Coherent Beats Expectations as AI Data Center Optics Drive a Bigger 2027 Outlook

Coherent exceeded Wall Street’s expectations on August 12, reporting quarterly revenue of $2.05 billion and adjusted earnings of $1.74 per share. The optical technology supplier also issued a first-quarter outlook well above its latest quarterly run rate.

The results confirm that artificial intelligence spending has reached another layer of the data center supply chain. Demand is moving beyond processors and switches toward the lasers, transceivers, and optical components that connect expanding clusters.

Yet this is not simply an earnings beat. Coherent must now convert exceptional demand into manufacturing output without sacrificing margins or delivery performance. That challenge puts the company beside Lumentum in a capacity race that neither supplier can win through laboratory progress alone.

Coherent’s Earnings Beat Was Broad, Not Cosmetic

Coherent beat expectations across revenue, adjusted earnings, margins, and its forward outlook.

Coherent reported the results after the US market closed on August 12. The fiscal fourth quarter ended June 30, making the announcement a completed financial event rather than an unconfirmed hot-list claim.

Quarterly revenue reached $2.05 billion, up 34% from one year earlier. Coherent calculated pro forma growth at 42% after excluding businesses sold during fiscal 2026.

The distinction matters because reported growth includes changes in the company’s portfolio. Pro forma growth compares the continuing operations on a more consistent basis.

Datacenter and Communications generated $1.62 billion, up from $1.02 billion one year earlier on Coherent’s pro forma presentation. The segment represented 79% of quarterly revenue, compared with 21% from Industrial.

That mix shows where the acceleration came from. Industrial revenue declined to $431 million from $511 million, while the data center business added nearly $600 million.

Coherent’s quarterly results also showed meaningful profit expansion. GAAP gross margin rose to 38.5%, an increase of 277 basis points from the prior year.

Adjusted gross margin reached 40.2%, up 215 basis points. Adjusted operating margin climbed to 21.8%, compared with 18% in the corresponding quarter.

Adjusted earnings reached $1.74 per diluted share. That exceeded the $1.62 estimate listed before the announcement in the week’s earnings calendar.

GAAP earnings were $1.19 per diluted share, compared with a loss one year earlier. GAAP net income attributable to the company reached $241 million.

The adjusted figure removes several items, including share-based compensation, acquired-intangible amortization, restructuring costs, and asset impairments. Investors should therefore avoid treating adjusted earnings as interchangeable with GAAP income.

Still, the improvement was not created solely by adjustments. Coherent’s GAAP gross margin, operating income, and net income all improved substantially.

The company also posted sequential growth. Revenue rose from $1.81 billion in the March quarter to $2.05 billion in June, an increase of roughly 13%.

Datacenter and Communications revenue advanced from $1.36 billion to $1.62 billion during that period. Industrial revenue declined from $444 million to $431 million.

That divergence makes the central story unusually clear. Coherent’s overall expansion currently depends on optical demand from data centers and communications networks.

Full-year revenue reached approximately $7.13 billion, compared with about $5.82 billion in fiscal 2025. Adjusted earnings per share increased to $5.61 from $3.53.

Chief executive Jim Anderson said adjusted earnings grew at more than twice the rate of revenue. The financial statements support that direction, although adjusted measures remain management-defined calculations.

The quarter therefore answered the immediate question behind the hot-list headline. The underlying event occurred on August 12, 2026, and the company’s official release confirms a broad financial beat.

More important, the results created a harder question. Can Coherent supply what customers now appear ready to order?

AI Data Centers Have Turned Optics Into a Scaling Constraint

The strongest result is not the revenue total; it is the evidence that optical connectivity has become essential infrastructure for larger AI systems.

Graphics processors perform the calculations inside AI clusters. Those processors create value only when data can move between chips, racks, and buildings at sufficient speed.

Copper connections remain practical across short distances. However, electrical links consume more power and lose signal quality as bandwidth and distance increase.

Optical links convert electrical signals into light for transmission through fiber. That conversion adds components and manufacturing complexity, but it improves reach and bandwidth efficiency.

As clusters grow, networking performance increasingly determines how effectively companies can use expensive accelerators. A delayed processor wastes capacity, even if its own theoretical performance remains unchanged.

Coherent supplies several components used in this transition. Its portfolio includes lasers, photodiodes, optical transceivers, indium phosphide devices, and silicon photonics technology.

A transceiver converts data between electrical and optical formats. Indium phosphide, often shortened to InP, is a semiconductor material used in high-speed lasers and related optical devices.

Coherent said its 1.6-terabit transceivers remain an existing growth engine. It is also developing 3.2-terabit and 6.4-terabit products for future network generations.

