Analysis of ZoomInfo's 2026 earnings reveals AI feature growth is outpacing revenue, exposing a persistent data quality gap for B2B and local businesses.

ZoomInfo's Q2 2026 earnings call reported a modest 1.2% year-over-year revenue increase to $310 million, while highlighting AI integrations with platforms like an LLM provider and Claude. However, the number of customers paying over $100k annually saw a sequential decrease of 9, and net revenue retention was 89%. This suggests that while AI features are being adopted, the core challenge remains the quality of underlying data, which decays at an estimated 22-40% annually.

TL;DR

ZoomInfo Q2 2026 revenue grew only 1.2% YoY to $310 million, while down-market ACV fell 12%.
B2B contact data decays at 22-40% annually, costing companies an average of $12.9 million per year.
Apollo.io competes with ZoomInfo on price, offering a free tier and plans starting at $49/month, but data accuracy is lower for enterprise US phone numbers.
Incumbent platforms like ZoomInfo and Apollo struggle with local business data, where public directory sourcing provides ~70% email deliverability.
AI adoption in sales is now 81-87%, but effectiveness is limited by static databases versus real-time, fact-based lead data.

What ZoomInfo's Q2 2026 Earnings Reveal About the AI Prospecting Market

ZoomInfo's Q2 2026 financial results present a narrative of slowing growth, with revenues reaching $310 million for a modest 1.2% year-over-year increase. While this figure narrowly surpassed analyst expectations, it signals a significant deceleration from the high-growth periods that previously defined the go-to-market intelligence sector. This slowdown is not occurring in a vacuum; it reflects a broader market maturation and difficult economic conditions impacting technology budgets. According to the Salesforce State of Sales 6th Edition (2024), a staggering 67% of sales representatives did not expect to meet their quotas in 2024, indicating widespread pressure on sales performance and, by extension, the tools they rely on. [18] This environment forces a flight to efficiency, where companies scrutinize every dollar of their tech stack spend. The modest top-line growth, paired with a challenging sales landscape, suggests that even established platforms like ZoomInfo are contending with market saturation and heightened buyer caution, making substantial new revenue growth increasingly difficult to achieve without significant product differentiation or market expansion. The era of automatic budget increases for sales technology appears to be giving way to a more measured, results-driven approach from customers.

In response to a challenging growth environment, ZoomInfo has strategically pivoted towards profitability and up-market consolidation, evidenced by an adjusted operating margin expansion to 35%. This move highlights a deliberate focus on operational efficiency rather than pursuing growth at all costs. The company's customer segment performance further illuminates this strategy: Annual Contract Value (ACV) from up-market customers grew by a respectable 3% year-over-year, while the down-market segment experienced a sharp 12% decline. This bifurcation shows a clear effort to shed smaller, less profitable accounts in favor of larger, more stable enterprise clients. Such a strategy aligns with broader SaaS trends where enterprise accounts deliver superior retention and expansion profiles. For instance, a 2024 analysis of SaaS benchmarks found that companies with Annual Contract Values between $100k-$250k achieved a median Net Revenue Retention of 113%, significantly higher than those with smaller deal sizes. [11] This disciplined focus on high-value customers and margin improvement, as detailed in their AI prospecting scoops, is a defensive maneuver to build a more resilient business model amidst uneven demand and a tightening market for go-to-market tools.

A key indicator of underlying friction is ZoomInfo's Net Revenue Retention (NRR) rate, which dipped to 89% from 90% in the prior quarter. While seemingly minor, this decline points to persistent challenges in customer retention and expansion, which are critical for sustainable SaaS growth. An NRR below 100% signifies that revenue from customer churn and downgrades is outpacing revenue from upsells and cross-sells, a concerning trend for any subscription business. [12] This issue is magnified by the fundamental problem of B2B data decay, which industry research estimates now occurs at an annual rate between 22.5% and 70.3%. [1] More recent reports from late 2024 even identified monthly decay rates as high as 3.6%, meaning a significant portion of a contact database can become obsolete in just a few months. [4, 5] For a platform whose primary value proposition is accurate data, this relentless decay creates a constant battle to maintain quality and demonstrate value. Even with advanced AI features, if the foundational data is flawed, the insights generated are compromised, leading to customer dissatisfaction and churn. This makes improving NRR not just a sales challenge, but a core product and data integrity challenge.

