Cognizant Technology Solutions Corporation (NASDAQ: CTSH) — A Buyback-Led EPS Story Awaiting Organic Proof

Analyst: Independent Research Date: August 21, 2026 Price: $60.92 (August 20 close) · Market capitalization: approximately $27.54 billion · Enterprise value: approximately $28.58 billion Shares outstanding: 452 million at June 30, 2026 · FY-end: December · CIK: 0001058290 · Headquarters: Teaneck, New Jersey Current valuation: approximately 12.4x TTM GAAP earnings, 10.6x midpoint FY2026 adjusted EPS, 7.2x TTM EV/EBITDA, 10.6x TTM free cash flow, 9.4% TTM FCF yield, and 2.2% dividend yield

⚡ Claude’s Take

Claude’s own subjective opinion; not investment advice. The institutional body below carries no position or price target.

Call: HOLD at $60.92. Conviction: MEDIUM. Accumulate below the low-$50s; require operating proof rather than multiple expansion above $70.

The June BUY worked for the wrong reason faster than the right reason could be tested. CTSH is up 39% from the June 18 report price and 64% from its June 30 trough. Q2 then showed that the EPS floor is sturdier than feared: Financial Services grew double digits, Cognizant retired 22.5 million shares at about $51, and the full-year adjusted-EPS guide rose to $5.70-$5.82. But the quarter also exposed weak growth quality. Of 4.1% constant-currency growth, about 1.0 point came from acquisitions and 1.7 points from third-party product resale, leaving only about 1.4% underlying service growth. The reported 40-basis-point adjusted margin expansion was more than explained by roughly 100 basis points of net rupee benefit. Bookings remain high, yet Q2 bookings fell 6%, the revenue guide was cut, gross margin remains pressured, and first-half free cash flow converted only about half of net income.

At roughly 10.6x forward adjusted EPS and a 9.4% trailing FCF yield, the stock is still inexpensive. It is no longer priced for terminal decline, however, and the balance sheet has moved from net cash to roughly $1.0 billion of net debt after $1.4 billion of acquisitions and accelerated buybacks. The core franchise still has no company-wide moat; TriZetto is a real, high-margin local moat inside a largely contestable labor-and-delivery business. The decisive evidence will arrive through the renewal clock management itself described: AI has been embedded into contracts since 2025, time-and-materials work reprices within 6-15 months, and fixed-price work within 24-36 months. By 2027, currency-neutral margins and revenue per booking should reveal whether Cognizant retains AI productivity or passes it to clients.

I would not sell a durable cash generator at 10-11x earnings simply because the rebound was fast. I also would not pay up while organic service growth is near 1%, margin quality is FX-dependent, and acquisition goodwill is doing more work than disclosed AI economics. The bull case flips back on with two or more quarters of accelerating organic service growth, stable currency-neutral gross margin, normalizing DSO, and bookings converting without longer duration. The bear case wins if organic service growth reaches zero while bookings remain elevated, proving that each booked dollar is carrying less revenue, or if Project LEAP savings fail to protect margins through the 2027 renewal cycle.

Changes since June 20, 2026

The operating update is more mixed and the valuation less asymmetric than in the June report.

Prior falsifier check. The June bull thesis would have failed on two consecutive quarters of flat-to-negative organic growth with elevated bookings. That has not happened, although normalized Q2 service growth moved uncomfortably close. The June bear thesis would have failed on several quarters of accelerating organic growth with stable pricing and margins. That has not happened either. Q2 therefore weakens the bull case without confirming the secular bear case.

📈 Stock Price Action — Five-Year Event Map

Price moves are facts; attributed causes are interpretations. This section contains no recommendation or target.

The arc. CTSH closed at $77.36 on August 23, 2021, peaked around $86 in April 2022, then fell through the attrition and leadership crisis to the high-$40s in November 2022. Ravi Kumar’s turnaround helped carry the stock to a five-year high of $90.82 on February 14, 2025. The next cycle reversed violently: sector AI-deflation fears, cautious discretionary spending, and successive guidance concerns drove the stock to $37.08 intraday on June 30, 2026. Q2 results then produced a large positive company-specific reaction and the stock recovered to $60.92 by August 20. The five-year price return remains negative even after the rebound.

FactorsToday reinforces that history. CTSH retains a strongly negative Momentum loading (-0.769), positive Value (+0.209), positive Low Volatility (+0.232), and 30.2% annual specific volatility. Its all-factor model explains only 48% of return variance, leaving company-specific outcomes important. Accenture is the closest factor-related stock at 0.908 similarity, consistent with the June sector read-through. The three-month recovery is powerful, but five-year annualized performance is still negative and the stock remains about one-third below its February 2025 high.

1. Executive Summary

Cognizant is a $21.6 billion-revenue professional-services company that combines consulting, application development, systems integration, engineering, cloud, infrastructure, and business-process operations. It serves a client base concentrated in North America while delivering predominantly from India. The economic model remains recognizable: revenue equals people, bill rates, utilization, project scope, and a growing amount of third-party product pass-through; profit equals the spread between those billings and labor, subcontractor, product, travel, and delivery costs.

The business is financially sound but strategically average. It is capital-light, has produced $2.6 billion of TTM free cash flow, and earns low-to-mid-teens reported ROIC. It also operates in a fragmented market with low supply-side barriers. Competitors hire from the same labor pools, deploy the same hyperscalers and foundation models, and bid for the same transformation programs. Cognizant’s 16% operating margin trails the low-20s margins of leading India-based peers despite similar delivery geography. That gap is direct evidence against a cost advantage. Account knowledge, regulated-domain experience, transition risk, and embedded platforms create customer captivity within particular accounts, but the company does not have a network effect, unique technology standard, or global scale moat.

Q2 2026 improved the visible EPS floor without proving the strategic bull case. Revenue rose 4.1% in constant currency, Financial Services rose 11.7%, adjusted margin reached 16.0%, and adjusted EPS rose 4.6%. Yet the mix matters. About 170 basis points of revenue growth came from reselling third-party products and about 100 basis points from acquisitions. Product resale expands the top line with less gross profit per dollar, while acquired revenue reflects capital deployed rather than organic demand. After stripping both, underlying service growth was only about 1.4%. The other three segments outside Financial Services were near flat to low-single-digit constant-currency growth.

Margin quality was similarly mixed. The rupee’s depreciation supplied about 180 basis points of benefit before 80 basis points of hedge cost, for roughly 100 basis points net. Adjusted operating margin expanded only 40 basis points. Product resale, compensation, and acquisition mix pressured gross margin, while Project LEAP is still producing charges and savings that management intends largely to reinvest. This does not mean margins are deteriorating irreversibly; it means the quarter cannot be cited as proof that AI productivity is reaching shareholders.

