Embedded Analytics: Multi-Tenancy, RLS, Tools & Pricing (2026)

Quick answer:

The embedded analytics platforms that natively support multi-tenancy and row-level security include Toucan, Luzmo, GoodData, Qrvey, and Explo. Most traditional BI tools (Tableau, Power BI, Metabase) require significant custom engineering to achieve proper tenant data isolation. Pricing for these purpose-built platforms ranges from $1,500–$5,000/month for most SaaS use cases.

Why "per tenant" pricing is the right lens for SaaS companies

If you're building a SaaS product, your cost unit isn't a named employee — it's a customer account. You might have 200 customers, each with 50 of their own users accessing your dashboards. Buying 10,000 individual viewer seats from a traditional BI tool isn't a viable model.

That's why SaaS teams should think in cost-per-tenant: what does it cost to serve one customer organization with analytics? Everything else flows from there — margin, packaging, upsell potential.

Understanding the full ROI picture matters here too. The platform fee is just one part of the equation.

Which embedded analytics tools natively support multi-tenancy and row-level security?

For SaaS companies, multi-tenancy and row-level security (RLS) aren't nice-to-haves — they're non-negotiable. Every customer tenant must see only their own data. The question is whether your analytics vendor handles this natively, or whether your engineering team has to build it.

What multi-tenancy means in embedded analytics: Each customer organization (tenant) gets an isolated view of the platform — their own dashboards, their own data, their own branding — without any data leaking across accounts. This is typically enforced through token-based authentication (JWT/SAML/OIDC) combined with row-level filters applied at query time.

What row-level security means: RLS restricts which rows of data a given user can see, based on their identity or role. In a SaaS context: a merchant using your platform should only see their own transaction data, not a competitor's. RLS enforces this at the database query level, not in the UI.

Here's how major platforms handle it:

How Toucan handles tenant isolation specifically: When a SaaS app embeds Toucan, the parent application authenticates the user and generates a secure token (JWT) encoding the user identity, tenant/organization ID, and optional filters. Toucan resolves this token server-side and enforces data isolation at query time — meaning no tenant can ever query another tenant's rows, regardless of UI manipulation. This happens without custom engineering on the SaaS team's side.

Why this matters for pricing: Platforms that don't handle multi-tenancy natively push the implementation cost onto your engineering team. A "cheap" $500/month tool that requires 3 engineering sprints to build proper RLS isn't cheap at all.

The four main pricing models — and what they cost per tenant

Per-user / seat-based

You pay for every end user who accesses analytics. Simple in theory, brutal at scale.

Tableau Embedded runs around $420/viewer/year. Metabase charges $12.50/external user/month on top of a $575/month platform fee. ThoughtSpot Pro sits at $50/end user/month — meaning 1,000 users across your customer base runs $50,000/month.

For a SaaS company with tenants that each have 30+ users, seat-based pricing compounds fast. If your average tenant has 40 users and you're paying $15/user/month, that's $600/tenant/month before a single feature add-on.

Capacity / usage-based

Platforms like Power BI Embedded charge for compute capacity rather than users. The entry-level A1 tier on Azure starts around $750/month. The model is appealing if your dashboards have low concurrent usage — but it's genuinely hard to predict, and can spike when adoption grows.

Qlik's premium plan starts at $2,750/month with additional data throughput fees. Domo uses a credit-based consumption model tied to data rows, refresh rates, and users — costs vary so widely that meaningful per-tenant estimates are nearly impossible without a custom quote.

Per-tenant / flat-tier

Some platforms purpose-built for embedding price directly by tenant count or offer flat tiers that encompass unlimited end users. This is the model most aligned with SaaS unit economics.

Purpose-built embedded analytics platforms like Toucan typically range from $1,500–$5,000/month for most SaaS use cases, with tiered plans (Start, Grow, Scale, Enterprise) that adjust based on usage and features. The math per tenant depends entirely on how many customers you're serving — at 100 tenants on a $2,000/month plan, you're at $20/tenant/month. At 500 tenants, it's $4.

This is why tenant-based or flat-rate pricing tends to win for growing SaaS companies. Your analytics cost doesn't linearly track your revenue growth.

Feature-based tiers

Some vendors gate features — white-labeling, row-level security, multi-tenancy controls — behind higher plan tiers. You might start at $1,995/month (Explo's entry point, now migrating to Omni) and find that the features you actually need for multi-tenant isolation are only in a custom enterprise plan.