The company’s investor presentation identifies additional opportunities in optical circuit switches, data center interconnect products, thermal systems, and co-packaged optics.

An optical circuit switch redirects light paths without repeatedly converting traffic back into electrical signals. It can help data centers reorganize network capacity more efficiently.

Co-packaged optics places optical interfaces closer to a switch or processor package. The approach aims to reduce the electrical distance traveled by high-speed signals.

These technologies address different parts of the same constraint. More computing capacity creates more traffic, while denser racks make power consumption and heat harder to control.

Coherent said Datacenter and Communications revenue grew 59% year over year on a pro forma basis. That pace far exceeded the 1% pro forma growth reported for Industrial.

The result does not prove every proposed optical architecture will reach mass adoption. It does show that current optical products are already benefiting from the data center buildout.

That difference is important. Investors have often valued optical suppliers according to products that remain several years from broad deployment.

This quarter relied on products shipping now, including transceivers and components for existing network designs. Future technologies add potential, but the current business no longer depends entirely on a distant transition.

NVIDIA’s behavior offers another demand signal. In March, NVIDIA announced a multiyear agreement involving advanced lasers, optical networking products, and future capacity rights.

The agreement was nonexclusive, meaning NVIDIA retained the ability to work with other suppliers. It included a multibillion-dollar purchase commitment and a separate investment in Coherent.

NVIDIA’s optics partnership expanded access across several Coherent product families. The companies tied the work to next-generation AI infrastructure and US manufacturing.

This arrangement does not guarantee specific future revenue. Purchase schedules, product qualifications, and customer deployment timing can still change.

However, it demonstrates that a leading system designer views optical supply as strategically important. NVIDIA did not limit its response to a conventional supplier contract.

The AI data center buildout has therefore changed the status of photonics. Optics is moving from a specialized networking category toward a constraint that major customers actively finance and secure.

The Real Contest Is Coherent Versus Lumentum on Capacity

Coherent’s primary opponent is Lumentum, but the decisive contest concerns manufacturing capacity rather than a single product benchmark.

Lumentum and Coherent both supply high-performance lasers and optical components. Both participate in markets shaped by cloud networking, telecommunications, and AI infrastructure.

Their overlap makes product comparisons tempting. Yet customers need qualified components in volume, not isolated demonstrations with favorable laboratory measurements.

NVIDIA reinforced this point by announcing strategic arrangements with both suppliers. It committed capital and purchase volume without selecting one exclusive winner.

That choice changes how the competition should be understood. Coherent does not need to eliminate Lumentum to benefit from the transition toward optical networking.

Instead, each supplier must expand enough qualified capacity to capture demand. Execution determines how much of the available market each company can serve.

Coherent’s advantage is breadth. It operates across materials, devices, components, modules, and systems, giving it exposure to several layers of the optical chain.

Vertical integration can improve coordination when one component becomes scarce. It can also allow design work across lasers, detectors, packaging, and complete modules.

That breadth creates complexity, however. Managing several manufacturing steps requires capital, technical specialists, and consistent yields across different facilities.

Lumentum presents credible competitive pressure because it also has deep laser expertise and established data center relationships. Its own fiscal fourth-quarter performance showed that optical demand was not unique to Coherent.

This parallel strength supports the industry thesis, but it limits claims that one supplier has captured the market. Customers appear willing to fund more than one manufacturing route.

Coherent’s reported capacity plans show the scale of the response. Management said it remains on track to double internal indium phosphide output by the end of 2026.

The company plans to more than double that output again during 2027. It is also moving production toward six-inch wafers for electro-absorption modulated lasers, continuous-wave lasers, and photodiodes.

An electro-absorption modulated laser combines a light source with a component that rapidly encodes data onto that light. The device is widely relevant to high-speed optical links.

Larger wafers can produce more devices per manufacturing cycle. They can reduce unit costs when yields remain strong, but a larger format does not automatically guarantee reliable output.

Coherent says its six-inch platform currently produces higher yields than its three-inch lines. That statement comes from management and still requires validation through sustained volume production.

The manufacturing footprint matters for another reason. NVIDIA is supporting Coherent’s expansion in Sherman, Texas, where the company is modernizing an existing facility.

Coherent has also signed a letter of intent for up to $50 million in proposed CHIPS Act funding. That remains preliminary rather than a completed award.

The company previously received support through Texas programs and local economic development arrangements. Those resources reduce some expansion pressure but do not remove execution risk.

NVIDIA’s investment has a clearer legal record. A Coherent SEC filing says NVIDIA acquired 7,788,161 shares through a private placement.

The filing also says the companies expanded collaboration across five additional product families related to co-packaged optics. NVIDIA received future access and capacity rights under the broader arrangement.