Bookings remain supportive but no longer accelerating. TTM bookings were $29.1 billion, up 5% and equal to 1.3 times revenue. Q2 bookings declined 6%; average annual contract value fell modestly because deal terms lengthened. Longer duration can improve visibility, but it also inflates total contract value relative to near-term revenue. The full-year constant-currency revenue guide was narrowed to 4.0%-5.5%, with 150 basis points of inorganic contribution unchanged. The midpoint therefore implies roughly 3.25% organic growth. Q3 midpoint organic growth is closer to 2.55%. These figures are compatible with a stable franchise, not a broad acceleration.

Capital allocation is the principal near-term support and the principal new risk. Cognizant repurchased 22.5 million shares for $1.153 billion in Q2, including a $500 million accelerated repurchase at $51.54. It bought inexpensive shares and reduced quarter-end shares to 452 million from 479 million at year-end. At the same time, it spent roughly $1.39 billion on 3Cloud and Astreya. Preliminary accounting assigned approximately $1.00 billion, or 72% of their combined price, to goodwill. The company drew $1.0 billion on its revolver; cash plus short-term investments fell to $1.05 billion and debt including leases reached $2.09 billion. Leverage remains modest, but the old net-cash margin of safety is gone.

The embedded expectations are no longer extreme. Current equity value is about $27.5 billion and enterprise value about $28.6 billion. Against TTM EBITDA of $4.0 billion and FCF of $2.6 billion, CTSH trades at 7.2x EV/EBITDA and 10.6x FCF. Against the $5.76 midpoint adjusted-EPS guide, it trades at 10.6x. Those are cheap multiples for a durable cash generator and reasonable multiples for a no-moat company with low organic service growth and an uncertain AI renewal cycle.

The central question is measurable: can Cognizant separate revenue growth from labor hours while retaining a fair share of productivity? Management says over 40% of software development is AI-assisted and cites more than 8,000 AI engagements. Neither statistic tells investors what clients pay. Time-and-materials contracts reprice in 6-15 months, and fixed-price contracts typically in 24-36 months. By 2027, bookings conversion, constant-currency service growth, gross margin excluding FX, and revenue per employee should answer whether AI expands the market faster than it deflates delivery.

2. Business Overview

2.1 Revenue model and service mix

Cognizant provides consulting and technology services, engineering and research-and-development services, infrastructure and cloud operations, and business-process services. The company combines client-facing account teams with a global delivery pyramid. At year-end 2025 it employed approximately 351,600 people; June 2026 headcount was 356,700, including acquired employees, even as organic headcount declined sequentially.

The contract mix is evolving. In Q2 2026, time-and-materials revenue was $2.305 billion, 42.1% of total; fixed-price revenue was $2.690 billion, 49.1%; and transaction- or volume-based revenue was $486 million, 8.9%. Fixed-price mix has risen over three years. That change is strategically logical because clients want outcomes rather than hours. Economically it changes the timing of AI benefits. On an existing fixed-price contract, faster delivery can lift provider margin. At renewal, competitive bidding transfers some or all of that productivity to the client. Time-and-materials contracts pass lower hours through more directly and reprice sooner. Outcome-based contracts can be the most defensible when Cognizant owns workflow expertise or a platform such as TriZetto.

Third-party product resale complicates revenue interpretation. Cognizant increasingly combines licenses, cloud capacity, hardware, and external products with implementation. This can deepen client relationships and increase wallet share. It also creates revenue that has lower gross margin and does not represent Cognizant labor or intellectual property. Product resale contributed 170 basis points to Q2 growth and 160 basis points to first-half growth. Investors should track gross profit and service growth, not revenue alone.

2.2 Segments

Financial Services is doing the work. Even after subtracting the approximate 250-basis-point product contribution, its growth was around 9%, consistent with modernization, payments, data, risk, and AI demand. The other segments were near flat. Health Sciences remains economically superior because it combines delivery relationships with TriZetto products and workflows. Management disclosed that TriZetto now exceeds $1.1 billion in annual revenue, grew faster than Cognizant in the first half, and carries a substantially higher margin. It is the clearest instance where domain software and workflow captivity move the model beyond labor arbitrage.

Products & Resources includes Belcan’s aerospace and engineering work. It broadens total addressable work into engineering R&D and can create longer-duration relationships, but its 12.5% margin is dilutive. CMT serves sophisticated technology and media clients that often have strong internal capabilities and purchasing leverage. Its margin improved from 12.8% to 14.4%, but 350 basis points of its Q2 revenue growth came from products, meaning underlying service revenue likely declined.

2.3 Geography and concentration

North America generated $4.127 billion, or 75.3%, of Q2 revenue; Europe generated $1.027 billion, or 18.7%; the rest of the world generated $327 million. Demand is therefore concentrated in US enterprise budgets, while supply is concentrated in India. Cognizant has no disclosed customer representing a dangerous share of total revenue, and its vertical mix is balanced. Geography nonetheless exposes results to US discretionary spending, financial-services cycles, visa policy, rupee movements, Indian wage inflation, and local labor regulation.

2.4 Unit economics

The primary unit is a deployed associate or delivery team. The company has begun reporting revenue and operating income per associate, but Q2 comparables excluded trainees not fully deployed. That definition change is sensible for productivity management and reduces time-series usefulness. Reported revenue per associate and operating income per associate rose 4.6% and 7.1%, respectively, after recasting comparables. The correct test is whether those gains persist after controlling for acquisitions, product resale, currency, and trainee definitions.

The labor pyramid remains central. Management expects to hire about 20,000 graduates by year-end while Project LEAP removes roles and keeps total headcount roughly range-bound. This is consistent with replacing expensive or underutilized layers with lower-cost, AI-enabled juniors. It can protect margin, but it is also an admission that AI alters the supply curve. The firm must capture the productivity before competitors and clients price it away.

2.5 Customer economics and revenue durability

Cognizant does not disclose subscription-style net retention, remaining performance obligations by segment, or top-ten client concentration. Revenue durability must therefore be inferred from contract structure, segment behavior, transition requirements, and cash collection. The evidence is favorable but not absolute. Managed infrastructure, application maintenance, claims administration, and business-process operations are difficult to remove abruptly. A client can reduce discretionary scope within a quarter, but replacing an incumbent across critical systems requires knowledge transfer, parallel running, security approvals, testing, and operational-risk acceptance. That produces a gradually declining tail rather than a sudden churn event.

Large-deal vendor consolidation strengthens and weakens durability at the same time. Cognizant gains a larger share of the client’s estate and often signs a multi-year contract. The client awards that share precisely to lower total cost, which increases pricing pressure and can require upfront transition spending. Long duration therefore improves visibility but not necessarily lifetime economics. Investors should distinguish retention created by unique workflow value from retention purchased through contractual savings.

Health Sciences illustrates the stronger form. TriZetto embeds Cognizant in payer administration, claims, and data flows where error costs and regulation make transition unattractive. Financial Services combines regulatory knowledge and legacy-system complexity, but large banks have sophisticated procurement and internal engineering. P&R and CMT clients can be more cyclical or technically self-sufficient. The segment-margin dispersion—Health Sciences near 20%, P&R near 12.5%—is consistent with different levels of captivity.

