Bumper today expanded access to its podcast analytics platform, making the Bumper Dashboard available to all podcasters through a new free tier, and introducing an MCP server for enterprise subscribers - a move that brings natural-language AI queries to podcast data for the first time on the platform.

A dashboard that no longer requires a budget

The announcement, dated June 8, 2026, restructures the Bumper Dashboard around three distinct access levels. The free tier gives any podcaster access to the most recent three months of data at no cost. The Pro plan, priced at $50 per month per show, unlocks the full analytics suite, including complete episode history and consumption metrics. Enterprise subscriptions, priced at $500 per month, cover up to 10 podcasts, offer a discount for additional shows, and now include access to the newly launched Bumper MCP server.

According to Bumper, the platform aggregates listeners, viewers, listen time, playback, followers, and downloads into a single automatically updated view, covering both show and episode level data. The service connects to multiple hosting providers, podcast apps, and attribution platforms, consolidating what has historically required manual management across separate dashboards and CSV exports.

"Better data shouldn't be a premium feature. The free tier in the Bumper Dashboard exists because we want the floor for podcast measurement to be higher across the whole industry, not just for the publishers who can afford enterprise tools," said Jonas Woost, co-founder at Bumper.

That framing reflects a structural tension in podcast publishing that has persisted for years. Audience data sits scattered across too many systems - hosting platforms, distribution apps, attribution services - and smaller publishers have rarely had the resources to consolidate it. Free access to a unified view represents a meaningful lowering of the barrier, even if the free tier is bounded to a 90-day window.

The Bumper Score, now available to all plans

Alongside the pricing changes, Bumper confirmed that the Bumper Score - its benchmark tool for measuring a podcast's performance against comparable shows - is now included across all plans, including the free tier. The Score had previously been positioned as a premium capability.

The Bumper Score uses verified consumption data rather than raw download counts to assess how a show is performing in context. The distinction is significant. According to Bumper, many publishers have seen download numbers decline in recent years due to changes in industry measurement definitions and platform updates, while their verified listener counts have continued to grow. The Bumper Score is designed to reflect the latter - actual audience engagement rather than a figure shaped by methodology shifts.

The Bumper Score launched on April 21, 2026, as a 0-to-200 verified audience metric built on first-party data. Research from Oxford Road, cited alongside that launch, estimated that implementing better data and verification across the industry could unlock an additional $1 billion in podcast advertising spend. That figure framed the product as a commercial argument as much as a measurement tool.

Making the Score available at no cost is a different kind of commercial argument - one aimed at market-wide adoption rather than premium upsell. If more shows carry a Bumper Score, the benchmark gains comparability. A score is only as useful as the number of shows it covers.

What the MCP server actually does

For Enterprise subscribers, the more technically substantive change is the introduction of the Bumper MCP server. Model Context Protocol, or MCP, is an open standard developed by Anthropic that defines how AI models communicate with external tools and data sources. In practical terms, it lets an AI assistant connect to a live data system and answer questions about it in plain language, without the user having to navigate a dashboard, export files, or write queries.

According to Bumper, Enterprise users will be able to ask natural-language questions about their podcast performance and receive instant, data-driven answers, without leaving the tools they already work in. The capability applies to the full range of data in the Bumper Dashboard: playback, verified audience, listen hours, downloads, followers, and episode-level breakdowns.

The architecture is straightforward. The MCP server sits between the user's AI assistant and the Bumper Dashboard data. When a user poses a question - say, which episodes drove the most verified listeners last month, or how a specific show's performance compares to its category average - the MCP server translates that into a data query, retrieves the relevant metrics from the dashboard, and returns the result inside the AI tool. No separate login. No export. The data comes to the workflow rather than requiring the user to go to the data.

Bumper is entering an MCP landscape that has become significantly more crowded over the past 12 months. Google Analytics released an experimental MCP server on July 22, 2025, enabling natural language queries against analytics data. AppsFlyer launched its own MCP tool on July 17, 2025, connecting its attribution platform to large language models for marketing teams. Google released an open-source MCP server for its Ads API on October 7, 2025, with read-only reporting and diagnostics capabilities. Amazon Ads launched a closed beta MCP server on November 13, 2025, covering sponsored ads, DSP campaigns, and Amazon Marketing Cloud analytics. Meta opened its ad system to AI agents through MCP connectors in April 2026, the first time the platform opened advertising infrastructure to third-party AI systems at that level. Pacvue launched its own MCP server on May 14, 2026, connecting its retail media platform to ChatGPT, Copilot, Gemini, and Claude.

Bumper's implementation follows a pattern common to these deployments: read access to proprietary data through a standardized AI interface. What distinguishes it is the domain. Podcast data has not previously been accessible through MCP, and the specific combination of verified audience metrics, consumption data, and benchmark scoring is not replicated elsewhere in the available MCP ecosystem for audio publishers.

Why podcast measurement matters to ad buyers

The timing of this announcement sits inside a period of material commercial growth for podcast advertising. Podcast ad spending hit $408 million in December 2025, the highest single month recorded in Magellan AI's dataset. Spending grew 26% year-over-year in Q3 2025 and accelerated to 32% year-over-year growth in Q4 2025. New brand entry is accelerating too - Germany added 259% more new brands to podcast advertising in Q1 2026 compared to Q1 2025, and the United States grew new brand entry by 27% over the same period.