AI Features vs. Financial Reality: A Platform in Transition

ZoomInfo is aggressively repositioning itself as a 'headless context layer' for AI agents, a strategic pivot away from its legacy as a seat-based data provider. The company's Q2 2026 earnings materials emphasize this transformation, highlighting the launch of GTM.AI and a Model Context Protocol to embed its data within external AI workflows. [3, 4] This strategy includes native integrations with major platforms like Microsoft Copilot and those from leading AI labs, designed to make ZoomInfo's data the foundational intelligence for go-to-market activities, regardless of the front-end application. [3] This shift acknowledges a broader industry trend where, according to the Salesforce State of Sales 6th Edition (2024), 81% of sales teams are already using or evaluating AI. [14] The core value proposition is no longer just the database itself, but its ability to be operationalized by autonomous agents and AI-driven systems. By becoming the data backbone for these emerging technologies, as detailed in their AI prospecting scoops blog, ZoomInfo aims to create a new consumption-based revenue model that is less reliant on per-user licenses and more aligned with the data-intensive needs of AI applications. [5]

Despite the strategic pivot to AI and a modest 1.2% year-over-year revenue increase to $310.4 million in Q2 2026, ZoomInfo's financial results reveal significant underlying pressures. [2] A critical indicator of this strain is the sequential decrease in its most valuable customer cohort; the number of clients with an annual contract value (ACV) over $100,000 fell by 9, from 1,900 in the prior quarter to 1,891. [2, 7] This decline in high-value accounts, occurring for the second consecutive quarter, directly contributed to a net revenue retention rate of 89%, signaling that churn and down-sells are outpacing expansion revenue from existing customers. [6, 8] This financial reality suggests that while the new AI features are being rolled out, they have not yet been sufficient to counteract the headwinds of a challenging software market and increased competition. The upmarket segment, which has been a pillar of ZoomInfo's strategy, showed decelerating ACV growth of just 3% year-over-year, indicating that even large enterprise customers are exercising caution. [7] The combination of slowing growth, customer contraction, and low retention paints a picture of a company in a difficult transition, where the promise of an AI-driven future has yet to translate into tangible financial momentum.

The most telling sign of ZoomInfo's difficult transition was the disclosure of a non-cash goodwill impairment charge of approximately $651 million in its Q2 2026 results. [1, 6, 8] This massive write-down, which drove a GAAP operating loss of $622 million, is a direct result of the decline in the company's market capitalization and serves as a formal acknowledgment that past acquisitions are not expected to generate their originally anticipated value. [2, 3] Such an impairment indicates that the strategic and financial assumptions underpinning previous M&A activity have been fundamentally re-evaluated and found wanting in the current market. [7] While the charge does not affect immediate cash flow, it reflects a significant reset of long-term expectations for the business. [1] This financial maneuver underscores the immense pressure on the company to prove the viability of its new AI-centric strategy. It is a stark admission that the old model of growth through acquisition and seat-based sales is no longer sufficient, forcing a painful but necessary recalibration of the company's asset values and future outlook as it navigates its pivot to a consumption-based, AI-focused platform.

The core challenge limiting the return on investment for platforms like ZoomInfo remains the fundamental problem of data quality, a persistent issue that new AI features alone cannot solve. While a high percentage of sales teams, 81% according to a 2024 Salesforce study of 5,500 professionals, have adopted AI, their effectiveness is capped by the accuracy of the underlying data. [14, 19] Industry analysis indicates that B2B contact data decays at a staggering rate, with estimates ranging from 22.5% to over 40% annually, as contacts change jobs, phone numbers become disconnected, and companies are acquired. [11, 12] This means that a significant portion of a purchased database can become obsolete within a year, a problem that AI systems can inadvertently amplify by automating outreach based on flawed information. According to Salesforce research, sales representatives still spend approximately 70% of their time on non-selling activities, a figure that includes manual data verification and CRM updates to counteract this decay. [14] This operational drag highlights the disconnect between the promise of AI efficiency and the on-the-ground reality of managing stale data, a problem that requires continuous, automated data hygiene rather than just a more intelligent interface.