Receivables provide a commercial-quality check. DSO of 88 days is not alarming for a global enterprise-services firm with large contracts and product pass-through, but the five-day year-over-year increase shows that reported revenue is consuming more working capital. If the increase reverses as related payables settle, management’s mix explanation is validated. If it persists alongside longer contract duration, the company may be financing client transitions or accepting less favorable terms to win work.

Business-overview verdict: a diversified, durable, capital-light franchise with one meaningful platform asset and many sticky accounts, but still predominantly a people-and-process business. Product resale and acquisitions make revenue growth look stronger than the organic service engine.

3. Industry Dynamics

3.1 Structure and competitive intensity

Global IT and business-process services is large, fragmented, and structurally contestable. Accenture, TCS, Infosys, Cognizant, Capgemini, HCLTech, IBM Consulting, Wipro, EPAM, Genpact, the Big Four, hyperscaler service units, and many specialists compete by account and work package. No provider controls the market. The main production inputs—engineers, domain experts, cloud platforms, packaged software, and foundation models—are mobile or available to competitors.

Greenwald’s moat taxonomy yields no broad supply advantage. Cognizant uses the same Indian talent pool as firms with superior margins, so it does not own a unique cost curve. There is no network effect: another client’s use of Cognizant does not inherently increase value for an existing client. Economies of scale help with recruiting, training, compliance, sales coverage, and partner certifications, but the market is global enough that several competitors share similar scale. The meaningful barrier is demand-side customer captivity within an account: systems knowledge, security approvals, regulated workflow expertise, transition cost, and relationships.

Competitive intensity is rising for three reasons. First, clients continue building global capability centers in lower-cost markets, internalizing work that once went to vendors. Second, AI tools increase effective engineering capacity, which is economically similar to an outward shift in industry supply. Third, hyperscalers and model vendors supply the key technology and can influence economics upstream. Partnerships with Microsoft, Google, AWS, NVIDIA, Anthropic, and OpenAI are table stakes rather than exclusive advantages.

3.2 The AI capital cycle

The traditional IT-services capital cycle centered on labor. High demand led providers to hire, wage inflation followed, and utilization later normalized. AI changes the relevant capacity unit from headcount to output. If a developer using a coding assistant produces more code, industry supply expands even without hiring. In a competitive market, the long-run benefit generally flows to customers through lower prices or greater scope.

That is the bear mechanism: fewer hours per task reduce time-and-materials revenue; fixed-price renewals reset lower; clients build more internally; provider differentiation shrinks because every firm accesses similar models. The countervailing bull mechanism is demand elasticity. Lower cost can unlock projects that were previously uneconomic, legacy estates require modernization before AI can operate, regulated enterprises need governance and integration, and autonomous workflows enlarge the domain from systems integration into business operations.

Cognizant argues that its reachable market could expand from approximately $1 trillion of systems integration into several trillion dollars of enterprise operations. That is a management hypothesis, not an addressable-revenue fact. The credible portion is narrower: enterprises need data cleanup, application modernization, agent integration, cybersecurity, evaluation, observability, change management, and industry controls. The uncertain portion is value capture. Model vendors, software platforms, clients, and integrators all claim the same productivity pool.

The renewal schedule turns theory into evidence. Time-and-materials work generally renews or reprices within 6-15 months, while fixed-price contracts average 24-36 months. Cognizant says active AI embedding began in 2025 and roughly half of the book remains to renew. If margins stay stable after currency normalization and service revenue accelerates through 2027, demand expansion and provider value capture are winning. If bookings stay elevated while organic service revenue stalls, the deflation mechanism is winning.

3.3 Peer signals

Recent peer work is consistent with pressure, not extinction. Infosys has observed AI productivity being passed to clients and growth supported by realization rather than volume. EPAM has described AI-native delivery as replacing revenue before it creates incremental growth; its organic growth slowed and guidance fell. IBM Consulting faces aggressive pricing and average people-business economics. Accenture retains the strongest brand, scale, and C-suite position but still experiences booking volatility and restructuring.

These cross-reads matter because CTSH’s Q2 is not isolated. Industry clients remain cautious on discretionary programs and prioritize infrastructure and data readiness. Large cost-takeout deals can lift bookings while delaying revenue and compressing unit price. A provider can gain share and still show low organic growth if the market’s price per unit falls.

3.4 Supply response

Project LEAP, lower lateral hiring, graduate recruitment, and flatter headcount are Cognizant’s supply response. Rivals are doing versions of the same. This should reduce wage pressure and prevent catastrophic margin decline. It does not create a durable advantage if every provider can redesign its pyramid. Acquisitions such as 3Cloud and Astreya add scarce skills and partner credentials faster than internal development, but high goodwill signals that Cognizant is paying sellers for those expected rents.

Industry verdict: structurally average and becoming more competitive. AI expands the volume of possible transformation while increasing delivery supply and client purchasing power. The industry can grow without supporting higher provider returns. Cognizant’s outcome depends on local account captivity and execution, not a favorable industry structure.

4. Competitive Position

4.1 Moat map

The broad conclusion remains no company-wide moat. Cognizant can earn attractive returns because the business requires little tangible capital, not because competitors cannot enter. Reported ROIC around 13% in the ROIC.ai TTM feed remains above a reasonable cost of capital, but it has fallen from the high-teens framing in the June report as acquisition goodwill and balance-sheet deployment increased. Goodwill of $8.083 billion now equals 56% of equity. Returns on tangible operating capital remain high because people and relationships are expensed rather than capitalized.

4.2 TriZetto as the exception

TriZetto supplies healthcare-payer software, claims, administration, and workflow capabilities. Its more than $1.1 billion revenue base, above-company growth, and superior margin provide rare quantitative evidence of proprietary value. Healthcare integration is expensive, regulated, and operationally risky. Clients face switching costs from data conversion, workflow redesign, testing, payer/provider connectivity, and compliance. This is a narrow demand-side moat and a reason Health Sciences earns nearly 20% segment margin.

TriZetto is not large enough to define the enterprise. At roughly 5% of company revenue, it supports Health Sciences quality but does not protect Financial Services, engineering, cloud, or infrastructure work. Management could enhance value by reporting platform revenue, recurring mix, retention, margin, and organic growth separately. Without that disclosure, investors cannot distinguish platform economics from bundled services.

4.3 Turnaround execution

Ravi Kumar has improved large-deal sourcing, attrition, account focus, partner coverage, and capital return since 2023. Financial Services growth and new-logo wins suggest better sales execution. Seven Q2 deals exceeded $100 million, including three new logos. New and expansion bookings grew mid-teens in the first half even as total Q2 bookings fell.

The turnaround earns credit for operational stabilization, not a new moat. Bookings can be won through price, vendor consolidation, and longer duration. Project LEAP should improve organizational speed and the labor pyramid, but restructuring is imitable. The key question is whether Cognizant wins a disproportionate share of AI-era work at attractive gross profit.

