Nearly every SaaS analytics tool on the market ships the same way: connect your billing data, and you get a fixed set of charts - MRR, churn, ARPU, a subscriber count. The charts are well-designed. The metrics are the right ones, in the abstract. And within a month, most founders find themselves staring at a dashboard that's technically accurate and still missing the one number they actually needed this week.
That's the gap between a fixed dashboard and a custom one - and it's the biggest structural difference between how most SaaS analytics tools work and how Chartsy is built. This post covers what that difference actually means in practice.
The Fixed-Dashboard Model
Most SaaS analytics tools - and every payment processor's native dashboard - are built around a fixed model:
- A product team decides which metrics matter for a typical subscription business.
- Those metrics become the dashboard - MRR, ARR, churn, LTV, ARPU, a trend line for each.
- You get some filtering (date range, plan, sometimes a saved view), but the underlying set of questions the dashboard can answer is fixed at however many chart types the product ships.
This isn't a design flaw - it's a reasonable default, and it's genuinely useful for the first 80% of what any SaaS business needs to track. The problem shows up in the remaining 20%: the question specific to your business, this month, that no vendor could have anticipated when they built the fixed set.
| Tool | Dashboard Model |
|---|---|
| Stripe's native dashboard | Fixed - revenue, subscribers, payments |
| Baremetrics | Fixed, well-designed - MRR, churn, LTV, ARPU |
| ChartMogul | Fixed core metrics + cohort views |
| Paddle's built-in analytics (ex-ProfitWell) | Fixed - MRR, churn, subscriber counts |
| Chartsy | Fixed templates to start, fully custom via AI from there |
What "Custom" Actually Requires
A dashboard being genuinely custom - not just filterable - requires three things most fixed-dashboard tools don't offer:
The ability to ask a question the vendor never anticipated. Not a filter on an existing chart, but an entirely new chart, built from a question specific to your business - a pricing migration, a discount cohort, a customer segment defined by a metadata tag only you use.
No dependency on knowing the tool's data model. If building a new chart requires understanding which field is called what and how metrics and dimensions can be crossed, most founders will never actually use that flexibility - they'll stay inside the fixed set because it's the path of least resistance.
A dashboard that reshapes itself over time. Your questions in month one of a SaaS business aren't your questions in month eighteen. A truly custom dashboard should be able to gain and lose charts as your business changes, not just be configured once and left alone.
How Chartsy Combines AI and Custom Dashboards
This is Chartsy's core bet: the two features that matter most - ChartsyAI and Custom Dashboards - aren't separate. They're the same mechanism. Every chart on a Chartsy dashboard exists because someone asked a plain-English question and saved the result. There's no separate chart-builder UI to learn, and no dropdown menu standing between a question and an answer.
"Show me MRR by plan for the last 12 months." "Which customers upgraded after we changed pricing in March?" "What's churn for customers tagged with 'enterprise' in Stripe metadata?"
Each becomes a chart in seconds. Save the ones worth checking regularly, and the dashboard grows into something no fixed template would ever produce - because it was built entirely from your specific questions, not a vendor's general-purpose defaults.
This also means the AI isn't a bolt-on chatbot sitting next to a fixed dashboard - it's the dashboard's construction method. Ask a question, get a chart, decide whether to keep it. Repeat, and the dashboard becomes fully yours.
What You Lose with a Fixed Dashboard (Even a Well-Designed One)
The specific question always requires leaving the tool. When a fixed dashboard doesn't cover what you need, the fallback is usually exporting data to a spreadsheet, or - for Stripe users specifically - writing SQL in Stripe Sigma. Both work, but both mean stepping outside the dashboard entirely for anything non-standard.
Metadata and tags go unused. Most SaaS founders are already tagging customers and subscriptions with useful context - acquisition source, company size, plan tier - directly in Stripe or Paddle metadata. A fixed dashboard has no way to expose that data unless the vendor specifically built a widget for it. It sits there, unused, because there's no path from "I tagged this" to "I can see this on a chart."
The dashboard doesn't evolve with the business. A pricing migration, a new customer segment, a discount code experiment - all of these create a temporary but real need for a specific chart. A fixed dashboard has no mechanism to gain that chart for the three months it matters and lose it once it doesn't.
Comparison at a Glance
| Fixed Dashboard (Baremetrics, ChartMogul, Stripe, Paddle) | AI-Built Custom Dashboard (Chartsy) | |
|---|---|---|
| New chart for a specific question | Not possible without a separate tool (SQL, spreadsheet) | Ask ChartsyAI directly, save in one click |
| Uses your Stripe/Paddle metadata | Rarely, unless the vendor built a specific widget | Yes - any tagged field becomes a filter/dimension |
| Requires learning the tool's data model | No, but also no flexibility beyond presets | No - you just ask in plain English |
| Adapts as your business changes | Requires waiting on the vendor's roadmap | Add or remove charts anytime |
| Starting point for a new account | Fixed default view | Choose a template or start blank - both lead to fully custom |
Frequently Asked Questions
Isn't a well-designed fixed dashboard good enough for most SaaS businesses? For the standard, universal metrics - MRR, churn, ARPU - yes, a fixed dashboard covers most of what any subscription business needs. The gap shows up specifically for business-specific questions: a pricing migration, a discount cohort, a metadata segment unique to your company. Fixed dashboards have no mechanism for those; custom, AI-built ones do.
Does Chartsy have any fixed dashboards at all? Yes - four starter templates are available the moment your first sync completes, so day one isn't a blank screen. But every template is a starting point, not a ceiling - any chart can be removed and new ones added via ChartsyAI at any time.
Do I need to know SQL to build a custom chart in Chartsy? No. Every custom chart is created by asking a question in plain English through ChartsyAI - there's no SQL, no chart-builder dropdowns, and no need to understand Chartsy's underlying data model.
Can a custom dashboard use data I've tagged in Stripe or Paddle metadata? Yes. Chartsy automatically indexes existing metadata fields, so tags like acquisition source, company size, or coupon code become usable filters and dimensions in any chart you ask ChartsyAI for.
What's the actual difference between this and Stripe Sigma's SQL-based flexibility? Stripe Sigma also lets you go beyond a fixed dashboard, but requires writing and maintaining SQL for every question. Chartsy's Custom Dashboards achieve the same flexibility - any question, not just the preset ones - through plain English instead of a query language.
Connect Stripe or Paddle and start building a dashboard that's actually yours →
Related: Custom Dashboards feature overview · ChartsyAI Visualization · Custom SaaS Dashboards: How to Build Exactly the Dashboard You Need with AI
Written by
Chartsy TeamThe Chartsy Team writes guides, product updates, and resources to help SaaS and eCommerce founders make sense of their metrics, without SQL or spreadsheets.
Chartsy