Metric	Q4 2025	Q1 2026	Q2 2026	Commentary
Revenue ($M)	$319.1M	$310.2M	$310.4M	Revenue growth has stalled, showing a slight sequential decline then flattening. [2, 15]
Customers >$100k ACV	1,921	1,900	1,891	High-value customer count has decreased for two consecutive quarters. [2, 15, 16]
Net Revenue Retention	90%	90%	89%	NRR has dipped below 90%, indicating customer churn and down-sells are outpacing expansion. [1, 16]
YoY Revenue Growth	3.0%	1.5%	1.2%	Year-over-year growth has decelerated significantly over the last three quarters. [2, 15, 16]
Goodwill Impairment ($M)	$0	$0	$651M	A significant one-time charge reflecting a revaluation of past acquisitions. [1, 8]
Adjusted Operating Margin	38%	35%	35%	Profitability remains strong and stable despite revenue headwinds. [2, 15, 16]

The Elephant in the Room: B2B Data Decays at 30% Annually

The foundational challenge for any AI-powered prospecting tool is the rapid and continuous decay of its underlying B2B contact data. Industry analysis from 2026 indicates that B2B data degrades at an alarming rate, with annual decay estimated between 22.5% and 30%. [10] Email addresses are particularly volatile; some studies noted a monthly decay rate of 3.6% in late 2024, a significant acceleration from historical norms. [9, 10] This decay is not a slow leak but a constant churn driven by job changes, corporate restructuring, and technology migrations. For instance, job title and function changes can affect 28-35% of a contact list annually. [12] This means that a static list of 100,000 prospects purchased in January could contain over 22,000 invalid contacts by December, rendering a substantial portion of a sales team's outreach efforts useless before they even begin. [10] The problem is so pervasive that even sophisticated AI models designed for prospecting are fundamentally handicapped if they are trained on and operate with stale information, a reality that directly impacts their ability to generate meaningful pipeline and revenue.

The financial consequences of this data degradation are staggering, imposing a heavy, often underestimated, tax on go-to-market operations. A frequently cited 2021 Gartner analysis estimates that poor data quality costs organizations an average of $12.9 million annually. [7] Other research from sources like the MIT Sloan Management Review suggests the toll could be even higher, potentially costing companies between 15% and 25% of their total revenue. [15] These costs are not abstract figures; they manifest as wasted marketing spend on campaigns targeting nonexistent contacts, significant operational drag as data teams spend up to 80% of their time cleaning and reconciling information, and lost sales productivity when representatives pursue leads based on flawed intelligence. [7, 13] For example, every outreach attempt made to an invalid email address or a person who has changed roles represents a direct opportunity cost and consumes valuable sales cycles that could have been allocated to verified, in-market buyers. The issue compounds as flawed data propagates through integrated systems like CRMs and marketing automation platforms, corrupting analytics and leading to misguided strategic decisions. [8]

A core part of the data decay problem stems from how incumbent data providers have historically sourced their information. Many legacy databases are built on compiled or crowdsourced data, which introduces significant latency and accuracy issues from the outset. [14] Compiled data, which merges information from various public records and directories, can be up to 18 months old by the time a sales team uses it. [14] Crowdsourcing, which incentivizes users to share contact lists, is also prone to error and lacks real-time verification. [16] This contrasts sharply with modern approaches that prioritize timely, verifiable signals over static attributes. For example, the methodology behind intent data platforms like Bombora's Company Surge, which monitors content consumption to identify active buyers, shifts the focus from who a person is to what they are doing now. This pivot is critical because the value of a lead is no longer in its static profile but in the accuracy and timing of the data, as highlighted in the latest AI prospecting signals from GTM earnings calls.

