Insights from E-commerce Transformation on the Transition to Software/AI

Investing.com - According to the latest industry research report from Stifel, the challenging period experienced by enterprise software stocks may not be a temporary decline but rather the beginning of a long-term restructuring.

Analysts compare the current situation to the e-commerce disruption of the late 1990s, believing that even if AI-driven concerns are exaggerated, it is prudent for investors to remain cautious.

The brokerage maps the software industry landscape to the retail archetypes of the e-commerce era: large existing companies striving to maintain their lead (Walmart-type), high-growth challengers positioning for dominance in the next cycle (Costco-type), survivors unlikely to thrive (Macy’s-type), and a key point—no expected cases of public bankruptcy (Bed Bath & Beyond-type).

Real-time tracking of major market movements and analyst opinions - Up to 55% off

Analysts state: “Similar to the trading patterns of ‘traditional’ retailers in the early 2000s,” they expect “many software stocks will not see any V-shaped recovery in the coming quarters.”

The example of Microsoft is cautionary. After reaching a peak of about $60 in December 1999, the stock fell below $40 in April 2000 and bottomed out around $20 in December 2000.

It only returned to $40 in April 2014, after 14 years, despite revenue growing from $22 billion to $83 billion at a nearly 10% compound annual growth rate, and earnings per share increasing at an average of about 8% annually.

Analysts point out that the stock’s recovery only occurred after new management took over and Azure began to accelerate.

They note that investors’ current concerns are not about tools from Anthropic or OpenAI immediately replacing Salesforce or ServiceNow installations worth hundreds of millions or billions of dollars.

The real concern is whether existing companies can monetize AI capabilities or will be forced to bundle AI agents into existing contracts to fend off competition, rather than using it as a new revenue driver.

In terms of profit margins, AI costs are expected to pressure the software industry’s historically best gross margins, similar to the transition from on-premises deployment to SaaS.

Stifel analysts believe that LLM providers may currently be subsidizing some customer usage, with certain prompt activities resulting in negative gross margins for providers.

As these subsidies fade and hyperscale cloud providers adjust infrastructure pricing to justify massive capital expenditures, a significant reset of profit margins may occur.

Valuations reflect this uncertainty. The enterprise value/future 12-month revenue ratio of the iShares Expanded Tech-Software ETF (IGV) has compressed from a peak of over 16x to 3.9x, although Stifel analysts note that, from a 20-year perspective, the sector is reverting to the range seen from 2005 to 2017.

For enterprise value/future 12-month free cash flow, IGV’s trading multiple is 22.8x, compared to a historical average of 38.2x, indicating it is cheaper but not in distress. Analysts add: “The broader sector may remain range-bound in the foreseeable future.”

Private equity is unlikely to play a rescue role during the SaaS transition, as PE’s share in institutional allocations has already risen above 20%, up from single digits two decades ago, coupled with higher debt costs and the difficulty of returning capital from existing holdings.

Strategic mergers and acquisitions are also expected to remain subdued, with IBM being an exception—continuing to acquire open-source infrastructure assets after acquiring Red Hat and HashiCorp.

In the short term, analysts favor stocks related to data, infrastructure consumption, and security, citing recent strong performances from Cloudflare and Datadog.

In terms of applications, they expect established SaaS companies with strong data attraction and deep domain expertise to be winners in ready-to-use agent workflows.

Top picks include CrowdStrike, Cloudflare, Palo Alto Networks, Salesforce, Guidewire Software, HubSpot, Braze, Titan Machinery, Datadog, MongoDB, and Snowflake.

This article was translated with the assistance of artificial intelligence. For more information, see our Terms of Use.