4.4 Share stability and relative economics

Cognizant lost growth and prestige during the 2019-2022 leadership and attrition period, then recovered. That history fails the strict Greenwald share-stability test. A protected franchise should not lose ground so easily when management execution deteriorates. Conversely, the rapid recovery shows the client base was not destroyed. Account captivity gave management time to repair delivery and sales.

The margin gap remains the simplest competitive scoreboard. Cognizant’s adjusted margin near 16% is respectable, but TCS and Infosys have historically earned more from similar offshore delivery. Some difference reflects portfolio—Cognizant’s North American onshore mix, acquired engineering, and product resale—but not all. Until the gap narrows through organic productivity rather than FX, Cognizant should not receive a quality-premium multiple.

Competitive-position verdict: localized customer captivity and one significant platform franchise inside a broadly no-moat provider. Execution has improved; structural advantage has not.

5. Growth History and Forward Opportunities

5.1 Historical bridge

Revenue grew from roughly $18.5 billion in 2021 to $21.1 billion in 2025, a low-single-digit compound rate. Growth was uneven: the post-pandemic surge gave way to the 2023 discretionary slowdown, followed by recovery aided by Belcan and other acquisitions. FY2025 reported growth was 7.0% and constant-currency growth 6.4%, but Belcan contributed materially. Q1 2026 grew 3.9% in constant currency; Q2 grew 4.1%.

The current run rate must be decomposed:

• Approximately 1.0 percentage point of Q2 growth came from acquisitions.
• Approximately 1.7 points came from third-party product resale.
• The residual underlying service growth was approximately 1.4%.
• Financial Services supplied most of the organic growth.
• Astreya closed June 22 and contributed only about one week, so its full quarterly inorganic effect arrives in Q3.

This is not a collapse. It is a narrower engine than the reported 4.1% suggests.

5.2 Bookings and conversion

TTM bookings of $29.1 billion represent 1.3 times revenue. This appears ample, but total contract value is not revenue. Contract duration, renewals, pass-through products, cancellation terms, productivity commitments, and transition periods affect conversion. Q2 ACV decreased modestly because duration lengthened. That means book-to-bill can remain high while near-term revenue growth slows.

The cleanest monitoring measure is a conversion triangle:

Q2 produced a healthy bookings level but warnings in quarterly bookings, organic service growth, gross-margin quality, and DSO.

5.3 AI and modernization

Cognizant reports more than 8,000 AI engagements and AI assistance in over 40% of software development. The company is training 5,000 Frontier-certified engineers and 10,000 business operators and building broad partnerships with model and cloud vendors. These activities show adoption and capability formation. They do not show revenue or profit.

Revenue opportunities include legacy application modernization, data engineering, cloud migration, model evaluation, agent orchestration, cybersecurity, observability, governance, and business-process redesign. The highest-value work sits at the intersection of industry context and technology. Financial Services and Health Sciences are logical beneficiaries because data, regulation, and workflow complexity create implementation demand.

The threat is that AI reduces coding and testing hours faster than new scope appears. Cognizant’s move toward fixed-price and outcome-based work is directionally correct because it monetizes results rather than labor. Long-term economics depend on contract design: the provider must share enough productivity to win while keeping enough to protect margin.

5.4 Cloud and infrastructure M&A

3Cloud adds Microsoft Azure, data, and AI capability. Astreya adds infrastructure lifecycle, data-center, workplace, and cloud operations. Combined with Belcan, Thirdera, and prior acquisitions, they expand Cognizant into engineering, cloud, ServiceNow, and AI infrastructure. These markets can grow faster than legacy application maintenance.

Acquisition growth is not free. 3Cloud and Astreya required approximately $1.39 billion, and 72% of purchase price became goodwill. Customer relationships represent another $380 million. This accounting says much of the value resides in people, relationships, and future synergy rather than separable technology. If employees leave or clients do not cross-buy, returns disappoint.

5.5 Guidance bridge

Full-year revenue guidance of $22.04-$22.35 billion implies 4.4%-5.9% reported growth and 4.0%-5.5% constant-currency growth. With 150 basis points of acquisition contribution, organic constant-currency growth at midpoint is approximately 3.25%. Q3 guidance of 3.8%-5.3% constant currency includes approximately 200 basis points inorganic, placing organic midpoint near 2.55%.

The upper end assumes some fourth-quarter improvement in short-cycle discretionary work. Management explicitly said discretionary demand remained weak, which caused the guide cut. The lower end is compatible with continued Financial Services strength and little recovery elsewhere.

Growth verdict: durable low-single-digit organic growth with strong Financial Services and meaningful acquisition support, but no broad acceleration. AI and modernization create real volume opportunities; current public KPIs do not prove monetization.

6. Financial Quality

6.1 Income statement

Q2 operating income included an $81 million benefit from reversing part of an India defined-contribution obligation and $84 million of Project LEAP restructuring charges. Both were excluded from adjusted results, making the non-GAAP comparison reasonable. The more important adjustment is economic rather than accounting: currency delivered about 100 basis points of net margin benefit. Without it, the underlying margin direction was negative.

TTM revenue was $21.642 billion, operating income $3.520 billion, EBITDA $3.995 billion, and net income $2.220 billion. ROIC.ai reports a 33.4% gross margin, 16.3% operating margin, 10.3% net margin, and 13.2% ROIC. The effective tax rate in the TTM feed is elevated by prior discrete tax effects, so adjusted EPS is a better near-term guide than mechanically annualizing GAAP net income.

6.2 Cash conversion

First-half net income was $1.298 billion, operating cash flow $832 million, and capex $175 million. Simple FCF was therefore $657 million; management’s adjusted measure was $652 million. Conversion of roughly 51% is weak for a capital-light services business. Management guides full-year FCF to 90%-100% of net income, requiring a pronounced second-half release.

TTM cash flow remains strong: operating cash flow $2.917 billion, capex $319 million, and FCF $2.598 billion. The discrepancy between healthy TTM cash flow and weak first-half conversion reflects seasonality and working capital, but it deserves monitoring. Accounts receivable increased to $4.780 billion, and Q2 DSO reached 88 days, five days higher year over year. Management says product and deal mix created a matching payable increase and is cash neutral. The claim is plausible but should be tested through the H2 catch-up.

6.3 Balance sheet

At June 30, Cognizant held $1.038 billion cash and $13 million short-term investments. Long-term borrowings were $1.527 billion; short-term debt was $33 million; lease liabilities were approximately $534 million. Including leases, total debt was about $2.094 billion and net debt about $1.043 billion. Current assets of $7.559 billion covered current liabilities of $3.464 billion more than twice.

There is no solvency problem. Net debt is about 0.26x TTM EBITDA, interest coverage is ample, and the revolver provides liquidity. The issue is optionality. In June the balance sheet could support acquisitions and aggressive repurchases from cash; now further large capital deployment competes with maintaining liquidity and restoring cash conversion.