Ultimately, the persistent decay of B2B data forces a strategic shift from prioritizing list size to demanding verifiable accuracy and timing. The true value of a prospect is not a probabilistic 'fit score' derived from a narrative of past attributes, but a collection of verifiable facts that signal current opportunity. A lead becomes valuable when you can confirm, with high confidence, that a specific individual at a target account just started researching a relevant solution, or that a company just received a new round of funding and is hiring for a specific role. This is where the next generation of AI prospecting tools must excel, not by generating larger lists from decaying databases, but by continuously verifying critical data points in near real-time. As noted in Salesforce's recent discussions on AI, success starts with a unified and trustworthy data strategy. The emphasis moves from a massive, decaying rolodex to a dynamic, event-driven system where accuracy and timeliness are the primary metrics of success, ensuring that sales teams engage with the right person, at the right company, at the exact moment of need.

The Incumbent Blind Spot: Why Local Business Prospecting Fails

Incumbent B2B data platforms like ZoomInfo and Apollo are architecturally designed for prospecting 'people at companies', resulting in near-zero effective coverage of named owners at local, single-location service businesses. These platforms build their databases by scraping professional social networks, corporate websites, and press releases, a methodology that excels at identifying employees within structured corporate hierarchies but fails to capture the sole proprietors and small partnerships that dominate local markets. The core data model is centered on the individual as an employee, not the business entity itself. This creates a structural blind spot for the millions of small businesses that form the backbone of local economies. While these platforms introduce AI features for scoring and engagement, as seen in ZoomInfo's recent announcements, the underlying data asset for the local segment is largely missing. This is a critical failure of scope; the platforms are optimized for a specific type of B2B sale that involves navigating departments and multiple stakeholders, a reality that is completely absent when the target is a single-owner plumbing business or a local accounting firm. The problem isn't the technology, but the foundational assumption about what a 'business' is.

A prospecting strategy that begins with public business directories yields demonstrably higher data quality and deliverability for the local small and medium-sized business (SMB) segment. Keendai's approach, which starts with government and industry business registries, achieves approximately 70% verified email deliverability for named local business owners. This method inverts the model used by corporate B2B platforms. Instead of starting with a massive list of people and attempting to map them to employers, this strategy starts with a verified business entity and then identifies the registered owner or principal. This 'business-first' approach is more resilient to the data decay that plagues corporate contact lists. According to research, B2B contact data can decay at a rate of 22.5% to over 70% annually, with job changes being a primary driver. [1, 5] Local business ownership, in contrast, is a much more stable data point. By focusing on public, often legally mandated, registration data, this methodology provides a more accurate and durable foundation for outreach, sidestepping the churn and noise inherent in employee-centric databases. This approach turns local business directories into a reliable lead generation machine by focusing on ownership, not employment. [10]

For sales teams targeting local businesses, a plain-facts lead containing a business name, a verified owner, and a verified email or phone number is more valuable than an AI-scored B2B contact with a high probability of being outdated. The obsession with complex lead scoring and intent signals, powered by platforms like Bombora, is often irrelevant in the local SMB context. A local electrician's primary 'intent signal' is their business license; they are in business to provide a service. The value is not in predicting their need, but in successfully contacting the owner. B2B data decays at a staggering rate, with some estimates showing monthly decay reaching 3.6% in late 2024, pushing the effective annual decay rate over 35% for many lists. [2, 3] This means a lead from a typical B2B database has a one-in-three chance of being inaccurate within a year. A verified local lead, while simpler, provides a direct path to a conversation, eliminating the 70% of prospecting effort that can be wasted on decayed or inaccurate data. [4] As the 2024 State of Salesforce report from IBM highlights, high-quality, proprietary data is the prerequisite for any successful AI or sales strategy, a principle that is amplified in the high-volume, low-data-quality world of local prospecting. [7]

