AI Demand Meets a Selective Consumer Cycle

The opening snapshot is not a single macro trade. It is a test of whether earnings growth and resilient demand can support a mixed group of software, AI infrastructure, auto retail and home-furnishings companies over the next year—or whether the market is beginning to separate durable consumption from expensive promises.

The tape is sending two messages

The broad Wednesday close was constructive: SPY finished at 772.31, up 0.34%, QQQ at 725.61, up 0.60%, and DIA at 524.68, up 0.27%, all as of the 16:00 ET regular close on October 28.^[1] But the pre-market signal on October 29 is more selective. SNOW was at 325.76, down 4.16% versus its 16:00 ET close, while AN was at 160.80, down 2.06%; RH was at 147.18, up 1.46%, and WSM at 252.00, up 0.76%, with those prints timestamped between 07:07 and 07:20 ET.^[1]

That dispersion matters more than the index direction. Reuters reported that software shares fell in pre-market trading after Fortinet and ServiceNow results revived concerns about AI disruption and demanding expectations.^[2] The immediate lesson is not that enterprise software demand has disappeared. It is that “good growth” now has to survive a tougher test around pricing, product durability, margins and the economics of AI substitution.

What the six-month record says—and does not say

The last six months through October 28 show a wide gap inside the research scope:

These figures are from one comparison window, not a forecast. DDOG and SNOW outperformed the group, while CRWV and AN lagged materially; the spread is evidence that the market is pricing business-model and balance-sheet differences, not simply buying “growth.”^[3]

The operating hypothesis has a real supporting case

The macro backdrop is not recessionary in the latest snapshot: unemployment was 4.2%, real GDP growth was 2.1% year over year, and industrial production was up 2.25% year over year as of September 2026.^[4] That is compatible with ongoing enterprise technology budgets and with selective high-income consumption.

There is also evidence that the consumer side is not uniformly weak. WSM’s latest reported quarter was described as showing revenue growth of 6.7% and a raised outlook, while RH’s investor materials reported second-quarter revenue of $922.2 million, up 2.6%, alongside an 18.3% EBITDA margin.^[5] Those are company-specific results, not proof of a broad housing recovery, but they support the narrower interpretation that brand, assortment and execution can offset a difficult backdrop.

For the software names, the hypothesis is more conditional. DDOG’s next earnings date is listed as November 4, estimated, before the open; SNOW’s is December 2, estimated, after the close.^[6] The market will be looking for evidence that consumption, platform expansion and AI-related workloads are producing durable revenue—not merely a short-lived spending wave.

The countercase is becoming harder to ignore

The cost of capital is still meaningful. The 10-year Treasury stood at 4.99%, the fed-funds rate at 3.75%, and high-yield credit spreads at 3.18% in the latest macro snapshot. Consumer sentiment was just 48.1, down 12.7% year over year.^[4] That combination can coexist with solid GDP while still pressuring discretionary purchases and making highly capital-intensive growth less forgiving.

CRWV is the clearest stress test. Its latest regular-session close was 63.93, up 7.32% on October 28, but its six-month return was still -49.03% with a -60.69% maximum drawdown.^[1][3] A strong day does not settle the debate over infrastructure demand; it shows how quickly sentiment can move when expectations, financing and capacity utilization are all being repriced.

The same distinction applies to RH, WSM, AN, ETH and LZB. A resilient demand thesis for these businesses would need to show up in traffic, conversion, unit volumes, pricing and margins—not just in a temporary share-price bounce. For ETH and LZB specifically, the available earnings calendar does not provide a confirmed date for ETH, while LZB’s November 17 report is listed as estimated and after the close.^[6]

What would have to be true for each side to be right?

For the constructive case: enterprise AI and observability workloads continue to broaden; DDOG and SNOW convert usage into sustained revenue and margin; CRWV funds capacity without overwhelming shareholders or cash flow; and WSM, RH, AN, ETH and LZB protect demand through differentiated brands, inventory discipline or market-share gains.

For the cautious case: software buyers consolidate vendors faster than usage expands; AI substitutes for portions of incumbent software spend; data-center capacity outruns monetization; and higher yields plus weak sentiment delay big-ticket household purchases. Under that outcome, headline revenue growth could remain positive while the market compresses the multiple and penalizes margin slippage.

What to watch next

• October 30: AN’s earnings report is listed as estimated and before the open.^[6]
• November 4: DDOG’s report is listed as estimated and before the open.^[6]
• November 17: LZB’s report is listed as estimated and after the close.^[6]
• November 18 and December 2: WSM and SNOW are listed as estimated report dates, with WSM before the open and SNOW after the close.^[6]
• Across every report: separate demand growth from price increases, backlog from recognized revenue, and AI enthusiasm from cash returns.

The balanced conclusion is narrower than “growth wins.” Current evidence supports a split thesis: demand is resilient enough to keep the opportunity alive, but the next year will likely reward proof of monetization and execution more than exposure to a theme alone. No single pre-market print resolves that question.