Sticky Software, Defensible from AI?

Clarivate owns benchmarks and databases that institutions are afraid to replace. That creates reliable recurring revenue and pricing power. But is it enough to defend them against AI software innovation?

Today, I’m digging into Clarivate (CLVT).

This is a company I'd never heard of. They sell a variety of software to education and other institutions. They have built up a portfolio of products via mostly acquisitions.

They are often described as an "information services provider". That's awfully vague so I dug into the actual products themselves.

  • Web of Science: A core research database used to find academic papers and track citations.

  • Journal Citation Reports: The system behind Journal Impact Factor, used to rank academic journals.

  • ProQuest: Large academic content libraries used by universities and government institutions.

  • EndNote: Citation and reference management software used by researchers.

  • Derwent Innovation: Patent search and analysis tools for IP teams and law firms.

  • CompuMark and Darts-ip: Trademark search and IP case-law databases.

  • IP renewal services: Ongoing patent and trademark renewals that generate re-occurring revenue.

  • Cortellis and DRG: Life-sciences tools used by pharma companies to analyze drug pipelines, trials, and markets.

  • AI features layered on top: Research assistants, summaries, and risk tools embedded inside existing products.

Their tools tend to be systems of record which makes software sticky. If you're building software, aim to build a system of record.

In all honestly, I'm having a hard time judging the impact of AI on this business. The tools seem ripe to benefit from an AI layer on top, as mentioned above, but the more I use AI analysis tools, the more impressed I am with the ability to build custom tools and workflows.

They have a potential AI moat due to their proprietary benchmarks but, if I were them, I'd aim for network effects (somehow). That's one of the better shields against AI software disruption. You should do the same.

With that, I'll see you tomorrow!

Nick

TL;DR

  • Clarivate is a B2B information-services and workflow-software company serving academia, intellectual-property professionals, and life-sciences teams.

  • The business is built on subscription and re-occurring revenue tied to data-heavy, mission-critical workflows with high switching costs.

  • Financial performance looks strong on adjusted EBITDA and cash flow, but GAAP profitability remains weak due to amortization, impairments, and leverage.

  • Entrepreneurs can learn how owning benchmarks, datasets, and embedded workflows creates durable customer lock-in.

The 30,000-Foot View

Clarivate sits at the intersection of data, software, and institutional decision-making. Its platforms help customers discover knowledge, protect intellectual property, and guide research, development, and commercialization decisions. Customers are typically institutions rather than individuals, which leads to slower sales cycles but very sticky renewals.

The company does not just sell raw data. It packages proprietary and licensed datasets into workflow tools that customers rely on daily. Once embedded into processes like patent filing, journal evaluation, grant discovery, or competitive intelligence, Clarivate’s products become difficult to replace.

How Clarivate makes money

  • Subscription revenue: Annual or multi-year contracts for access to software platforms and curated datasets.

  • Re-occurring revenue: Predictable renewals tied to IP maintenance, including patent and trademark renewals.

  • Transactional revenue: One-off services such as consulting, implementations, and project-based data sales.

Revenue mix (FY2024)

  • Subscription: $1,626.8M (63.6%)

  • Re-occurring: $429.8M (16.8%)

  • Transactional: $500.1M (19.6%)

Management commonly frames the business as roughly 80% recurring when subscription and re-occurring revenue are combined.

Key stats

  • Market cap: $2.14B

  • TTM Revenue: ~$2.5B

  • TTM Gross Margin: ~66%

  • Employees: ~12,000

  • Industry: Information-services and technology-enabled business services

Company History

  • 2016: Thomson Reuters sells its Intellectual Property & Science business to private-equity buyers, forming the foundation of Clarivate.

  • 2019: Clarivate goes public via a SPAC merger with Churchill Capital.

  • 2020: Acquisition of Decision Resources Group expands life-sciences intelligence capabilities.

  • 2020: Acquisition of CPA Global significantly scales the IP segment and renewal-based revenue.

  • 2021: Acquisition of ProQuest strengthens academic and library workflows.

  • 2022: Leadership transition and growing focus on portfolio rationalization.

  • 2022–2024: Divestitures and cleanup of overlapping assets, including MarkMonitor and ScholarOne.

  • 2024: New CEO appointed and increased emphasis on AI-powered academic and IP tools.

Show Me the Money

Standout financial features:

  • Lower gross margin than typical software due to cost of services.

  • Adjusted EBITDA is strong, but GAAP results are distorted by amortization and impairment charges.

