL with authentication (e.g., clientname.neongrowth.com)
Scope of Work
1. Funnel.io Configuration
1. Connect 8 ad platform sources (Meta, Google Ads, TikTok, Snapchat, Apple Search Ads, Reddit, Remerge, RTBHouse)
2. Build transformation rules to normalize metric names across platforms (e.g., unify “Amount Spent,” “Cost,” “Spend” into a single field)
3. Create initiative tagging logic based on campaign naming conventions (e.g., BAU - Static, BAU - Search, BAU - UGC/Video, AI Coach, Family Plan)
4. Configure region grouping (US vs. International) based on campaign structure
5. Set up scheduled export to BigQuery (daily refresh)
2. BigQuery Setup
1. Create BigQuery project, dataset, and table structure for incoming Funnel.io data
2. Build SQL views for weekly aggregation at the channel level (Impressions, Clicks, Sign Ups, Cost per Sign Up, SUBs, CPSUB, SUB Conv Rate)
3. Build SQL views for weekly aggregation at the initiative × region level (same metrics, grouped by Initiative and Region)
4. Build a goals/targets view (joining a Google Sheet with monthly targets)
5. Configure BI Engine reservation for sub-second query performance
3. Looker Studio Dashboards
We need 4 dashboard tabs, modeled after a reference spreadsheet we’ll provide:
• Goals View: MTD and Month-End actuals vs. targets with % to Goal for Spend, Sign Ups, SUBs, and ROAS, broken out by US and International
• Channel Performance Weekly: Rolling 6-week table with columns for This Week through 5 prior weeks, plus % Change vs. Last Week. Rows grouped by channel (Meta, Google, TikTok, Snap, ASA, Reddit, Remerge, RTBHouse). Repeated for each metric: Impressions, Clicks, Sign Ups, Sign Up Conv Rate, Cost per Sign Up, SUBs, CPSUB, SUB Conv Rate
• Initiative Level Weekly: Same 6-week rolling structure, but rows grouped by Region (US/Intl) × Initiative (BAU - Static, BAU - Search, BAU - UGC/Video, AI Coach, Family Plan). Same metrics as Channel Performance
• Charts: Spend by channel with Cost per SUB and Cost per Sign Up overlays. Channel-specific trend charts (last 8 weeks) for Spend and Cost Pers for each of the 8 channels
Requirements
Must-Have Experience
• Funnel.io configuration and data transformation (not just Supermetrics)
• BigQuery - table design, SQL views, scheduled queries, BI Engine
• Looker Studio - pivot tables, calculated fields, multi-tab dashboards connected to BigQuery
• Performance marketing data (you understand what ROAS, CPA, CPSUB, and conversion rates mean in context)
Nice-to-Have
• Experience with Cloudflare Access or similar auth gating for embedded dashboards
• Prior work with multi-channel agencies (not just single-brand in-house)
• Ability to document the setup so our team can replicate it for additional clients
Deliverables
1. Fully configured Funnel.io workspace with all 8 sources connected, normalized, and exporting to BigQuery
2. BigQuery dataset with production-ready SQL views for all dashboard tabs
3. Looker Studio dashboard (4 tabs) loading in under 5 seconds with months of historical data
4. Brief setup documentation so we can onboard additional clients using the same template
Timeline: 3-4 weeks from kickoff to final delivery
Engagement type: Fixed-price project. We will provide platform access (Funnel.io, BigQuery, Looker Studio) and a reference spreadsheet with the exact reporting format.
Apply Now
Apply Now