← back to ranking

RJF · Raymond James Financial, Inc.

Financial - Capital Markets · mkt cap $28.6B · calls: Q2 FY2026 vs Q1 FY2026
47.0 conviction · conf-adj 47

conf 2/10 partial

enthusiasm:21.0 · trend:8 · quantifies:0 · impact:0 · under_radar:14 · credibility:0 · business_impact:4 · disruption:0 · commitment:0 · confirmation:0

Enthusiasm latest 7 / prev 6 (rising)

Raymond James frames AI as an adviser-enablement and internal-productivity layer—not a replacement for human advice—centered on the generative AI operations agent Ray/RA, broad associate/developer AI use, and future Agentic AI for process automation. Enthusiasm is substantive and rising (launch → pilot expansion → Agentic AI roadmap), but management repeatedly refuses to quantify revenue, margin, or cost savings from AI. Credibility is moderate: concrete deployments and adoption counts exist, yet business impact remains qualitative and explicitly "too preliminary" to model.

GROUNDED NEXT-FY IMPACT vs CONSENSUS

Grounded on actual base — revenue $15.9B · net income $2.1B · net margin 13.4% · diluted EPS 10.3

These are next-fiscal-year annual uplift estimates, not next-quarter numbers.

Aggregate next-FY est. rev uplift: % · next-FY EPS uplift: % · vs analysts: unclear · priced in: low (model's call-read: medium; verdict above is the hard-data one used for ranking) · confidence: 2/10

ClaimFigureArithmeticNext-FY Rev %Next-FY EPS %
~$1.1B FY2026 annual technology spend (automation + process improvement + AI, not AI-only)
cost · soft		>$1.1B annual spendDisclosed SPEND/input, not a saving or incremental revenue. $1,100M ÷ $15,912M FY2025 revenue = 6.91% of revenue (intensity only); = 51.5% of $2,135M NI but it is cash OUT and bundles non-AI automation. No disclosed opex reduction, ROI, or AI-only $ benefit → cannot compute next-FY rev_uplift_pct or eps_uplift_pct.
Over 10,000 associates using AI regularly (internal adoption)
engagement · soft		>10,000 associatesAdoption headcount only; no hours saved, FTE change, wage/opex base, or $ productivity value disclosed → cannot map to revenue or compute after-tax saving.
>3M lines of code/month written with AI assistance (developer productivity)
productivity · soft		>3M LOC/month with technologist oversightThroughput metric only; no developer FTE count, fully-loaded cost, or realized productivity % → saving_$ unknown; cannot apply eps_uplift_pct = 100 × (saving_$ × 0.79) ÷ $2,135M NI. Oversight implies augmentation, not replacement.
Ray adviser AI operations agent rolled out to a few hundred adviser teams
engagement · soft		a few hundred adviser teamsEarly pilot count, not a numeric numerator vs total advisers; no fee lift, AUM, revenue-per-adviser, or admin-hour $ savings disclosed → no anchored uplift.

Assumptions: Base: FY2025 (2025-09-30) revenue $15,912M, net income $2,135M, diluted EPS $10.30, net margin 13.42% (2,135÷15,912). EPS % would size against this earnings base (≈ consensus FY25 EPS). 21% tax and 13.42% incremental net margin reserved for any future $ benefit claim — not used, because no claim carries a dollar saving or revenue figure. The $1.1B is an investment outflow bundling non-AI automation, not an AI benefit, so it is not flowed to EPS. No phasing/bookings conversion (no incremental figure to phase). All quantified claims are adopter-side; zero supplier-side AI revenue.

Top line: No quantifiable topline AI impact. Every disclosed AI metric is an input or adoption count (>10,000 associates, >3M LOC/month, a few hundred advisers on Ray) — none ties to a revenue line. The only hard ratio is tech-spend intensity, $1,100M ÷ $15,912M = 6.91% of FY2025 revenue (investment, not incremental topline). Forward statements are explicitly qualitative ('significant opportunities,' 'still early innings'). Topline uplift = 0 estimable.

Bottom line: No dimensionable EPS impact. The only dollar figure, >$1.1B technology spend (6.91% of revenue, 51.5% of NI), is a cost OUTLAY bundling automation/process/AI — not an AI saving — so it cannot be flowed as an uplift. Management explicitly states the margin/cost-curve benefit is 'not yet dimensionable.' Productivity signals (>3M AI-assisted LOC/month, >10k AI users) are real and directionally margin-supportive, but with no disclosed FTE displacement or $ saving there is no honest after-tax number. EPS uplift = null (undimensioned, not absent).

