Why Your Looker Bill Doubled (and How to Cut It in Half with Superset)

The Looker Cost Trap: Why Your Bill Keeps Growing

You signed up for Looker three years ago. The pitch was clean: enterprise business intelligence, seamless Google Cloud integration, world-class support. The first year cost $60K. Fair enough for a platform that promised to democratize data across your organization.

Then something happened.

Year two: $95K. Year three: $140K. Now you’re looking at a renewal notice that makes your CFO wince. You haven’t added that many users. Your data volume hasn’t exploded. So why is your bill doubling every 18 months?

This isn’t a coincidence. This is how Looker’s licensing model works—and it’s by design.

Looker pricing operates on a per-user, per-year basis, starting at $60K annually for a standard instance. But the real cost drivers hide in the fine print: viewer licenses, API calls, scheduled queries, embed costs, and the hidden tax of platform lock-in. As your organization scales, as you add dashboards, as you push more queries through the system, costs compound in ways that aren’t immediately visible on the invoice.

According to detailed analysis of Looker pricing structures, typical mid-market deployments see costs ranging from $100K to $300K annually, with some organizations reporting bills exceeding $500K once embedded analytics and high-volume API usage are factored in. The platform’s licensing model penalizes scale—exactly when you’d expect economies of scale to kick in.

For data and analytics leaders at scale-ups and mid-market companies, this creates a painful choice: keep paying the escalating tax, or explore alternatives. Secoda’s research on Looker cost containment shows that many teams attempt to manage costs through data model streamlining and caching strategies, but these are band-aids on a systemic problem.

There’s a better path forward—one that doesn’t require abandoning your data infrastructure or retraining your entire analytics team.

Understanding Looker’s Hidden Cost Drivers

Before you can cut your bill in half, you need to understand exactly where the money is going. Looker’s pricing appears simple on the surface but becomes complex the moment you operationalize it.

The Base License Trap

Looker’s starting price of $60K per year sounds reasonable until you realize what that covers. You get a single instance with a baseline number of users. But “users” is where Looker gets creative.

There are three user tiers:

Standard Users — Full access to create and edit content. These are your analysts and data engineers. At roughly $2,000 per user annually, a team of 10 costs $20K before you’ve even turned on the lights.

Viewer Users — Read-only access to dashboards and reports. Looker charges $500–$1,000 per viewer annually, depending on your contract. If you’re embedding dashboards for customers or end-users, this tier becomes the dominant cost. A SaaS company embedding analytics for 500 customers pays $250K–$500K in viewer licenses alone.

Developer Users — API-heavy access for integrations and embedded use cases. These are expensive and often require custom pricing negotiations.

The trap: as your organization grows, you need more users. But Looker doesn’t scale linearly—it scales exponentially. Each new user adds $500–$2,000 to your annual bill. For a growing company, this is a built-in cost spiral.

Query and API Costs

Looker charges separately for heavy API usage. If you’re embedding dashboards, running scheduled queries, or building custom integrations, you’re burning through API calls. Some organizations negotiate API call limits into their contracts; others face overage charges that can reach tens of thousands of dollars annually.

Scheduled reports—a feature you’d think would be included—often trigger additional costs. Running a dashboard refresh every hour across 50 dashboards for 100 users means thousands of queries per day. Looker’s billing can penalize you for this operational necessity.

The Embed Tax

If you’re embedding Looker dashboards into your product, you’re in the most expensive segment of Looker’s pricing matrix. Embedded analytics require viewer licenses, API calls, and often custom development support. Analysis of embedded Looker costs shows that teams embedding dashboards for external users typically spend $200K–$500K+ annually, with some enterprise deployments exceeding $1M.

This is the core issue for product-led data teams: Looker’s pricing model was built for internal BI, not for embedding analytics at scale. If your business model depends on embedding dashboards—whether for customers, partners, or internal stakeholders—Looker becomes prohibitively expensive.

Support and Professional Services

Looker’s enterprise support plans add $20K–$50K annually. If you need custom development, data modeling help, or performance tuning, you’re paying Google’s consulting rates. Many mid-market organizations end up spending as much on support as they do on the platform itself.

Why Your Costs Keep Climbing (And Why Looker Wants Them To)

Looker’s pricing model is intentionally designed to increase over time. This isn’t a bug—it’s the business model.