6.4 Accounting quality

Accounting is generally clean for a services company. Stock compensation is modest, capex is low, restructuring is separately identified, and adjusted EPS retains ordinary operating costs. The company changed segment-cost allocations beginning in 2026 and recast 2025, limiting older segment-margin comparability. Associate productivity comparables now exclude undeployed trainees. These changes are not necessarily aggressive, but they require consolidated and definition-consistent analysis.

Goodwill is the main quality issue. It rose to $8.083 billion from $7.106 billion at year-end, while other intangibles reached $1.675 billion. Goodwill plus intangibles is two-thirds of equity. This does not create a cash liability, but it means reported book value and ROIC depend on management’s acquisition assumptions. A future impairment would confirm poor allocation rather than cause the economic loss.

6.5 Multi-year earnings-quality scorecard

The five-year record is more stable than the latest half-year cash result. Revenue has moved upward with only shallow organic contraction, while capex has remained near 1%-2% of sales. Depreciation and amortization is larger than capex because acquired intangibles are amortized and the physical asset base is small. That difference helps cash conversion but is not free: it reflects prior acquisition spending outside operating cash flow.

Share count is the most important bridge from enterprise earnings to per-share earnings. Average diluted shares in the TTM period were approximately 478 million, while quarter-end shares were 452 million and the FY2026 diluted guide is about 460 million. A simple TTM P/E based on TTM average-share EPS understates the earnings power of the smaller current share base, while a forward adjusted P/E can overstate operating progress because it incorporates the full denominator benefit. Both should be presented, as this report does.

Gross profit is the cleanest near-term operating measure. Q2 gross profit increased about 3.6%, slower than 4.5% revenue, even with currency help. Operating income grew faster because SG&A benefited from efficiency and the one-time India obligation reversal in GAAP results. Adjusted operating margin removes the one-time reversal and restructuring, but it cannot remove all mix and currency effects. The hierarchy of evidence is therefore: gross-profit dollars and margin excluding currency; adjusted operating profit; then adjusted EPS after share count.

Balance-sheet quality remains sufficient for the business model. There is no inventory risk, physical capex is low, the current ratio exceeds 2x, and net leverage is small. The main assets are receivables and goodwill, so the failure modes are commercial and allocative rather than financial: customers pay more slowly, acquired people leave, cross-selling fails, or contract prices decline. Those risks emerge gradually and should be visible before they threaten liquidity.

6.5 Financial-quality verdict

The business remains capital-light and cash generative with modest leverage. TTM earnings are cash-backed. Q2 growth and margin composition are weaker than reported, and first-half conversion is below standard. The financial profile supports durability, not premium quality.

7. Capital Allocation

7.1 Framework and recent deployment

Management balances dividends, repurchases, and capability M&A. In the first half of 2026 it used $1.607 billion for buybacks, approximately $316 million for dividends, and $1.334 billion net cash for acquisitions. These uses exceeded internally generated cash, leading to the $1.0 billion revolver draw and a lower cash balance.

The repurchase was economically attractive. Q2 purchases averaged roughly $51, including the ASR at $51.54. That price was about 9x midpoint adjusted EPS and near a double-digit normalized FCF yield. Retiring 22.5 million shares reduced the denominator by nearly 5% in one quarter. Management expects approximately 460 million diluted shares for FY2026, supporting adjusted EPS even when net income growth is slower.

The risk is extrapolation. A buyback creates value when shares trade below intrinsic value and the balance sheet remains resilient. The current price is about 18% above the Q2 average purchase price, and net debt has replaced net cash. Future repurchases should compete against debt repayment and acquisition integration rather than proceed automatically.

7.2 Acquisition scorecard

3Cloud generated approximately $596 million of goodwill; Astreya approximately $407 million. Together, $1.003 billion of goodwill and $380 million of customer relationships absorb nearly all purchase consideration. The companies may still earn attractive returns through cross-selling and capability scarcity. The burden of proof is high because accounting identifies few hard or proprietary assets.

An acquisition should clear three tests: organic growth above Cognizant, margin at or above the company after integration, and retention of key people and clients. Management has not yet supplied those KPIs. Astreya contributed only one week to Q2, so no conclusion is possible.

7.3 Project LEAP

Project LEAP is expected to cost $230-$320 million and save $200-$300 million during 2026, with fuller benefits in 2027. Q2 charges were $84 million. Management says savings will primarily fund investment and are as much about growth as margin. Analysts should therefore avoid adding the full savings amount to earnings.

Strategically, LEAP flattens management, changes the pyramid, reduces underutilized roles, and funds AI capability. Hiring 20,000 graduates while keeping headcount range-bound suggests substitution within the workforce rather than simple downsizing. If revenue per associate improves and gross margin holds without currency, the program is working. If savings merely offset pricing and acquisition dilution, it protects rather than creates value.

7.4 Dividend and dilution

The quarterly dividend is $0.33 per share, or $1.32 annualized, a 2.2% yield at the current price. The payout consumes roughly one-quarter of normalized FCF and is secure under the scenarios considered. Stock compensation is more than offset by repurchases. The board had $2.3 billion remaining under authorization at June 30.

7.5 Management incentives and insider behavior

Executive incentives emphasize revenue, adjusted margin, adjusted EPS, strategic goals, and relative/absolute shareholder return. Those measures are broadly aligned, although adjusted EPS can be enhanced by leverage-funded repurchases and acquisitions. Investors should emphasize organic growth, FCF per share, and ROIC after goodwill.

The June 10 director open-market purchase near $52 remains a mild positive signal. August sales by CFO Jatin Dalal and John Sunshin Kim occurred under 10b5-1 plans adopted in May and are not informative discretionary selling. No new open-market insider purchase appeared after June 20.

Capital-allocation verdict: Q2 repurchases were excellent arithmetic; recent M&A is strategically coherent but return-unproven. With net debt and weak first-half conversion, the hurdle rate for further deployment should rise.

8. Changes and Headwinds — Last Two Years

The turnaround since 2023 is real. Attrition normalized, large-deal sourcing improved, revenue returned to growth, margins recovered, and Financial Services reaccelerated. Cognizant added Belcan, Thirdera, 3Cloud, and Astreya, rebuilt cloud and engineering capability, and pushed account teams toward AI-enabled outcomes.

The last two years also changed the risk profile:

1. AI moved from demand narrative to contract economics. More than 40% of software development is AI-assisted, so productivity is already present. Renewals now determine who captures it.
2. Revenue quality weakened in Q2. Product resale and M&A supplied most headline growth.
3. Bookings duration increased. Large TCV gives visibility but less near-term conversion than the ratio alone implies.
4. Discretionary work stayed weak. Management cut the revenue-guide ceiling and still needs a better Q4 for the high end.
5. The balance sheet changed. Cash-funded optionality became modest net debt after simultaneous M&A and repurchases.
6. Gross margin is under pressure. Product resale, compensation, and acquisitions offset currency and efficiency.
7. The pyramid is being redesigned. Project LEAP plus graduate hiring should lower cost, but restructuring is also a signal that the old labor model needs adaptation.
8. Reported productivity definitions changed. Excluding undeployed trainees improves management relevance while complicating comparisons.
9. India listing optionality emerged. Management is evaluating a primary or secondary India listing, with no decision. It could broaden ownership but does not change intrinsic value by itself.
10. Governance noise remains low. A legacy derivative settlement received preliminary approval; no new operating liability was identified.