[impact n/m (all claims soft/unanchored)] No quantified adopter rev/EPS bridge to compare to consensus. Vs FY2025 actuals, consensus FY26 implies revenue −2.1% ($15,912M→~$15,582M), EPS +15.4% ($10.30→$11.88), NI +14.8% ($2,135M→~$2,451M) — i.e. margin expansion on roughly flat/down revenue, exactly the operating-leverage/efficiency shape AI+automation would produce, suggesting the broad efficiency thrust is already partly reflected. But because RJF discloses NO AI-specific dollar benefit, the math cannot be shown ABOVE consensus (would be 'low') nor confirmed fully captured ('high'); the gap is indeterminate → medium.

MODEL CONSENSUS (impact)

partial

Near-identical answers (adopter-side, all claims soft/null, confidence 2); only priced_in differed — kept X's well-argued 'medium' over Y's 'high'.

Conflicts reconciled
FieldOpus 4.8GPT-5.5
Rev uplift %––
EPS uplift %––
Priced inmedium–
vs analystsunclear–
Confidence2–
Top lineNo quantifiable topline AI impact. Every disclosed AI metric is an input or adoption count (>10,000 associates, >3M LOC/month, a few hundred advisers on Ray) — none ties to a revenue line. Management frames AI as helping advisers 'deepen client relationships' and serve 'a larger client base,' but offers no incremental-revenue figure. Forward statements are explicitly qualitative ('significant opportunities,' 'still early innings'). Topline uplift = 0 estimable.–
Bottom lineNo dimensionable EPS impact. The only dollar figure, >$1.1B technology spend (6.9% of revenue, 51.5% of net income), is a cost OUTLAY that bundles automation/process/AI — not an AI saving — so it cannot be flowed as an uplift. Management states the cost-curve/margin benefit is 'not yet dimensionable' and AI is 'still early innings.' Productivity signals (>3M AI-assisted LOC/month, >10k AI users) are real and directionally margin-supportive, but with no disclosed FTE displacement or $ saving there is no honest after-tax number to compute. EPS uplift = null, not zero — undimensioned, not absent.–
ReasoningConsensus FY26 already embeds revenue of $15.58B (-2.1% vs FY25 actual $15.91B) yet EPS of $11.88 (+15.4% vs $10.3) and NI of $2.451B (+14.8%) — i.e. analysts expect margin expansion on roughly flat/down revenue, which is exactly the operating-leverage/efficiency story AI+automation would produce. So the broad efficiency thrust appears already reflected in the EPS-up-on-flat-revenue shape. But because RJF has disclosed NO AI-specific dollar benefit, you cannot show the math points ABOVE consensus (would be 'low' priced-in) nor confirm it's fully captured ('high'). With nothing to quantify, the gap is indeterminate — medium.–

Rows highlighted where the two models disagreed.

QUANTIFICATIONS
Annual technology spend (includes automation, process improvement, AI — not AI-only): more than $1.1 billion / close to $1.1 billion (FY2026 (annual), both)
“We will continue investing in automation, process improvement and AI as part of our more than $1.1 billion annual technology spend to create efficiencies, give advisers more time to deepen client relationships and further enhance the client experience.”
Associates using AI regularly (internal adoption): over 10,000 (Q1 FY2026 (current at time of call), bottomline)
“We already have over 10,000 associates that are using AI on a regular basis in 1 way, shape or form.”
Code written with AI assistance (developer productivity): over 3 million lines of code per month (Q1 FY2026 (current at time of call), bottomline)
“I think we've -- over 3 million lines of code are written a month using AI with oversight from the technologist in the group.”
Advisers/teams with AI operations agent (Ray) rollout: a few hundred (Q2 FY2026 (current at time of call), both)
“This agent has been rolled out to a few hundred advisers and their team so far in addition to service focus groups at the home office.”
PAST (realized)
CURRENT (now)
FORWARD (guidance)
TRACK RECORD — PROMISE vs DELIVERY

—/100 (no quantified promises)   no-quantified-promises  6 calls reviewed

Across six earnings calls (Q1 FY2025–Q2 FY2026), Raymond James discusses AI substantively—Chief AI Officer, back-office AI, a proprietary search tool, the RA operations agent, and $1B+ annual tech spend—but never states a forward-looking AI target with both a number and a deadline. Management emphasizes early-stage exploration and gradual rollout, so there is no quantified promise-vs-delivery track record to score.