Strategic customer funding helps Coherent build earlier than normal purchase patterns might allow. It also creates closer alignment with a major architecture provider.

Yet dependence can move in both directions. Large customers gain influence when they provide capital, purchase commitments, and product requirements.

The most useful comparison is therefore not which company announced the fastest component. It is which supplier converts customer-backed expansion into stable yields and dependable deliveries.

Coherent’s earnings provide an early advantage because revenue and margins rose together. The capacity race will determine whether that advantage persists.

Coherent’s Guidance Raises the Execution Bar

The outlook implies another step up in demand, leaving Coherent less room for production delays or uneven yields.

For the first quarter of fiscal 2027, Coherent forecast revenue between $2.2 billion and $2.4 billion. The midpoint represents approximately 12% sequential growth from the June quarter.

The company expects adjusted gross margin between 39.5% and 41.5%. It forecast adjusted operating expenses between $400 million and $420 million.

Adjusted earnings guidance ranges from $1.85 to $2.05 per diluted share. Even the low end exceeds the $1.74 reported for the fourth quarter.

Management did not provide corresponding GAAP guidance for several measures. Coherent said it could not reasonably forecast items needed for those reconciliations.

Those items include restructuring expenses, integration costs, foreign exchange effects, and share-based compensation. Their variability can materially change GAAP results.

The guidance therefore communicates strong operating expectations, but it does not offer a complete view of future statutory profit. Readers should preserve that distinction.

Still, the revenue outlook carries substantial information. Its low end stands $154 million above the June-quarter result, while the high end implies a much larger sequential increase.

Coherent expects new products to contribute at different times. Its roadmap places early thermal-solutions revenue in the second half of calendar 2026.

Management expects additional growth platforms during 2027, including co-packaged optics and near-packaged optics. Near-packaged optics positions optical modules close to switching silicon without fully integrating them into the package.

That schedule means the near-term forecast still relies heavily on current products and ongoing capacity additions. Future architectures cannot explain every dollar of the first-quarter increase.

The quarter also improved Coherent’s ability to finance expansion internally. Adjusted operating income reached $446 million, up from $275 million one year earlier.

GAAP operating income was $254 million, compared with $6 million in the prior-year quarter. The gap between the two measures remained significant.

Coherent excluded $54 million in share-based compensation and $70 million in intangible amortization from adjusted operating income. It also excluded restructuring, impairment, and integration-related items.

Those exclusions do not invalidate the operating improvement. They do show why margin comparisons should use consistent accounting definitions.

The company’s broader financial structure also deserves attention. Coherent was formed through a large acquisition strategy, including the combination of II-VI and the former Coherent business.

That history left the company with debt and a complicated operating portfolio. Management has since sold businesses, reduced organizational overlap, and concentrated investment around stronger markets.

Portfolio changes can improve margins by removing lower-return operations. They can also make historical revenue comparisons less intuitive.

That is why the company presents pro forma growth alongside reported growth. Readers should track both rather than choosing whichever produces the larger percentage.

The central reversal is now visible. Coherent previously had to prove that its broad portfolio could generate acceptable returns after years of integration and balance-sheet pressure.

Today, demand is strong enough that insufficient capacity presents a greater near-term risk than insufficient product breadth. The operating challenge has changed from finding growth to fulfilling it.

That shift helps explain why the market focused on guidance. A strong quarter confirms what happened; a much higher forecast indicates that management expects the demand surge to continue.

What the Numbers Still Do Not Prove

The report validates demand, but it does not establish that current growth, customer concentration, or adjusted margins will remain stable.

The first uncertainty concerns manufacturing yield. Doubling nominal wafer capacity does not double sellable output when qualification delays or defect rates intervene.

New production lines usually pass through installation, process tuning, customer qualification, and volume ramp stages. Problems at any stage can delay revenue.

Coherent says its six-inch indium phosphide platform already delivers higher yields than older three-inch lines. Investors still need several quarters of evidence at higher production volumes.

The second uncertainty concerns product timing. Co-packaged optics promises lower power consumption and higher bandwidth density, but deployment requires changes across switches, packaging, cooling, and service models.

Large customers can test multiple architectures before committing to a standard. Some networks may continue using pluggable transceivers longer than suppliers expect.

That outcome would not eliminate Coherent’s opportunity. The company sells products for both current pluggable designs and emerging integrated architectures.

However, it would change the timing and margin profile of newer growth platforms. A delayed architecture transition can strand specialized investment or postpone expected returns.

The third uncertainty is customer concentration. Coherent does not disclose every buyer’s quarterly contribution in its earnings presentation.

NVIDIA’s investment and purchase commitment make the relationship strategically valuable. They also highlight how strongly supplier plans can depend on a small number of hyperscale and system customers.