The net effect is a financially stronger EPS floor but a more complicated read on organic economics. Shareholders benefit from reduced shares today; future value still depends on service growth and margins.

9. Risk Analysis

The risk distribution is secular and economic rather than solvency-related. A realistic adverse outcome is a long period of flat revenue, lower margins, and low valuation, not a sudden failure. Diversification, low capex, and embedded accounts give management time to adapt. That same durability can conceal value destruction if buybacks offset shrinking operating value per share.

10. Valuation Discussion — Embedded Expectations

10.1 Reconciled current valuation

ROIC.ai’s June-quarter valuation table used the June 30 close of $38.73. For an August 21 report, enterprise value must be rebuilt from current price and current shares. At $60.92 and 452 million shares, market capitalization is $27.536 billion. Adding $2.094 billion of debt including leases and subtracting $1.051 billion of cash and short-term investments gives enterprise value of approximately $28.579 billion.

The discount remains visible. On comparable current inputs, Accenture trades around 9.1x EV/EBITDA and 14.2x TTM earnings; EPAM around 6.7x EV/EBITDA and 13.7x earnings; IBM much higher because of software mix. CTSH deserves a discount to Accenture’s franchise quality and may deserve a premium to a contracting engineering peer if Financial Services remains strong. The peer range does not imply a dramatic mispricing.

10.2 Embedded operating expectations

At 10.6x midpoint adjusted EPS, the market does not require high growth. With a 2.2% dividend and continued repurchases, flat aggregate earnings could still produce modest per-share growth. The valuation roughly embeds low-single-digit organic revenue, a margin near 16%, normal cash conversion, and no severe AI-driven decline.

That differs from June, when a 7.8x forward multiple and 12.6% FCF yield embedded persistent erosion. The current price recognizes durability and buyback accretion. Upside now requires either better organic economics or a further multiple re-rating without proof.

10.3 Scenario framework through 2028

These scenarios express enterprise and equity values, not per-share targets.

A probability-weighted center using 25% bear, 50% base, and 25% bull lies close to current equity value. The range is wide because AI value capture is unresolved. The base case does not require a moat; it requires account captivity, adequate cost action, and continued low-single-digit demand. The bull case requires evidence not yet visible in product-adjusted service growth. The bear case requires persistent revenue-per-booking decline and currency-neutral margin compression.

10.4 Earnings-power and reverse-expectations cross-check

An earnings-power approach is appropriate because Cognizant is mature, capital-light, and not dependent on large maintenance capex. Normalizing TTM operating income near $3.5 billion for a mid-20s cash tax rate and adding back after-tax interest produces operating earnings broadly consistent with $2.5-$2.7 billion of sustainable owner earnings when working capital is normal. That is also close to TTM FCF. Capitalizing that stream at 10%-11% before growth yields an enterprise value in the mid-$20 billions; modest durable growth moves value into the low-$30 billions. Current enterprise value sits between no-growth earnings power and a modest-growth case.

The reverse question is more useful than choosing a single multiple. At current equity value, approximately $2.5 billion of normalized FCF needs to remain durable and grow only modestly for a 9%-10% required return after dividends and repurchases. The market is not assuming collapse. It is also not capitalizing the $5-$6 trillion operations narrative, the 8,000 AI engagements, or a return to peer-leading margins. That is a reasonable expectations set given the available evidence.

The sensitivity is asymmetric around margin. A 100-basis-point operating-margin change on roughly $22 billion of revenue is about $220 million of pre-tax income, or roughly $160 million after tax. At an 11x capitalization, that is about $1.8 billion of equity value before second-order effects. Q2’s approximate 100-basis-point net currency benefit is therefore economically large enough that investors cannot treat the reported 40-basis-point expansion as noise. Currency-neutral margin is a valuation input, not a cosmetic adjustment.

10.5 Cash-return sensitivity

Repurchases at 9-11x earnings can add several percentage points to EPS growth if FCF stays near $2.5 billion. They cannot offset permanent operating decline indefinitely. Funding buybacks with debt creates a one-time acceleration and adds interest expense. The sustainable source of per-share value is FCF after maintaining capability and acquisition returns.

Valuation verdict: inexpensive, no longer distressed. Current price approximately balances a durable cash franchise against low-quality growth and AI-renewal risk. The next re-rating should be earned through organic service growth and currency-neutral margin evidence.

11. Variant Perception

Optimistic consensus argument. Cognizant has repaired execution, Financial Services is leading peers, bookings remain 1.3x revenue, AI modernization expands the market, and repurchases at low multiples create visible EPS growth. The Q2 price reaction suggests investors believe the EPS floor is more important than the guide cut.

Skeptical consensus argument. Headline growth is acquired or resold, adjusted margin is currency-supported, cash conversion is weak, and large deals are longer-duration cost takeouts. Cognizant is using M&A and leverage-funded repurchases to disguise a slow organic services melt.

Variant view. Both sides overstate their case. Q2 does not show a melting franchise: Financial Services growth is real, TriZetto is a genuine platform asset, TTM FCF is strong, and leverage is modest. Q2 also does not show an AI winner: service growth excluding products and acquisitions was around 1.4%, underlying margin likely contracted, and no AI revenue or gross-profit KPI exists. The most differentiated conclusion is that EPS durability has improved faster than enterprise value creation.

Five assumptions decide the outcome:

1. Bookings quality: new and expansion ACV must convert into revenue without persistent duration inflation.
2. AI value capture: revenue per outcome can fall, but gross profit per employee must rise enough to compensate.
3. Segment breadth: Financial Services strength must spread to Health Sciences, P&R, or CMT.
4. Capital allocation: 3Cloud and Astreya must earn above the cost of capital after goodwill, while repurchases remain below intrinsic value.
5. Cash conversion: H2 working capital must normalize without clients demanding structurally longer terms.

Factor positioning adds timing context. Momentum remains strongly negative despite the three-month rebound; value and low-volatility loadings are positive. The stock is not a crowded quality compounder. A further recovery can occur if company-specific evidence improves, but the easy mean-reversion from the June panic has already happened.

12. Fact vs. Interpretation

13. Open Questions

1. What percentage of bookings is renewal, expansion, and new logo by annual contract value, and how has duration changed?
2. What are organic revenue, gross margin, and retention for Belcan, 3Cloud, and Astreya?
3. How much Q2 gross profit came from third-party products, and what services attach rate follows resale?
4. Can management disclose AI revenue, AI-influenced bookings, price per outcome, or gross-profit uplift rather than engagement counts?
5. How much of the approximately 100-basis-point rupee benefit reverses under current hedges?
6. What bridge takes first-half FCF conversion from 51% to the full-year 90%-100% target?
7. Does DSO return toward the low-80s as product payables settle, or is 88 days the new large-deal normal?
8. What is TriZetto’s recurring revenue, organic growth, retention, and standalone margin?
9. How much of Project LEAP savings is reinvested, how much protects pricing, and how much reaches operating profit?
10. What cash floor and leverage limit govern the remaining $2.3 billion repurchase authorization?
11. Would an India listing create incremental demand or merely administrative complexity?
12. What final monetary or governance terms arise from the derivative-action settlement?