Google acquired Looker in 2019 for $2.6 billion. The company needed to justify that acquisition through revenue growth. One way to drive revenue: make the platform expensive for existing customers and charge more as they scale. This creates a revenue ratchet: each year, more users, more queries, more API calls, and a higher bill.

The second mechanism is lock-in. Once you’ve built 200 dashboards in Looker, trained 50 analysts on the platform, and embedded dashboards in your product, switching costs are enormous. Looker knows this. They price accordingly.

Third, Looker’s architecture is cloud-native and Google-centric. If you’re on Google Cloud Platform (GCP), Looker integrates seamlessly with BigQuery, Cloud SQL, and other GCP services. But this integration comes with a cost: your queries run through Google’s infrastructure, and you’re paying both Looker licensing and GCP compute. Google Cloud’s own documentation on visualizing costs shows that many organizations use Looker to track their GCP bills—a reminder that you’re often paying Looker to visualize how much you’re spending on Google’s other services.

For organizations on AWS or multi-cloud environments, Looker becomes even more expensive because you lose the native integration advantage. AWS documentation on analyzing cost reports with Looker shows that AWS users often need custom data pipelines and external tools to achieve the same functionality that GCP users get natively.

The result: a pricing model that punishes scale, rewards lock-in, and makes it nearly impossible to predict your annual bill.

The Apache Superset Alternative: What You’re Actually Paying For

Apache Superset is an open-source business intelligence platform that does 80% of what Looker does at 20% of the cost. More importantly, it’s designed for the modern data stack—modular, API-first, and built for embedding.

Superset’s core advantage isn’t that it’s free (though it is). It’s that it’s open-source and self-hosted. You control the infrastructure, the data connections, and the deployment model. There are no per-user licenses, no viewer fees, no API call overages, and no lock-in.

The Cost Comparison

Let’s build a realistic scenario: a mid-market company with 30 internal analysts, 200 viewer users (for dashboards), and 10 embedded use cases (dashboards embedded in products or customer portals).

Looker Cost Breakdown:

• 30 Standard Users @ $2,000/year = $60,000
• 200 Viewer Users @ $750/year = $150,000
• Embedded Analytics (custom pricing) = $100,000–$200,000
• API Overages and Scheduled Queries = $20,000–$50,000
• Enterprise Support = $30,000
• Total: $360,000–$490,000 annually

Apache Superset Cost (via D23 Managed Service):

• Hosted Superset instance = $15,000–$30,000 annually (depending on compute and data volume)
• Data consulting and custom development = $20,000–$40,000 (as needed)
• Total: $35,000–$70,000 annually

That’s an 80–85% cost reduction.

But the comparison gets more interesting when you factor in switching costs and implementation. At D23, we manage Apache Superset for teams that need production-grade analytics without platform overhead. We handle the hosting, security, backups, and upgrades. You get the benefits of open-source—unlimited users, unlimited dashboards, unlimited API calls—without the operational burden.

The real advantage of Superset isn’t just cost. It’s flexibility. You’re not locked into a single cloud provider. You’re not paying per-user. You’re not penalized for embedding. You can integrate with any data warehouse, any data lake, any API. You own your dashboards and your data model. If you want to switch platforms in five years, you can export your work and move it elsewhere. Try doing that with Looker.

How Superset Handles the Use Cases Looker Charges For

Looker’s pricing model works by charging for features that Superset includes by default. Let’s walk through the major ones:

Embedded Analytics Without the Embed Tax

Looker charges $500–$2,000+ per embedded dashboard. Superset’s architecture is built for embedding. You can embed dashboards in your product, your customer portal, or your internal applications using Superset’s native embed API. There are no per-embed fees, no viewer license taxes, no API call overages. You pay once for the platform and embed as many dashboards as you want.

This is why product teams and SaaS companies are migrating to Superset. If your business model depends on embedding analytics, Looker’s pricing becomes a profit killer. Superset’s API-first design (outlined in D23’s platform documentation) makes embedding a first-class feature, not an expensive add-on.

Unlimited Users and Viewers

Superset doesn’t charge per user. You can give dashboard access to your entire organization, your customers, your partners—and pay the same amount. This is revolutionary for companies that want to democratize data but are held back by Looker’s viewer license costs.