That commercial growth carries a measurement problem with it. When a new advertiser enters podcast for the first time, typically spending around $33,900 on average according to Magellan AI Q4 2025 data, the due diligence they need to perform - understanding audience composition, verifying listener counts, comparing shows against category benchmarks - requires navigating tools that do not talk to each other. Download data comes from the hosting platform. Audience composition requires a separate subscription. Attribution lives elsewhere again.

Audio advertising's full-funnel measurement challenges remain one of the documented structural constraints on the channel's growth. According to industry research cited by PPC Land, consumers dedicate 31% of their media time to audio, while advertisers allocate only 9% of their budgets to audio platforms. That 22-percentage-point gap reflects, in part, the difficulty of defending audio spend when measurement is fragmented.

Triton Digital launched a Podcast Metrics Industry Feature Set in February 2026 that gave publishers competitive download benchmarks across the top 20 podcast markets. Nielsen integrated Triton Digital's Demos+ data into Nielsen Media Impact in April 2026, enabling podcast data to sit alongside television, radio, and digital channels inside the same cross-media planning tool. Spotify introduced a verified badge for podcasters in May 2026, tying its identity program to verified audience authenticity as a prerequisite for badge eligibility.

Bumper's move today adds a different layer to this stack. The Bumper Dashboard is not a planning tool for advertisers - it is a performance tool for publishers. The free tier targets independent creators. The Pro tier targets professional shows and growing networks. The Enterprise tier, and its MCP server, targets media companies managing multiple shows who also want to query their performance data through AI.

Three billion hours as a baseline

According to Bumper, the platform has tracked more than 3 billion hours of podcast consumption to date. That figure provides the dataset from which the Bumper Score draws its benchmarks. Without a comparable population of shows and hours against which to calibrate, a score from 0 to 200 would have no reference class. The scale of the consumption database is the underlying justification for making the benchmark broadly available rather than restricting it to premium tiers.

Bumper was founded by Dan Misener and Jonas Woost. The company describes its focus as podcast data and growth, working with creators and brands. The decision to restructure around a free tier is consistent with an approach to building market coverage - a benchmark tool achieves more commercial utility when it covers a large portion of the podcast market than when it covers only the publishers who can afford a paid subscription.

The 3 billion hours figure is not broken down by time period or geography in the announcement, so it is not possible to assess how current or geographically distributed the underlying data is. Bumper positions the figure as context for the scale of the infrastructure underpinning the Score.

Practical implications for publishers and advertisers

For an independent podcaster, the practical value of the free tier is a consolidated view of audience metrics without the operational overhead of logging into multiple platforms. The three-month data window limits historical analysis but covers current performance adequately for most editorial and scheduling decisions.

For a professional creator or small network on the Pro plan, the full episode history and consumption data at $50 per month per show gives a complete performance record alongside the Bumper Score. The pricing is competitive with other analytics tools in the podcast space, and the combination of aggregated data and verified benchmarking is a differentiator relative to hosting-platform analytics alone.

For Enterprise subscribers at $500 per month, the MCP server changes the operational model more significantly. Rather than pulling reports and interpreting dashboards, a team working inside Claude, ChatGPT, or another AI assistant can ask direct questions about their show portfolio and receive answers grounded in live Bumper data. The IAB Tech Lab Agent Registry, as of March 2026, held 10 active MCP entries, all operating under the standard Anthropic developed. Bumper's server adds a podcast-native data source to that growing registry of AI-accessible marketing tools.

Whether the MCP functionality drives enterprise adoption at scale will depend partly on whether podcasters already work inside AI tools - and partly on the quality of the data that the server surfaces. A natural-language interface is only as useful as the data behind it. Bumper's claim that the dashboard draws from across hosting providers, apps, and attribution platforms simultaneously is the critical premise. If the aggregation is complete, the AI layer becomes genuinely useful. If it is partial, the conversational interface surfaces the same fragmentation problem it was designed to solve.

Timeline

Summary

Who: Bumper, a podcast data and growth company co-founded by Dan Misener and Jonas Woost, headquartered in New York and Toronto.

What: Bumper expanded access to its Bumper Dashboard by introducing a free tier covering the most recent three months of data, restructured its pricing into three tiers (Free, Pro at $50 per month per show, Enterprise at $500 per month), made the Bumper Score available across all plans, and launched an MCP server for Enterprise subscribers that enables natural-language AI queries against podcast performance data.

When: The announcement was published on June 8, 2026.

Where: The Bumper Dashboard is an online platform aggregating data from hosting providers, podcast apps, and attribution services. The MCP server is accessible to Enterprise subscribers and connects to AI tools that support the Model Context Protocol standard.

Why: Podcast measurement remains fragmented across too many platforms for publishers to manage manually. The free tier lowers access to consolidated analytics for independent creators. The Bumper Score's expanded availability increases its benchmarking utility by widening coverage across the market. The MCP server addresses a growing enterprise demand to query proprietary performance data through AI assistants without adding another platform to an already complex workflow - a need that has driven MCP adoption across advertising platforms including Google Analytics, AppsFlyer, Amazon Ads, Meta, and Pacvue over the past 12 months.