PRICED-IN (REFINED)
LOW (room left)

Est. revisions falling  ·  Fwd P/E 15.4  ·  EV/Sales 1.4x

AI claim maps to Private Client Group, Asset Management Segment, Capital Markets

Estimate momentum is weak to negative: price targets have stepped down from $178 (last year) to $163 (last quarter) to $155 (last month), outweighing a modest shift in monthly ratings (strongBuy 2→3, hold 9→8). Forward consensus still embeds solid EPS growth (FY25–FY26), but that is ordinary trajectory, not an AI-specific ramp, and the broad consensus remains Hold. Valuation is not stretched for a capital-markets/wealth name (fwd P/E ~15.4, EV/Sales ~1.4, PEG ~1), so falling/flat revisions on a non-rich multiple leave room for an AI productivity or PCG/capital-markets thesis to surprise rather than being fully discounted.
COVERAGE — ENTHUSIASM TRAJECTORY + CATALYSTS
2Q1 FY20255Q2 FY20255Q3 FY20256Q4 FY20257Q1 FY20268Q2 FY2026

AI enthusiasm across 6 calls — trend ↗ rising

From generic tech spend to Chief AI Officer, RA agent launch, and phased adviser rollout with feedback.

RECENT AI CATALYSTS & NEWS
BUSINESS IMPACT - QUALITATIVE MATERIALITY

5/10 qualitative impact   moderate  medium-term · soft evidence

Where AI matters: PCG adviser productivity and internal ops

Ray and broad associate/developer AI use are real deployments on the core adviser workflow, but management will not tie them to revenue or margin and Ray remains a few-hundred-team pilot versus the full adviser base.

Caveats: No quantified AI ROI despite >$1.1B annual tech spend bundling AI; Ray adoption still early vs total advisers; Agentic cash-sweep or robo tools could erode ancillary spread and simple-advice economics; Rising AI spend is competitive table stakes before proven net margin benefit

AI DISRUPTION / CANNIBALIZATION RISK  tailwind · 3/10

AI may compress sweep spreads and commoditize basic advice or research, but RJF's fee and AUM model rests on relationship-led wealth management that management and clients still treat as human-adviser-led rather than bot-replaceable.

OPTIONS / MARKET STRUCTURE

option liquidity: fair

ATM IV
—
TYPICAL BID-ASK
—
OPEN INTEREST
—

proxy inputs — dollar-ADV $206M · beta 0.995 · px $146.75

source: proxy (no options chain on FMP)
FMP /stable/ exposes no options-chain endpoint on this key, so ATM IV, bid-ask spread and open interest are unavailable. Liquidity below is a PROXY from dollar-ADV, beta and price level (a stand-in for option depth), not measured option-market data.

