Cloud Demand Is Holding; the Consumer Signal Is Selective

The proposed durable-demand thesis is not resolving as one trade. The cleaner evidence is in enterprise software consumption tied to AI and cloud adoption; the consumer side still depends on housing turnover, financing conditions, promotions and confidence.

The opening snapshot: a rate-sensitive market after record highs

U.S. equities pulled back modestly on October 7 after the S&P 500 and Nasdaq reached record closing highs the prior session. Reuters reported that the S&P 500 fell 0.23%, the Nasdaq Composite 0.24% and the Dow 0.69% as long-dated Treasury yields resumed their climb. Housing and homebuilders were among the weaker groups as mortgage rates moved higher.^[1]

The selected basket was mixed at the 16:00 ET close: DDOG fell 2.48% to $271.34, SNOW fell 0.92% to $332.85, RH fell 1.68% to $115.19 and WSM fell 0.74% to $240.46. LZB was almost unchanged at $29.64, while LESL’s latest available close was $0.102 and TPX’s quote record is not current. ETH was returned as a $24.50 equity quote by the feed, so it is not treated here as a confirmed cryptocurrency spot instrument. These observations are from the FMP snapshot; the regular close is the relevant tape because the U.S. session was closed.^[2]

That is a useful starting point, not a breadth claim: software and home-related names did not move as one block, and two symbols have coverage limitations.

Software provides the strongest evidence for the thesis

Datadog’s latest available Q2 FY2026 call offered operating evidence beyond an AI narrative. Management said revenue reached $1.12 billion, up 36% year over year, with roughly 33,400 customers and about 4,720 customers above $100,000 of ARR. It also described high-20s growth among non-AI customers and more than 750 AI customers using Datadog.^[3]

The important transmission mechanism is usage. Management said AI activity is accelerating cloud and modern-technology usage, while agentic activity and MCP tool calls were growing rapidly. That does not prove the next year’s result, but it is the kind of customer and workload evidence that can support durable revenue if it persists.^[3]

Snowflake shows a similar pattern with a different risk profile. In its Q2 FY2027 call, management reported 37% year-over-year product-revenue growth for a third consecutive quarter, a 15% non-GAAP operating margin, and a higher full-year product-revenue outlook of 36% growth. Management linked AI adoption to new workloads and greater platform consumption.^[4]

The counterpoint is embedded in Snowflake’s consumption model. Earlier management discussion acknowledged the possibility of sticker shock and optimization as AI agents create more users, applications and workloads. Growth can therefore be real while the path to efficient monetization remains under observation.^[4]

![Enterprise cloud infrastructure is translating AI adoption into observable workload demand.]

Home and discretionary demand require more conditions to align

RH’s second-quarter 2026 filing reported revenue growth of 2.6% to $922.2 million, but its adjusted EBITDA included a $55.1 million tariff benefit; normalized adjusted EBITDA margin was 13.4%. That distinction matters because headline profitability can conceal how much the result depends on temporary or non-operating conditions.^[5]

WSM is a stronger recent operating datapoint than the broader category: its Q2 comparable-brand revenue increased 6.2%, and the company raised its full-year outlook. But one good quarter does not establish a uniform furniture recovery, especially while mortgage rates and consumer confidence remain restrictive.^[5]

![Housing-sensitive demand still depends on households converting intent into financed furniture purchases.]

For LZB, LESL and TPX, the current pass is more about what must be verified than about declaring a direction. LZB was nearly flat in the latest close; LESL’s feed is stale and showed a large extended-hours discrepancy against an old close; TPX’s observation is dated February 2025. Data quality is part of the investment question here: an unavailable observation is neither bullish nor bearish.

Macro: growth is resilient, but the financing channel is restrictive

The September macro snapshot showed 4.2% unemployment, 2.1% year-over-year real GDP growth and a 3.75% federal-funds rate. Against that, CPI inflation was 3.35%, the 10-year Treasury yield was 5.31%, high-yield credit spreads were 3.24%, and consumer sentiment was 51.7—down 11.17% year over year.^[6]

This is a two-speed backdrop. It can support software businesses whose customers are funding mission-critical infrastructure while still discouraging large discretionary purchases. A lower long-term yield would remove one constraint, but the broader consumer thesis also needs improving purchase activity and margins—not just a better valuation multiple.

What would confirm—or weaken—the hypothesis?

What to watch next

• DDOG and SNOW earnings evidence. DDOG is scheduled for November 5, 2026, before the open, and SNOW for December 2, 2026, after the close. Both dates are marked estimated by the calendar source. Watch usage, large-account additions, retention and margin mix rather than headline AI language.^[7]
• The consumer sequence. WSM is scheduled for November 18 before the open, RH for December 10 after the close, LZB for November 17 after the close and LESL for December 1 after the close; each date is estimated. TPX has no confirmed date in the current calendar.^[7]
• Rates and confidence together. The thesis needs more than a lower yield: it needs evidence that households are willing and able to transact at acceptable margins.
• Instrument and data validation. Confirm what ETH represents in the quote universe and refresh TPX and LESL before using them as comparative evidence.

Bottom line

The hypothesis is partly supported, not proven. DDOG and SNOW show measurable earnings growth and demand connected to AI and cloud adoption. The consumer names face a higher evidentiary bar because revenue depends on housing, financing, promotion and confidence aligning at the same time. The most defensible conclusion is therefore selective: durable demand is easier to observe in mission-critical software consumption than in discretionary home purchases.