Imagine you’re a portfolio company owned by a private equity firm. You want to share KPI dashboards with all 500 employees. Under Looker, that’s $250K–$500K in viewer licenses. Under Superset, it’s included in your base platform cost.

Text-to-SQL and AI-Powered Analytics

Looker offers limited AI capabilities, mostly through integrations with external tools. Superset’s AI integration—available through D23’s MCP server for analytics—enables text-to-SQL queries and natural language analytics. Users can ask questions in plain English and get SQL queries and visualizations automatically. This is a $50K–$100K+ feature in enterprise Looker deployments. In Superset, it’s integrated into the platform.

Scheduled Queries and Reports

Superset handles scheduled queries and automated reports without additional licensing. You can set up hundreds of scheduled queries, automated exports, and email reports—all included in your base cost. Looker charges separately for high-volume scheduling.

API-First Architecture

Superset’s API is comprehensive and unlimited. You can build custom integrations, embed dashboards, pull data programmatically, and automate workflows without worrying about API call limits or overage charges. This is critical for engineering teams building analytics into their products or platforms.

The Migration Path: From Looker to Superset

Migrating from Looker to Superset is not a rip-and-replace operation. It’s a deliberate, phased process that can be completed in 6–12 weeks with minimal disruption.

Phase 1: Assessment and Planning (Week 1–2)

Start by documenting your current Looker environment:

• Dashboard Inventory: How many dashboards do you have? Which are actively used? Which are orphaned?
• User Base: How many Standard, Viewer, and Developer users do you have? Which teams depend on which dashboards?
• Data Connections: Which databases and data warehouses do you connect to? BigQuery, Snowflake, Redshift, PostgreSQL?
• Custom Code: How much LookML have you written? How many custom blocks or extensions do you have?
• Embedded Dashboards: How many dashboards are embedded in products or customer portals?
• API Usage: How many API calls do you make monthly? What integrations depend on Looker’s API?

This assessment tells you the true scope of migration. Most organizations discover that 30–40% of their dashboards are unused or outdated. This is your opportunity to clean house.

Phase 2: Proof of Concept (Week 3–4)

Build a small Superset instance and migrate 5–10 of your most critical dashboards. The goal is to validate that Superset can handle your use cases and that your team can learn the platform.

Key Testing Areas:

• Data connections: Can Superset connect to all your data sources?
• Dashboard complexity: Can Superset replicate your most complex dashboards?
• Performance: Do queries run as fast as in Looker? (Usually faster, because Superset is closer to your data.)
• User experience: Can your analysts and viewers navigate Superset intuitively?
• Embedded dashboards: Can you embed dashboards in your product or portal?

During this phase, work with a Superset expert (like D23’s consulting team) to validate architecture and identify any custom work needed.

Phase 3: Pilot Deployment (Week 5–8)

Once the PoC validates your use cases, deploy a production Superset instance. Start with one team or business unit—typically the team that’s most data-driven and least dependent on custom Looker features.

Migrate their dashboards, train them on Superset’s interface, and run both systems in parallel for 2–4 weeks. This parallel run is crucial. It lets you validate that Superset is delivering the same insights as Looker and that users are comfortable with the transition.

Migration Checklist for Each Dashboard:

• Recreate data model (or import from Looker if possible)
• Build visualizations in Superset
• Set up filters and interactivity
• Configure drill-down and cross-filtering
• Test performance under load
• Train users on new interface
• Document dashboard logic and ownership

Phase 4: Full Migration (Week 9–12)

Once the pilot team is confident, scale to the rest of your organization. Migrate dashboards in waves, prioritizing by usage and criticality.

Parallel Run Strategy:

Keep Looker running during migration. This gives teams time to adjust and lets you validate that Superset is delivering equivalent or better results. Plan to sunset Looker 4–8 weeks after your last dashboard migrates to Superset.

Training and Change Management:

Superset’s interface is different from Looker’s. Invest in training:

• Dashboard Viewers: 30-minute orientation on navigating dashboards, using filters, and exporting data.
• Dashboard Creators: 4-hour training on Superset’s data model, visualization builder, and dashboard design.
• Admins: 8-hour training on user management, data connections, security, and performance tuning.

Phase 5: Optimization (Ongoing)

Once you’re fully migrated, optimize for performance and cost:

• Query Caching: Configure Superset’s caching layer to reduce database load.
• Data Model Optimization: Review your data model and optimize for the most common queries.
• Compute Sizing: Right-size your Superset instance based on actual usage patterns.
• API Optimization: If you’re using Superset’s API, optimize queries and cache responses.

This is where you realize the true cost savings. In Looker, optimization often means paying for better support or custom development. In Superset, optimization is built into the platform and doesn’t trigger additional costs.

Real-World Migration Example: A Mid-Market SaaS Company

Let’s walk through a real scenario. Imagine you’re a B2B SaaS company with 150 employees, $20M ARR, and a customer base of 500 accounts. You’ve been using Looker for three years.

Current Situation:

• 25 internal analysts and data engineers (Standard Users)
• 50 business users viewing dashboards (Viewer Users)
• 20 embedded dashboards in your customer portal (Embed Users)
• 15 API integrations pulling data from Looker
• Annual Looker bill: $285,000

Migration Plan:

Week 1–2: Assessment

• Audit 200 Looker dashboards; identify 80 actively used
• Document 15 data sources and 50 LookML models
• Map API integrations and dependencies
• Cost: Internal time only

Week 3–4: PoC

• Deploy Superset on AWS (managed by D23)
• Migrate 10 dashboards (mix of simple and complex)
• Test data connections and performance
• Cost: $5,000 (D23 consulting)

Week 5–8: Pilot

• Full Superset deployment with 100 GB data warehouse
• Migrate dashboards for data team and finance team
• Run parallel with Looker
• Cost: $8,000/month (D23 managed Superset) + $4,000 consulting

Week 9–12: Full Migration

• Migrate remaining 70 dashboards
• Train all users
• Decommission Looker
• Cost: $8,000/month (D23) + $6,000 consulting

Total Migration Cost: $35,000–$40,000

First-Year Savings:

• Looker cost: $285,000
• Superset cost (D23 managed): $96,000 ($8,000/month × 12)
• Net savings: $189,000

Five-Year Savings:

• Looker (assuming 15% annual increase): $1,425,000
• Superset (flat cost): $480,000
• Total savings: $945,000

The migration pays for itself in the first month. By year two, you’re ahead by $200K+.

But the financial savings aren’t the only benefit. You also gain:

• Flexibility: Switch data warehouses, cloud providers, or platforms without re-licensing
• Control: Own your data model and dashboards; no vendor lock-in
• Speed: Deploy dashboards faster without waiting for Looker’s UI or hitting API limits
• Scale: Embed dashboards for unlimited users without per-user fees
• AI Integration: Use text-to-SQL and natural language analytics without paying extra

Addressing Common Migration Concerns

”Won’t Our Team Resist Switching?”

Yes, initially. But Superset’s interface is intuitive, and the benefits are immediate. Analysts love the flexibility. Business users appreciate the speed. The biggest resistance usually comes from teams heavily invested in LookML, but even that resolves quickly once they see Superset’s data model approach.

Key to adoption: involve your team early, train thoroughly, and run parallel systems long enough for confidence to build.

”What About Custom Looker Features We’ve Built?”

Looker’s custom features—custom blocks, extensions, and LookML logic—don’t port directly to Superset. You’ll need to rebuild some of this. However, most custom features are workarounds for Looker’s limitations. Superset’s flexibility often makes the custom work unnecessary.

Example: A client had built a complex LookML model to handle multi-tenant data isolation in Looker. In Superset, this was solved with a simple row-level security (RLS) configuration. The custom code became unnecessary.

”What If We Have Looker Embedded in Our Product?”

This is actually where Superset shines. Looker’s embed API is restrictive and expensive. Superset’s embed API is powerful and unlimited. You’ll spend time rebuilding embeds, but the result is better performance, more functionality, and zero per-embed costs.

Many product teams find that switching to Superset for embedded analytics actually improves their product because they can iterate faster and offer more customization to customers.

”What About Data Security and Compliance?”

Superset supports row-level security (RLS), column-level security, LDAP/SAML authentication, and audit logging. Hosted by D23, Superset instances are SOC 2 compliant, encrypted in transit and at rest, and backed by daily snapshots.

For most organizations, Superset’s security is equivalent to or better than Looker’s, especially because you control the infrastructure and don’t rely on Google’s ecosystem.