CONFIRMATION — INSIDERS · 13F · LANGUAGE
Undercutting — insiders selling, institutions flat, management language 5/10 measured.
INSIDERS selling 2 open-market sell(s) vs 0 buy(s) — net distribution
INSTITUTIONS (13F) flat as of 2026-03-31: 100 new / 123 closed positions; 398 increased / 363 reduced; institutional ownership -0.67pp; -26 net 13F holders
MGMT LANGUAGE 5/10 measured Brief AI discussion; live agent with rollout numbers, but expansion hedged and no AI-specific spend or timelines.
commit “We will continue investing in automation, process improvement and AI as part of our more than $1.1 billion annual technology spend”
commit “This agent has been rolled out to a few hundred advisers and their team so far”
hedge “we'll continue to expand adviser and associate access over time”
VERBATIM AI QUOTES
“We continue to make investments and implement solutions to automate and streamline processes that provide advisers with incremental time to invest in their client relationships. Highlighting this is our newly launched proprietary digital AI operations agent named RA, which builds on our service-focused long-term AI strategy. The firm's suite of AI-based tools and technologies is focused on empowering financial advisers and professionals across the firm by applying artificial intelligence to enhance service models in secure, scalable applications.”
— Paul Shoukry, Q1 FY2026
“And coupling that with the platform, the technology, we've been investing close to $1.1 billion this year. the AI to support that, to help them save time, to help them make better decisions, to help them be more efficient in their operations with their clients and then the products.”
— Paul Shoukry, Q1 FY2026
“And that's hard to remain competitive when you can't invest in AI and the tools that you need to help advisers develop more efficiency in their businesses with their clients.”
— Paul Shoukry, Q1 FY2026
“if you look just at our investment in cybersecurity, the growth of AI investments and the development that we're doing with applications across all of our businesses, the infrastructure investment, there's a lot that goes into it.”
— Paul Shoukry, Q1 FY2026
“I do think long term, especially with AI, we will find more efficiencies in the cost structure as we deploy AI and automation. We're not going to dimension that or even put a time table on that now because it's still early innings. But we're starting to see some great benefits already. I mean we just launched Ray. We had a press release that came out Ray AI sort for kind of Raymond James a play on that. But it's natural language, sort of Q&A model, if you will, that uses generative AI that -- to answer questions for advisers and their sales systems and their teams -- that way, they don't even have to call in. And that's going to create efficiencies for them. That's going to allow our service people to spend their time on higher-value problems and solutions and opportunities. So again, we're not even in the first innings of those opportunities going forward.”
— Paul Shoukry, Q1 FY2026
“Really, we don't know yet. It's early. It's first inning of opportunities here and deployment. We already have over 10,000 associates that are using AI on a regular basis in 1 way, shape or form. So the penetration has been pretty significant. I think we've -- over 3 million lines of code are written a month using AI with oversight from the technologist in the group. So we are using AI to a pretty significant extent already, but I still think it's early innings. And the opportunities to expand that as these tools get smarter and more efficient is significant.”
— Paul Shoukry, Q1 FY2026
“But we won't rest on our laurels. We will continue investing in automation, process improvement and AI as part of our more than $1.1 billion annual technology spend to create efficiencies, give advisers more time to deepen client relationships and further enhance the client experience. For example, our proprietary AI operations agent provides curated natural language answers and guidance to operational questions while intelligently evolving based on user activities and preferences. This agent has been rolled out to a few hundred advisers and their team so far in addition to service focus groups at the home office. We are very encouraged by the strong initial feedback, and we'll continue to expand adviser and associate access over time.”
— Paul Shoukry, Q2 FY2026
“In a world being shaped by AI, technology and constant change we believe personal relationships will matter more, not less. Our strategy is to keep investing in the people, platforms and capabilities that help our financial professionals deliver more holistic more personalized advice to clients while staying true to the culture and long-term approach that have always differentiated Raymond James.”
— Paul Shoukry, Q2 FY2026
“I think it's already been helpful in our industry. And so we've had 3 client events in the last quarter with advisers and clients, 1 in Memphis, 1 in Atlanta, 1 in Miami, and I would tell you, when you see the adviser relationship with clients, there's no doubt that the deeply personal relationships that advisers have with clients trump any kind of technology or AI bot that may exist in the future.”
— Paul Shoukry, Q2 FY2026
“And so when we talk about AI we need to understand the value of those personal relationships advisers have with these families. It's not about transactions. It's not just about portfolio returns it's about really deeply understanding the family's financial objectives, and that's something that AI should help down the road because it will help advisers come up with more bespoke tailored insights that advice safe there, save them time on administrative tasks and allow them to spend more time developing those deeply personal relationships with their clients.”
— Paul Shoukry, Q2 FY2026
“a lot of the focus on AI across corporate America right now. And for us, it included, has been around the large language models and some of the sort of benefits of an efficiency and increased productivity that large language models can provide by synthesizing a lot of data and we rolled it out. We have a solution called [ Ray ] that we rolled out and that advisers and sales assistants can use to find -- to sit through a lot of information, self-service and very quickly find the answers the complicated questions. And so we're piloting it with a few hundred advisers in the early feedback has been extremely positive. But what we're all wondering is the next phase of AI really is around Agentic AI and what can Agentic AI do to improve processes and streamline processes and ultimately the cost curve across not only our industry, but all industries. And we're still -- I think Raymond James and all of corporate America is still early in that journey, frankly. And so we think that there will be significant opportunities. The compute power being invested is substantial and significant. We're using AI in a lot of areas already, whether it be in our cybersecurity area, although that's continuing to evolve as we saw with a new release a couple of weeks ago and some of the notifications from Washington around that. So we're using AI, and we're seeing a lot of benefits from AI, but it's hard to dimension the actual margin impact at this juncture. I think anyone who's talking about cost reductions or margin benefits from AI today at least I would be arbitrary and too preliminary in providing that type of specificity.”
— Paul Shoukry, