Brown Advisory: New Additions To  Mid-Cap Growth Portfolio

The article is bullish on three names: Guidewire Software is seen as having underappreciated competitive strength and durable growth, Toast is expected to keep growing locations by more than 15% for several years, and Reddit is benefiting from generative AI through lucrative data licensing agreements. The key takeaways are sustained growth visibility at Toast, improving conviction in Guidewire's positioning, and an AI monetization tailwind for Reddit.

Analysis

GWRE looks like the cleanest quality-duration compounder in the group: the market often underprices software businesses where retention is high, implementation is sticky, and pricing power shows up late in the cycle. The second-order implication is that incumbents and adjacent vertical SaaS vendors may face a tougher sell-side narrative on “AI disruption” because the real moat here is workflow embedding, not feature parity. If the growth durability thesis holds, multiple expansion can persist even without acceleration, which is usually where shorts get squeezed.

TOST’s >15% location growth thesis matters less as a pure unit story than as evidence that the small- and mid-sized restaurant base is still underpenetrated and operationally fragmented. That creates a longer runway for cross-sell in payments, payroll, and capital, while also making the business more resilient than a simple “restaurant beta” trade. The competitive risk is not another point solution, but a larger platform using bundled pricing to slow new logo wins; that pressure would likely show up first in new merchant quality before it appears in topline.

RDDT is the most interesting from a second-order standpoint because data licensing could become the market’s preferred “AI monetize the corpus” template, but the biggest risk is that investors extrapolate scarce, recurring AI revenue too far ahead of actual renewal cadence. If licensing becomes viewed as quasi-utility cash flow, the stock can rerate quickly; if deal concentration or model substitution risk emerges, the multiple can compress just as fast. The key debate is whether the market is pricing a durable platform tax on LLM training or a one-time monetization window.

The consensus may still be too focused on headline growth rates rather than the durability and optionality embedded in each model. The risk/reward is best where growth is steady enough to support valuation but not yet fully “known,” especially if AI exposure is additive rather than the core thesis. Near term, catalysts are less about quarterly beats and more about management commentary on retention, take rates, and expansion into adjacent products.

Key Decisions for Investors

• Long GWRE on pullbacks over the next 1-3 months; favor a quality-growth re-rating trade with limited fundamental downside if retention and conversion remain stable. Risk/reward improves if the market remains skeptical of vertical software durability.
• Long TOST vs. short a basket of restaurant-facing, low-moat software names for a 3-6 month pair trade; thesis is that location growth plus product expansion supports multiple expansion while peers face pricing pressure. Cover if net adds decelerate sharply or if churn inflects higher.
• For RDDT, buy upside through call spreads into the next 1-2 earnings cycles rather than outright stock; this captures rerating from AI licensing optionality while limiting downside if revenue concentration or contract visibility disappoints.
• Consider a relative-value long GWRE / short high-multiple unprofitable SaaS basket as a 6-12 month trade if rates stay elevated; the market should continue rewarding durable cash flow and punishing stories with weaker implementation moats.
• Set a tactical risk trigger on RDDT if management commentary shifts from repeatable licensing to one-off deals; that would argue for reducing exposure quickly because the stock is vulnerable to multiple compression.