”What If We’re Locked into Google Cloud?”

Many organizations are on GCP and feel locked into Looker for the BigQuery integration. This is a fair concern, but Superset connects to BigQuery just as effectively as Looker does. You don’t lose the integration; you just gain the ability to also connect to Snowflake, Redshift, or any other database without additional licensing.

In fact, cloud cost management documentation shows that many organizations use Looker specifically to visualize their GCP costs. With Superset, you can do the same thing at a fraction of the cost, and you can also visualize AWS, Azure, or multi-cloud costs in the same system.

The Hidden Benefits: Beyond Cost Savings

While the $150K–$300K annual savings is compelling, the real value of migrating to Superset goes deeper.

Speed to Insight

Superset dashboards load faster than Looker dashboards. Why? Because Superset is closer to your data. There’s no Google intermediary, no licensing check, no API call overhead. Queries that took 30 seconds in Looker often run in 5–10 seconds in Superset. This matters for real-time dashboards, embedded analytics, and interactive exploration.

Data Model Flexibility

Looker’s LookML is powerful but rigid. Your data model is locked into Looker’s paradigm. Superset’s data model is more flexible—it’s based on SQL, which your analysts already know. You can pivot to a new data warehouse, add new data sources, or change your data architecture without rebuilding your entire BI layer.

API-First Architecture

Superset’s API is comprehensive and designed for modern data stacks. You can programmatically create dashboards, manage users, pull data, and embed visualizations. This is critical for engineering teams building analytics into products or platforms. Looker’s API is functional but expensive. Superset’s API is unlimited.

AI and Automation

Superset’s integration with AI tools (like D23’s MCP server for analytics) enables text-to-SQL, natural language dashboards, and AI-assisted insights. Looker offers some AI features, but they’re add-ons and often require additional licensing. In Superset, AI is part of the platform.

Multi-Cloud and Hybrid Support

Looker is optimized for GCP. If you’re on AWS, Azure, or multi-cloud, you’re at a disadvantage. Superset treats all databases equally. This is critical for enterprises managing data across multiple cloud providers or for organizations planning future infrastructure changes.

Getting Started: Your Next Steps

If your Looker bill has doubled in the last two years, migration is worth serious consideration. Here’s how to start:

Step 1: Run the Numbers Pull your last three Looker invoices. Calculate your annual cost and your cost per user. If you’re paying more than $100K annually or more than $2,000 per analyst, Superset will save you significant money.

Step 2: Audit Your Dashboards How many dashboards do you actually use? Most organizations find that 30–50% of their dashboards are orphaned or outdated. Migration is a good time to clean house.

Step 3: Talk to a Superset Expert Don’t try to migrate alone. Work with a managed Superset provider like D23 that understands your use cases, can validate your architecture, and can handle the heavy lifting of migration and optimization.

Step 4: Run a Proof of Concept Migrate 5–10 dashboards and validate that Superset meets your needs. This typically takes 2–4 weeks and costs $5K–$10K. It’s the best investment you’ll make in your data infrastructure.

Step 5: Plan Your Migration Once the PoC validates your use cases, build a migration roadmap. Most organizations complete full migration in 8–12 weeks with minimal disruption.

Conclusion: The Economics of Open Source

Your Looker bill doubled because Looker’s pricing model is designed to increase as you scale. This isn’t unique to Looker—it’s how proprietary BI vendors work. They lock you in, they make the platform expensive, and they raise prices as your dependency grows.

Apache Superset represents a different model: open-source, community-driven, and designed for the modern data stack. When you choose Superset via a managed provider like D23, you get the benefits of open-source—unlimited users, unlimited dashboards, unlimited API calls—without the operational burden of managing it yourself.

The financial case is clear: 80–85% cost reduction, payback within weeks, and no lock-in. But the strategic case is stronger: you regain control of your data infrastructure, you gain the flexibility to adapt to future changes, and you access modern capabilities like AI-powered analytics and API-first architecture that Looker charges premium prices for.

If you’re tired of watching your Looker bill climb, it’s time to explore alternatives. Start with a proof of concept. Talk to a Superset expert. Run the numbers. You’ll likely find that cutting your BI costs in half isn’t just possible—it’s inevitable.

The question isn’t whether you should migrate. It’s when.