14. What Must Be True

Constructive case

• Organic constant-currency service growth rises from roughly 1%-3% toward mid-single digits as large deals convert.
• Gross margin and adjusted operating margin remain stable after removing currency, product mix, and one-time items.
• Financial Services growth broadens into at least one other segment.
• TTM bookings stay above revenue without further duration-driven ACV erosion.
• H2 cash conversion restores the full-year target and DSO normalizes.
• 3Cloud and Astreya grow faster than Cognizant and approach company margin within two years.
• Repurchases remain below conservative intrinsic value and do not push leverage materially higher.

Constructive-case falsifier: two consecutive quarters of flat-to-negative organic service growth with bookings still near 1.3x, combined with currency-neutral margin pressure. That would show declining revenue per booked scope.

Adverse case

• AI reduces time-and-materials hours and forces lower fixed-price renewals faster than new scope develops.
• Project LEAP savings are consumed by pricing and acquisition dilution.
• Financial Services slows before other segments recover.
• Product resale and M&A continue to supply most reported growth.
• Working capital remains structurally heavier and FCF settles below net income.
• Acquisition goodwill fails to produce incremental organic growth or margin.

Adverse-case falsifier: several quarters of accelerating organic service growth, stable or rising currency-neutral gross margin, improving revenue and operating income per associate on a consistent definition, and clean bookings conversion. That would show AI-enabled demand and productivity are outweighing deflation.

Monitoring dashboard

15. Source Appendix Summary

The analysis prioritizes Cognizant’s 2025 Form 10-K; Q1 and Q2 2026 Forms 10-Q; Q2 earnings release, supplement, and call; 2026 proxy; relevant Forms 8-K and 4; and the five-year EDGAR corpus. Current price and statements were cross-checked through ROIC.ai, with enterprise value recomputed from the August 20 price and June 30 balance sheet. Factor data came from FactorsToday. Sector context uses current public filings and calls from Accenture, EPAM, IBM, and Infosys. The complete list appears in Appendix B.

APPENDIX A — Standard Diligence Questionnaire

Cognizant Technology Solutions Corporation (NASDAQ: CTSH) · 2026-08-21

Supplemental to the research memo. Facts, interpretations, and assumptions are labeled where the distinction is material.

General

What thoughtful questions have other investors asked about this company?

The core question is whether generative AI expands the number and scope of enterprise transformations faster than it reduces billable labor. The sharper version is not “does AI help Cognizant?”—it plainly helps delivery—but “who keeps the productivity at renewal?” Other recurring questions are: Why is Cognizant’s margin near 16% when leading Indian peers often earn low-20s margins from similar labor? Are 1.3x bookings high-quality growth or long-duration cost-takeout contracts? Why is third-party product resale supplying so much revenue growth? Do 3Cloud and Astreya add scarce capabilities or purchase growth at high goodwill? Can Project LEAP improve the pyramid without damaging delivery? Was the June low a secular warning or a sector panic? After the rebound, is the share-count story running ahead of enterprise-value growth?

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? [INTERPRETATION] Operating earnings are mid-cycle. Revenue growth is below a healthy discretionary-spending cycle, while adjusted margin is near the high end of Cognizant’s recent range. The margin is not a clean peak because Q2 benefited by roughly 100 basis points from currency and still carries acquisition and product-mix pressure. Per-share earnings are benefiting from an unusually aggressive repurchase cycle.

Driven by the external environment or internal actions? Both. External drivers include cautious discretionary spending, financial-services modernization, currency, AI-driven price pressure, and hyperscaler infrastructure investment. Internal drivers include improved large-deal execution, the labor pyramid, Project LEAP, acquisitions, and repurchases.

How stable are revenues? [FACT/INTERPRETATION] Stable but not contractual like subscription software. Embedded run-work, regulated workflows, long transitions, and account knowledge make revenue behaviorally recurring. New discretionary projects, product resale, and time-and-materials hours are more volatile. Revenue did not collapse during the 2023 downturn, which demonstrates resilience.

Outlook for products and services? FY2026 constant-currency growth is guided to 4.0%-5.5%, including about 150 basis points inorganic. Organic midpoint is therefore roughly 3.25%. Financial Services is strong; discretionary work and the other segments are weak. Adjusted margin guidance is 16.0%-16.2%, and adjusted EPS guidance $5.70-$5.82.

How big will the market be; growing or shrinking; domestic or international? [INTERPRETATION] Global and growing in workloads, but potentially deflationary in revenue per workload. Cognizant serves primarily North American demand using international delivery, especially India. Management’s multi-trillion-dollar enterprise-operations framing is a hypothesis; the investable fact is that modernization, data, security, governance, and integration demand is large while price capture remains uncertain.

Business Quality & Competitive Moat

Is the industry getting more or less competitive? More competitive. GCC insourcing, AI-enabled capacity, hyperscaler influence, and aggressive vendor consolidation all increase client alternatives.

How profitable is the business? [FACT] TTM operating margin is 16.3%, net margin 10.3%, and ROIC.ai reports ROIC around 13.2%. TTM FCF is $2.598 billion on $21.642 billion revenue, a 12.0% margin. Returns on tangible capital are much higher because goodwill, not physical capital, dominates investment.

How profitable is the industry; how many competitors; what barriers? Several global providers earn attractive accounting returns because capex is low. Barriers at the industry level are modest: talent, cloud, models, and software are broadly available. The barriers live within accounts and regulated workflows, not across the market.

Can the business be easily understood? Yes. Cognizant sells technology and operational outcomes produced by people, tools, and partner platforms. Profit depends on price, scope, utilization, wage geography, labor pyramid, product mix, and productivity retained.

Can it be undermined by foreign low-cost labor? It is already an offshore model. The new threat is lower-cost AI capacity and clients’ own lower-cost global centers. Cognizant must move from labor arbitrage toward workflow, platform, and outcome economics.

Do brands matter? Moderately. Enterprise buyers value references, security, domain knowledge, and low transition risk. Accenture has a stronger C-suite brand; top Indian peers have strong delivery brands. Cognizant’s brand reduces perceived risk but does not eliminate price competition.

What is the nature of competition? Competitive tenders for transformations, renewals, and vendor consolidations; account-level incumbency for run-work; partner certifications and domain references for new work. Price, talent, transition capability, and industry context matter.

Customers’ switching costs? [INTERPRETATION] Meaningful for embedded systems and TriZetto, limited for incremental projects. Transitioning a critical service takes time and operational risk, but clients commonly multi-source and rebid work.

