Answers to the questions leadership actually asks.
Custom reporting app with your metrics, drill-downs, scheduled exports. Built on your data, not Looker's idea of it.
Who this is for
CFO or COO who needs answers BI tools don't give — Looker too rigid, Excel exports stale, and custom metrics nobody in engineering has bandwidth to build.
The pain today
- Looker/Tableau seat costs climbing without matching utility
- Excel exports stale by the time Monday staff meeting runs
- Custom metrics requiring an engineering ticket that never ships
- No single source of truth — each team reports differently
- No way to drill from executive summary into transaction detail
The outcome you get
- Executive dashboard with KPIs owned by leadership, not engineering
- Drill-down from summary metrics to transaction-level detail
- Scheduled email reports to leadership and board
- Cross-system data — CRM + billing + product + finance in one view
- Historical trending and forecasting for quarterly planning
Metrics architecture (source of truth, cubes, caching)
Custom reporting lives or dies on data architecture. Three patterns. One: direct query against operational DB — simple, fast for small data, stresses production at scale. Two: ETL into a warehouse (Snowflake, BigQuery) — clean separation, adds infrastructure, adds latency (usually acceptable). Three: OLAP cubes (Cube.js, Metabase's interpretation) — abstraction layer between raw data and reports, handles aggregations efficiently, enables self-serve metric definitions. For most mid-size companies, pattern two (warehouse + cached queries) is the sweet spot. Warehouse holds truth; custom dashboard hits a cached materialized view; dashboards feel fast without pounding production. I architect this in week 1; everything downstream depends on it.
Chart libraries I trust
Different charts for different jobs. Time series and trend: ECharts or uPlot, both handle dense data. Cohort analysis: custom D3 or dedicated cohort library. Funnel: ECharts or a funnel component in Nivo. Map: Mapbox for geo data. Table: AG Grid for anything beyond 100 rows (sorting, filtering, grouping, export). I avoid all-in-one dashboard frameworks (Grafana, Redash, Metabase) when the reporting needs are business-metric-specific rather than infrastructure-monitoring — those tools are strong for engineering metrics, weaker for executive business reporting. Custom built-in components win when the metrics are central to the business.
Export and scheduled email reports
Executives often consume data as an email every Monday morning, not in a dashboard. Scheduled reports: configurable cron (daily, weekly, monthly), configurable recipients, rendered to PDF or HTML email. Content: executive summary paragraph, top 5 KPIs with week-over-week and month-over-month change, drill-down link to dashboard. Board-facing reports (quarterly) with deeper analysis and commentary space for CFO. Email templates customizable without engineer involvement. For extra-sensitive reports (material financial data), distribution via authenticated portal only with email notifying of new report.
Case studies: Cuez and Imohub
Cuez's performance optimization (3s → 300ms API, 40% infra cost cut) gave the ops team real-time visibility into platform health. Imohub's admin (120k+ records, <0.5s query response, 70% infra cost cut) gave operators fast search and reporting on property data. Both projects share the reporting principle: operational visibility matters as much as operational throughput. A system running well that nobody can see running well still produces daily anxiety. Executive reporting on the same principle — leadership needs to see the metrics that matter, fast, with drill-down when they ask follow-up questions.
Pricing
Custom reporting dashboards fit the Applications Standard tier at $3,499/mo for typical executive dashboards (5–15 KPIs, 2–3 data sources). Pro at $4,500/mo for multi-department or enterprise reporting (deeper data integration, RBAC by department, advanced scheduled reporting). First-version timeline: 4–6 weeks. Subscription continues through iteration — reporting needs always evolve as leadership learns what they actually want to see. 14-day money-back, cancel anytime, Work Made for Hire. Warehouse costs (Snowflake, BigQuery) are separate and billed to you directly.
When Looker or Metabase is enough
Looker, Metabase, and Tableau cover the 80% case well. Custom becomes worth building when your metrics definitions are central to the business and specific to your domain (SaaS unit economics with specific definitions, subscription cohort analysis with your exact churn model, marketplace take-rate measurement with your exact attribution), when executive UX matters more than analyst self-serve (CEO isn't dragging columns in Looker Explore), or when you already own the warehouse infra and want a front-end the team actually uses. I'll tell you in the first call which direction fits. Custom is a commitment; off-the-shelf is often the right pragmatic answer.
Recent proof
A comparable engagement, delivered and documented.
Rescued a slow API that was blocking user growth
Refactored the backend architecture, making the system far more responsive and scalable for the growing user base.
Frequently asked questions
The questions prospects ask before they book.
- Do I need a data warehouse?
- For small data (under 100k rows per source, simple metrics) you can query operational DB directly. Above that, a warehouse (BigQuery, Snowflake) provides better isolation and aggregation performance. I architect based on your current scale plus expected growth — warehouse when it's justified, direct query when it's enough.
- Can leadership drill from summary into transaction details?
- Yes — every KPI drills down to contributing records. MRR drills into the specific subscriptions. Pipeline value drills into the specific deals. Cost metrics drill into the specific expenses. Drill-down respects RBAC — CFO sees everything, department head sees their department.
- How fresh is the data?
- Configurable per metric. Real-time (operational-DB-direct) for critical live metrics. Hourly refresh for standard dashboards. Daily for heavy analytical queries. Most executive reporting is fine with 15-minute or hourly refresh — dashboards don't need to update faster than leadership can digest.
- What about predictive and forecasting?
- Basic forecasting (linear regression, seasonal) built into standard tier. Advanced (ARIMA, Prophet, ML-based) available on Pro tier or as a follow-up engagement. For serious forecasting accuracy, integrate with dedicated tools (Mode, Hex, or custom Python pipelines) rather than rebuilding forecasting logic from scratch.
- Who maintains the dashboard long-term?
- During the subscription, I maintain and evolve it with leadership feedback. If you cancel the subscription, the dashboard is yours (Work Made for Hire) — another engineer can maintain it because the code, infrastructure, and documentation are clean and handoff-friendly. No lock-in.
Ready to start?
Tell me what you need in 60 seconds. Tailored proposal in your inbox within 6 hours.