Prospecting Method	Primary Data Source	Target Segment	Typical Data Accuracy (Est. Annual)	Key Weakness
Incumbent B2B (e.g., ZoomInfo)	Professional Profiles, Web Scraping	Corporate B2B	~70-77% (22.5% decay)	Poor coverage of local business owners.
Value B2B (e.g., Apollo.io)	Aggregated Professional Profiles	Corporate B2B & Startups	~65-75% (25-35% decay)	Same structural flaw as incumbents for local market.
Keendai Method	Public Business Directories	Local SMBs	~90-95% (ownership is stable)	Limited to publicly registered entities.
Legacy List Brokers	Purchased Static Lists	Mass Market	<50% (high initial decay)	High decay, low verification, potential compliance risk.
Manual Prospecting	Google Maps, Social Media	Hyper-Local/Niche	Variable (High if verified)	Not scalable without significant manual effort.
Intent Data (e.g., Bombora)	Bidstream, Content Networks	B2B Tech/High-Value	N/A (Measures Intent, not Contact Info)	Irrelevant for most local service businesses.

A Modern Framework for Evaluating Prospecting Tools

A modern framework for evaluating prospecting tools begins with scrutinizing contractual obligations and billing transparency, demanding flexible terms over rigid, long-term commitments. Vendors often push for annual or multi-year contracts with automatic renewal clauses, a practice that locks customers in regardless of performance or evolving business needs. Instead, procurement teams should champion agreements that offer per-lead bounce credits or pro-rated refunds for inaccurate data. This shifts the financial risk of data decay from the buyer to the vendor, creating a powerful incentive for the provider to maintain data quality. The problem is widespread; vague terms about deliverables and service levels are a common source of B2B contract disputes. A superior model involves paying for outcomes, such as successfully delivered emails or verified contact records, rather than for access to a platform. This approach directly counters the financial damage caused by stale data, which can cost an individual sales representative over 550 hours and $32,000 annually in wasted effort on poor leads. By rejecting inflexible, auto-renewing contracts and insisting on performance-based billing, sales organizations can build a more accountable and effective prospecting stack.

Prospecting tool evaluation must prioritize quantifiable data accuracy metrics over ambiguous marketing claims like 'verified' badges or green checkmarks. The critical metric is the actual, numeric email deliverability percentage, which should be guaranteed in the service-level agreement. B2B contact data decays at a startling rate, with some analyses showing a monthly decay of 3.6% as of late 2024, compounding to over 35% annually. This means that a significant portion of any purchased list becomes obsolete within months. A vague 'verified' status fails to capture this reality, whereas a specific deliverability promise, such as '95% deliverability or a pro-rated credit,' provides a clear standard for performance. For example, while the overall B2B email delivery rate remains high at 98.16%, this top-level number masks a collapse in actual inbox placement, a far more meaningful metric for sales teams. The financial toll of poor data quality is immense, costing U.S. businesses a collective $3.1 trillion annually and individual organizations an average of $12.9 to $15 million per year. Insisting on transparent, numeric, and guaranteed deliverability rates is the only way to ensure prospecting investments translate into tangible sales conversations rather than bounced emails and a damaged sender reputation.

A crucial distinction must be made between tools designed for 'searching' for factual contact data and those built for 'running' campaigns with AI-generated narratives. The foundation of any successful prospecting effort is the accuracy of the underlying data: correct names, titles, emails, and phone numbers. This is the 'searching' layer. Tools like Bombora's Company Surge, which identifies companies actively researching specific topics, provide intent signals that are only valuable if the associated contact data is pristine. The second layer, the 'running' of campaigns, increasingly involves AI to generate personalized outreach. However, as highlighted in Salesforce's 6th Edition "State of Sales" report from 2024, which surveyed 5,500 sales professionals, reps spend a mere 30% of their time on actual selling activities, with the rest consumed by administrative tasks. AI promises to alleviate this, but its effectiveness is entirely dependent on the quality of the initial data. AI prospecting tools that generate compelling stories are useless if those stories are sent to email addresses that hard bounce. Therefore, evaluation frameworks must first validate the data acquisition and verification methodology of a tool before even considering its AI-powered campaign features.