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? The trained workforce, account knowledge, delivery methods, partner certifications, internally developed accelerators, and organic client relationships. TriZetto’s economic value may exceed its carrying value.

Off-balance-sheet liabilities? No material hidden liability identified. Leases are capitalized. Routine tax, visa, labor, litigation, and indemnity exposures are disclosed. A legacy derivative action has preliminary settlement approval but is not thesis-driving.

How conservative is the accounting? [INTERPRETATION] Generally clean. Restructuring and the India liability reversal are separately disclosed; stock compensation is modest; FCF is independently calculable. Cautions are acquisition goodwill, changed segment-cost allocation, changed associate-productivity definitions, and revenue mix from low-margin product resale.

How CapEx-hungry is the business? Very light. TTM capex was $319 million, about 1.5% of revenue. The economic investment is hiring, training, sales, partner capability, and acquisition goodwill rather than fixed assets.

Balance-sheet condition? [FACT] Cash plus short-term investments was $1.051 billion; debt including leases approximately $2.094 billion; net debt approximately $1.043 billion. Current ratio exceeds 2x and net debt is only about 0.26x TTM EBITDA. Liquidity is strong, but the former net-cash cushion is gone.

Capital Allocation & Management

How much FCF does the business generate, how is it used, and what is the philosophy? TTM FCF was $2.598 billion. Management uses cash for a covered dividend, capability M&A, and buybacks. First-half uses—$1.607 billion repurchases, $1.334 billion acquisitions, and $316 million dividends—exceeded first-half FCF, so debt and cash funded the difference.

Significant acquisitions recently? Belcan, Thirdera, 3Cloud, and Astreya. 3Cloud and Astreya cost about $1.39 billion combined; approximately $1.00 billion became goodwill and $380 million customer relationships. Strategic fit is coherent; returns are unproven.

Buying back shares? Yes. Q2 repurchases were 22.5 million shares for $1.153 billion. The ASR average was $51.54, and shares outstanding fell to 452 million from 479 million at year-end. The economics were attractive at the purchase price.

Issuing large amounts of new shares to insiders? No. Stock compensation is modest and more than offset by repurchases. Share count is shrinking rapidly.

Compensation policy? Revenue, adjusted operating margin, adjusted EPS, strategic objectives, and shareholder-return metrics drive incentives. These are sensible, though adjusted EPS can benefit from financial engineering. Organic growth, FCF per share, and post-goodwill ROIC deserve more weight.

Management motivations? [INTERPRETATION] Ravi Kumar and Jatin Dalal are pursuing a visible turnaround: large deals, AI capability, acquisitions, pyramid redesign, and aggressive capital return. The strategy appears designed to prove growth and close the peer discount. The risk is overusing M&A and repurchases before organic economics are demonstrated.

Valuation & Market Data

Is the stock an ADR, MLP, or K-1 issuer? No. CTSH is Class A common stock of a Delaware US C-corporation listed on Nasdaq.

Dividend policy? $0.33 quarterly, $1.32 annualized, roughly 2.2% current yield. The payout is well covered and secondary to repurchases.

How profitable is the business? Approximately 16% operating margin, 12% FCF margin, and low-to-mid-teens reported ROIC. Good economics for a services company, below the best offshore peers.

Is net income diverging from cash from operations? TTM cash conversion is healthy: $2.917 billion CFO versus $2.220 billion net income. First-half 2026 conversion is weak: $832 million CFO versus $1.298 billion net income. DSO of 88 days and the required second-half catch-up are key monitors.

Current embedded expectations? At $60.92 the company trades around 10.6x midpoint adjusted EPS and 10.6x TTM FCF. The price assumes durable low growth and stable margins, not terminal decline and not AI leadership.

Risks & Downside

What factors would cause the stock to decline? Flat or negative organic service growth despite high bookings; currency-neutral margin compression; weaker Financial Services; DSO above 90 days; FCF conversion below 80%; acquisition underperformance; further debt-funded buybacks; loss of share to Accenture, Indian peers, or GCCs; or evidence that AI productivity is passed entirely to clients.

Risk of catastrophic loss? Low. The business is diversified, capital-light, liquid, and cash-generative. The realistic tail is a secularly declining services franchise with repeated restructuring and low valuation.

Chance of total loss? Negligible under normal conditions. A total loss would require an implausible combination of enterprise-services collapse, severe legal or cyber damage, failed liquidity, and inability to resize the workforce.

Recent News & Events

Has the business environment changed recently? Yes. Discretionary spending remains weak, AI productivity is entering renewal economics, infrastructure and data spending is stronger than application discretion, and product resale is a larger revenue component. CTSH moved from a June panic low to a rapid recovery after Q2.

Significant acquisitions? Astreya closed June 22; 3Cloud closed in January. Astreya contributed only about one week to Q2 and will add more inorganic growth in Q3.

Change in accounting policies? Segment-cost allocation changed, with 2025 recast for comparability. Associate productivity metrics now exclude trainees not fully deployed, with comparables recast. These changes require consistent definitions.

Recent changes—new markets, facilities, or management? Project LEAP is reshaping the organization; Cognizant is hiring graduates while reducing other roles. It is expanding partnerships across Anthropic, OpenAI, Google, Microsoft, ServiceNow, Snowflake, CrowdStrike, and others. It is evaluating an India listing but has made no decision. No thesis-changing management departure occurred.

APPENDIX B — Source Appendix

Cognizant Technology Solutions Corporation (NASDAQ: CTSH) · 2026-08-21

Primary sources precede secondary sources. Current valuation was recomputed from the August 20 price and June 30 shares, cash, and debt; stale period-end valuation fields were not reused.

A. Primary SEC filings — CIK 0001058290

B. Company earnings and public materials

C. Current market and quantitative sources

D. External press and industry context

E. Analytical frameworks

• Bruce Greenwald and Judd Kahn, Competition Demystified: supply, demand, scale-and-captivity moat tests; share stability; barriers to entry; earnings-power discipline.
• Edward Chancellor, ed., Capital Returns: supply-side capital cycle; effective capacity; asset growth and M&A; management response to changing industry economics.

F. Method notes

• “Underlying service growth” is an analytical normalization: 4.1% Q2 constant-currency growth less approximately 100 basis points from acquisitions and 170 basis points from product resale equals approximately 1.4%. It is not a company-reported KPI.
• Current enterprise value uses $60.92 multiplied by 452 million shares, plus $2.094 billion debt including leases, less $1.051 billion cash and short-term investments.
• Free cash flow equals cash from operations less capital expenditure. First-half simple FCF of $657 million differs slightly from management’s $652 million adjusted definition.
• Currency-normalized margin direction is an interpretation from management’s approximately 180-basis-point operating benefit before approximately 80 basis points of hedge offset.
• Indian ADR peer multiples were not recomputed from the ROIC.ai feed because statement currency and ADR market capitalization were inconsistent. Peer conclusions rely on correctly matched public data.