A large customer can alter deployment schedules for technical, financial, or regulatory reasons. That decision can move demand across quarters even when the long-term market remains intact.

The fourth uncertainty involves competition. NVIDIA’s parallel commitment to Lumentum shows that customers do not intend to rely on one optical supplier.

Dual sourcing reduces supply risk for buyers. It also limits the pricing leverage available to any single vendor once capacity catches up with demand.

Additional competition can come from component specialists, module manufacturers, and semiconductor companies developing silicon photonics. Different architectures distribute value across the supply chain in different ways.

Coherent’s broad portfolio offers several ways to participate. It also requires the company to decide where additional capital can produce the best return.

The fifth uncertainty concerns adjusted accounting. Coherent reported clear GAAP improvement, but its adjusted profit remained substantially higher than its GAAP result.

Share-based compensation, amortization, restructuring, and integration expenses are real economic considerations. Some are noncash, while others reflect continuing organizational change.

Investors should therefore track whether the gap narrows as the company completes portfolio work. Persistent exclusions would weaken the argument that reported operational progress fully reaches shareholders.

The company itself lists several relevant risks. Its presentation warns about demand forecasting, customer purchasing patterns, new-product acceptance, competing launches, trade restrictions, and returns from capacity investments.

Those warnings are standard legal disclosures, but they map directly to the current situation. Exceptional demand encourages aggressive expansion, which increases the cost of forecasting errors.

There is also a cyclical precedent. Optical markets have previously experienced periods when suppliers expanded for projected bandwidth demand, only to encounter inventory corrections.

The current AI cycle differs because hyperscalers are funding physical infrastructure at enormous scale. It does not abolish inventory cycles or customer schedule changes.

Coherent’s Industrial segment provides another useful check. Its revenue declined during a quarter when the company’s consolidated result appeared exceptionally strong.

That weakness shows how one fast-growing segment can dominate the headline. It also means the company has not achieved equally strong conditions across its entire portfolio.

None of these issues overturns the quarter. They define the evidence needed to distinguish durable operating improvement from a demand spike amplified by constrained supply.

Three Signals Will Decide Whether the Coherent Outlook Holds

The next phase will be judged by capacity output, margin retention, and the timing of new optical platforms.

The first signal is indium phosphide output. Coherent says it will double internal production by the end of 2026 and more than double it again during 2027.

Investors should look for evidence that sellable device volume rises with installed capacity. Management commentary on yields, qualifications, lead times, and supply constraints will matter more than factory announcements alone.

A clean ramp would strengthen the argument that Coherent can convert demand into revenue. Delays or continuing shortages would weaken the first-quarter outlook and limit later growth.

The second signal is adjusted gross margin. Coherent guided to a range of 39.5% to 41.5% after reporting 40.2%.

Holding near the midpoint while revenue expands would show that the company is managing startup costs and product mix effectively. A decline below the range would suggest that capacity expenses or pricing pressure arrived faster than expected.

GAAP margin deserves equal attention. Improvement in both measures would provide stronger evidence than adjusted expansion by itself.

The third signal is revenue from new platforms. Coherent placed thermal products, optical circuit switches, and packaging-related technologies on a staged roadmap through 2027.

Readers should watch for named customer qualifications or disclosed commercial shipments, not demonstrations alone. Early revenue would support management’s claim that growth is broadening beyond current transceiver demand.

A schedule slip would not necessarily damage the existing business. It would reduce confidence in the incremental market opportunity included in longer-term expectations.

Lumentum’s response will remain an important reference throughout this period. Its capacity growth, customer wins, and margins can reveal whether demand remains industry-wide or shifts between suppliers.

NVIDIA’s deployment schedule offers another reference. Products using silicon photonics and co-packaged optics must progress from announced architectures into shipping systems.

The strongest version of Coherent’s thesis requires all three signals to align. Capacity must increase, margins must hold, and new platforms must become commercial products.

The weakest version involves strong orders without timely output. That scenario creates spending and customer pressure without producing the expected revenue.

For developers and enterprise AI buyers, the consequences extend beyond one supplier’s stock. Optical availability affects when larger clusters arrive, how efficiently they operate, and what cloud providers charge for scarce capacity.

For researchers and knowledge workers tracking AI infrastructure, the quarter is a reminder to follow the components surrounding accelerators. Compute roadmaps depend on networks, power, cooling, memory, and manufacturing throughput.

Coherent has now supplied the strongest evidence in years that its optical portfolio sits inside that critical path. Its August results validate the demand shift and set a much higher near-term target.

The next question is measurable: can Coherent turn planned capacity into qualified products while preserving margins? Watch its next earnings report for output, yield, and shipment evidence before treating the new run rate as permanent.