Evaluating the true cost of a bad lead reveals that a cheap but undeliverable contact is far more expensive than a premium lead that connects. A B2B contact purchased for $0.02 that bounces or is directed to the wrong person is not just a sunk cost; it represents a cascade of wasted resources. According to industry analysis, sales departments lose an average of 550 hours and $32,000 per representative due to poor lead data, and each bad record can cost as much as $100 in downstream effects. This hidden cost includes the sales development representative's time, the marketing automation platform's resources, and the erosion of the company's sender reputation. In contrast, a $0.15 local lead that results in a conversation, even if it doesn't close, provides valuable market feedback and a potential future opportunity. The data decay crisis exacerbates this problem, with some estimates placing the annual decay rate of B2B data as high as 70.3%. This rapid degradation means that a significant portion of low-cost data is often outdated before the first outreach attempt is even made. A modern evaluation framework must therefore calculate the cost-per-connected-lead, not just the cost-per-lead, to accurately assess the ROI of any prospecting tool.

Frequently Asked Questions

Is ZoomInfo's AI Copilot worth the cost in 2026?

The value of ZoomInfo's AI Copilot in 2026 is questionable for teams without a clear strategy to manage data quality. While the AI can automate research and surface insights, its effectiveness is limited by the underlying data, which decays at over 22% annually. [1, 3] With contracts starting around $40,000 to access Copilot, the investment may not provide a positive return if the AI is operating on inaccurate information, a risk highlighted by the company's own modest growth and declining customer metrics. [12, 14, 19]

What is the main weakness of AI prospecting tools like Apollo and ZoomInfo?

The main weakness of AI prospecting tools is their dependency on underlying B2B contact data that is often inaccurate and decays rapidly. [11, 16] This data degrades by an estimated 22-40% per year as people change jobs and companies restructure, meaning the AI frequently works with outdated information. [4, 23] Consequently, AI-generated insights and outreach can be based on false premises, leading to wasted sales efforts and damaging a company's reputation. [29]

How can I get accurate contact information for local businesses?

Getting accurate local business contact information requires going beyond major data providers, whose databases often have gaps in this segment. A better approach is using a multi-source or "waterfall" enrichment process that queries multiple databases to fill in missing details. [1] For phone-heavy prospecting, using a dedicated provider specializing in verified direct-dial numbers can significantly increase connection rates compared to the general data offered by all-in-one platforms. [22] This layered strategy compensates for the structural weaknesses and data decay inherent in any single B2B database.

What is a good email deliverability rate for B2B sales prospecting?

A good B2B email deliverability rate is 97% or higher, which means keeping your bounce rate below 3%. [10] Top-performing sales teams often achieve deliverability of 98% or more by using verified contact lists and proper domain authentication. [8, 13] Exceeding a 5% bounce rate can damage your sender reputation, causing providers like Google and Yahoo to filter your messages to spam, making high deliverability a foundational metric for any successful outreach campaign. [5, 7]

Why is B2B contact data quality getting worse despite AI?

B2B data quality is declining because AI amplifies the quality of the data it's given, and the foundational data is decaying faster than ever. [1] AI tools cannot prevent the root causes of data decay, such as employees changing jobs, which invalidates contact records at a rate of 22% to 40% annually. [4] While AI can identify patterns, it struggles to verify information in real-time without clean, reliable sources, and many vendors are still incentivized to sell large, unaudited lists rather than guarantee accuracy. [18, 29]

What does ZoomInfo's slowing growth mean for the sales tech market?

ZoomInfo's slowing growth, with revenue forecasts declining for 2026, signals a market shift where customers are prioritizing proven data accuracy over excess features. [19, 20, 26] After a period of rapid expansion, buyers are now scrutinizing the high cost of platforms that fail to solve the core problem of data decay. [25] This trend suggests the sales technology market is maturing, with a growing demand for more flexible, ROI-focused tools that provide verified, real-time information rather than just a large, static database. [12]

Last updated: August 2026

AI Prospecting Signals From GTM Earnings Calls