Software Demand Holds While Geopolitical Costs Reach Home Retail

The market tell

The clearest signal in this week’s geopolitical tape is a divergence: software demand is accelerating, while goods businesses remain exposed to the cost of moving and sourcing physical products. Datadog reported Q2 FY2026 revenue growth of 36% year over year, with growth outside its AI-native cohort in the high 20s and enterprise new-logo bookings more than doubling year over year.^[1] At the same time, Williams-Sonoma said tariffs and higher fuel costs were still reducing merchandise margins, even as positive furniture and non-furniture comps supported its results.^[2]

That is a more useful read than a blanket “risk-off” label. The current shock is discriminating between businesses whose revenue is primarily digital and those whose economics run through factories, ports, fuel and household purchasing power.

The geopolitical transmission channel

Reuters reported that China’s refiners suspended or reduced October fuel exports to build domestic stocks, a move that could tighten supplies for parts of Asia and alter regional product flows.^[3] Separately, Reuters reported that Qatari LNG cargoes were still transiting the Strait of Hormuz despite continuing risks.^[3] Those developments do not establish a single inevitable energy-price outcome, but they show why shipping lanes, refined products and insurance risk remain relevant to retailers with long supply chains.

The channel into company results is straightforward: a disruption can raise bunker fuel and inland transport costs, delay inventory, complicate country-of-origin decisions and force price or assortment changes. It can also create uneven winners when a larger operator has more scale, alternative suppliers or negotiating leverage.

What the company evidence says

Digital demand is holding up—but usage is not risk-free

Datadog’s latest available earnings-call record points to broad-based demand rather than an AI-only story. Management described robust growth across customer sizes and industries, while its AI customer group continued to expand.^[1] Earlier commentary also cautioned that customers can optimize cloud and observability usage and that enterprise adoption depends partly on the bandwidth to move workloads into production.^[1] The implication is balanced: demand can remain resilient, but consumption-based models still carry sensitivity to customer behavior.

Snowflake is included in this scope, but this pass did not retrieve a comparable current transcript excerpt. The stock snapshot showed SNOW at $368.89 at the 16:00 ET close on October 9, with an extended-hours price of $368.90 at 19:59 ET.^[4] That price action is consistent with the software side of the divergence, but it is not by itself evidence that geopolitical risk has disappeared.

Home furnishings are gaining demand share while absorbing policy risk

Williams-Sonoma’s Q2 FY2026 transcript said management was not building a housing recovery into guidance. It also said the company expected tariff effects to moderate in the back half as it began comparing against the prior year’s tariff costs, while higher oil prices continued to pressure transportation and supplier costs.^[2] The company raised its comparable-brand revenue outlook to 4%–6.5% and operating-margin outlook to 17.8%–18.2%, but also disclosed a $200 million tariff refund and related payments to vendors and employees.^[2]

RH offers a different version of the same story. Its recent call described a $75 million backlog of backorders and special orders above normal levels, while earlier management commentary stressed that resourcing furniture, lighting and rugs takes longer than moving smaller accessories between countries.^[5] That supports a two-part interpretation: resilient demand can help, but supply-chain adaptation is slower and more capital-intensive for bulky, specialized goods.

The market snapshot on October 9 reflected that mixed picture: RH closed at $116.99, up 2.9% on the session; WSM closed at $241.78, up 1.2%; and LZB closed at $29.22, down 1.1%.^[4] These are end-of-day observations, not proof of a single catalyst.

Testing the broader thesis

The hypothesis for DDOG, SNOW, RH, WSM, ETH, LZB, LESL and TPX is that earnings growth and resilient demand can support the group over the next year. This research pass finds partial support, but not a uniform setup.

• Most supportive evidence: DDOG’s acceleration across customer groups and WSM’s positive category comps despite a weak housing backdrop and tariff pressure.^[1][2]
• The constraint: energy and shipping disruptions can reach margins even when demand is intact. WSM explicitly cited higher fuel and tariff costs; Reuters’ fuel-export and Hormuz reporting shows why that risk remains live.^[2][3]
• The uncertainty: the available quote snapshot is uneven. ETH and the home-furnishing names have recent regular-session observations, but LESL’s latest quote was dated October 5 and TPX’s was dated February 26, so those two symbols should not be treated as current market confirmation.^[4]

The base case is therefore conditional rather than broad: digital infrastructure demand can stay firm while physical-goods companies outperform only if they convert demand into margin after tariffs, fuel and freight. A softer energy or trade backdrop would improve that conversion; another supply shock would test it.

What to watch next

1. Refined-product and LNG flows: further changes in Chinese fuel exports or interruptions around Hormuz would be an early read on whether logistics risk is intensifying.^[3]
2. Retail margin language: watch whether WSM and peers describe tariff and fuel pressure as moderating, passing through to prices, or being absorbed by vendors and operators.^[2]
3. Housing and high-ticket conversion: RH’s backorders, special orders and delivery timing can distinguish genuine demand from orders delayed by sourcing constraints.^[5]
4. Enterprise consumption: for DDOG and SNOW, the important question is whether AI-related workloads broaden into durable enterprise usage without a corresponding wave of optimization. DDOG’s own commentary identifies that tension.^[1]
5. Data quality across the smaller names: refreshed quotes and company-specific earnings evidence are needed before drawing conclusions about ETH, LESL or TPX from this basket.

This is market research, not investment advice. The evidence supports monitoring the divergence between digital demand and physical supply-chain exposure; it does not establish a guaranteed direction for any security.