Always ask which tier includes tenant-level data isolation and white-label embedding before benchmarking prices. Two platforms at the same headline price can have very different feature floors.

The hidden costs that inflate your real per-tenant number

The platform fee is only part of it. Here's what vendors don't put on the pricing page:

• Onboarding and implementation fees — some vendors charge $5,000–$20,000+ for initial setup, especially if your data model is complex
• White-label add-ons — removing vendor branding is often a paid upgrade, sometimes adding 20–40% to your base cost
• Row-level security and multi-tenancy — proper tenant data isolation isn't always included. Building it yourself on top of a traditional BI tool takes engineering sprints
• API call overages — usage-based platforms often have soft caps; exceed them and per-call charges appear on your invoice
• Support tiers — enterprise SLAs, dedicated account management, and priority support can add $1,000–$5,000/month

A $750/month entry plan can realistically land at $3,000–$4,000/month once you add white-labeling, proper multi-tenant security, and a support contract worth having.

What's a reasonable benchmark per tenant in 2026?

Based on current market data, here's a rough range by scenario:

These are ballpark figures. Actual costs depend heavily on which pricing model you're on, how actively users engage with dashboards, and what features you need.

If you're still evaluating whether to build vs. buy, the full build vs. buy breakdown for embedded analytics is worth reading — building in-house typically costs $181,000–$310,000 in year one alone, which reframes "expensive" vendor pricing quickly.

How to evaluate pricing before talking to sales

A few things worth doing before you get on a call:

1. Define your tenant count now and at 2x and 5x growth. Run the math for each scenario under each pricing model. Seat-based pricing that looks fine at 50 tenants can be prohibitive at 300.
2. Ask for an all-in quote. Request pricing that includes white-labeling, row-level security, multi-tenancy, and your expected support tier. Compare those numbers, not the headline figures.
3. Check if multi-tenancy is native. Some platforms require engineering work to isolate tenant data properly. That hidden implementation cost belongs in your total cost of ownership.
4. Negotiate on annual commits. Most vendors discount 20–30% for annual contracts. If you're confident in the platform after a POC, this is the lever to pull.

For more on what to look for when monetizing your analytics layer as a product feature, that's worth a read alongside your vendor evaluation.

The pricing model that fits your growth matters more than the number

A $2,000/month platform with flat pricing that doesn't scale per user or per tenant is a very different investment than a $1,000/month platform that doubles in cost every time you add 100 customers.

For CPOs evaluating embedded analytics, the goal isn't finding the cheapest option — it's finding the model where analytics margin stays predictable as you grow. That usually means avoiding pure seat-based pricing, understanding exactly which features require which tier, and pressure-testing the vendor's multi-tenancy architecture before you sign.

The per-tenant cost you should care about is the one at your target scale, not the one on the pricing page today.

Frequently asked questions: embedded analytics multi-tenancy and pricing

Does Toucan support row-level security natively?

Yes. Toucan enforces row-level security at the query level using token-based authentication. Each tenant's data access is scoped through a JWT token passed at embed time, ensuring no cross-tenant data leakage without custom engineering.

What's the difference between multi-tenancy and row-level security in embedded analytics?

Multi-tenancy refers to the ability to serve multiple customer organizations from the same platform instance, each with isolated data and configuration. Row-level security (RLS) is a specific data access control mechanism that restricts which rows a user can query, based on their identity or role. Both are required for production SaaS deployments.

Which embedded analytics platforms don't require custom engineering for multi-tenancy?

Purpose-built embedded analytics platforms like Toucan, Luzmo, and GoodData handle multi-tenancy natively. Traditional BI tools like Tableau, Power BI, and Metabase require significant custom development to achieve equivalent tenant isolation.

How does per-tenant pricing interact with multi-tenancy features?

Platforms that price by tenant (flat-rate) typically include native multi-tenancy and RLS in their base plan. Platforms with per-user/seat pricing often gate multi-tenant isolation features behind higher enterprise tiers — adding cost on top of per-user fees.

What authentication methods do embedded analytics platforms use for tenant isolation?

Most purpose-built embedded analytics tools support JWT (JSON Web Tokens), SAML, and OIDC. The parent SaaS application generates a signed token encoding the user's tenant ID and permissions, which the analytics platform validates server-side before executing any query.