  • Significant debt load creates a persistent interest-expense drag.

  • Cash generation is stronger than net income suggests, which management emphasizes.

Financial Data

Metric

FY2022

FY2023

FY2024

TTM

Revenue

$2.66B

$2.63B

$2.56B

$2.50B

Gross Profit

$1.71B

$1.72B

$1.69B

$1.64B

Gross Margin

64.1%

65.5%

66.0%

65.8%

Ops Profit

-$3.93B

-$0.73B

-$0.28B

~0.03B

Ops Margin

-147.6%

-27.9%

-10.8%

-1.3%

CapEx

$0.20B

$0.24B

$0.29B

$0.27B

Net Debt

$4.65B

$4.35B

$4.22B

$4.10B

The N.O.O.B. Nine — Competitive Powers

The Nerd Out on Business Nine is made up of Hamliton Helmer's famous "7 Powers" of competitive advantage (Scale Economies, Network Economies, Counter-Positioning, Switching Costs, Branding, Cornered Resource, and Process Power) combined with two of my own (Data Flywheel and Distribution Advantage).

Power

Score

Rationale

Branding

4/5

Brands like Web of Science, ProQuest, and Journal Impact Factor are category-defining within their niches.

Data Flywheel

3/5

Usage data improves tools and AI features, but much core content originates externally.

Process Power

4/5

Years of editorial standards, taxonomy-building, and data-governance processes create durable operational advantages.

Scale Economies

4/5

High fixed costs in data licensing and curation create leverage at scale that smaller competitors struggle to match.

Switching Costs

5/5

Products are deeply embedded in research, IP, and institutional workflows, making switching risky and expensive.

Cornered Resource

4/5

Proprietary datasets, long-term licenses, and curated indices are difficult to replicate quickly.

Network Economies

2/5

User growth does not meaningfully increase product value, improvements come mainly from internal data and curation.

Counter-Positioning

2/5

Clarivate competes at the enterprise tier with enterprise pricing, not with a structurally disruptive model.

Distribution Advantage

4/5

Deep relationships with universities, law firms, publishers, and pharma create incumbent access and renewal leverage.

Average Score: 3.6/5 - A defensible moat built on switching costs, brand authority, and data depth, without strong network effects or true counter-positioning.

Memorable Marketing

Clarivate markets itself as a neutral authority and industry scorekeeper. Instead of flashy consumer advertising, it relies on benchmarks, rankings, and reports that customers reference publicly, turning users into distribution channels.

Key campaigns and tactics

  • Top 100 Global Innovators

    • Annual patent-based ranking.

    • Earned-media driven, amplified by corporate PR teams.

    • Reinforces Clarivate as the referee of innovation quality.

  • Highly Cited Researchers List

    • Recognizes top academic researchers by citation impact.

    • Distributed organically by universities.

    • Anchors Web of Science as the default citation benchmark.

  • Journal Citation Reports

    • Annual release of Journal Impact Factor data.

    • Creates predictable spikes in relevance and renewal value.

  • AI Upgrades to Core Products

    • Positioned as productivity improvements, not abstract AI hype.

    • Strengthens existing workflows rather than launching standalone tools.

Tactical takeaways

  1. Own a benchmark customers want to share.

  2. Create annual content moments instead of constant noise.

  3. Let customers do the distribution.

  4. Tie AI messaging to concrete workflow gains.

  5. Design marketing assets that survive committee buying.

AI Uses & Opportunities

Current uses

  • Generative AI assistants embedded in academic-research and library-discovery tools.

  • AI-powered features in EndNote for summarization and journal selection.

  • Machine-learning models for IP risk and trademark analysis.

Future opportunities

  • Decision-support copilots that generate evidence-backed memos for IP and R&D teams.

  • AI-driven churn prediction and renewal-defense tooling.

  • Automation of data normalization and quality control to reduce editorial costs.

  • Vertical-specific AI agents sold as premium, auditable add-ons.

Bumps in the Road

  • Large goodwill and intangible-asset impairments have repeatedly crushed GAAP earnings.

  • High leverage limits flexibility during periods of slow growth or higher interest rates.

  • Product sprawl from acquisitions increases integration complexity.

  • Some rankings have faced criticism around transparency and gaming.

  • Dependence on third-party data sources introduces ongoing licensing risk.

Your Swipe File

  • Build systems of record.

  • Use benchmarks as compounding marketing assets. I like this one a lot.

  • Do not confuse recurring revenue with growth.

  • Be disciplined with acquisitions.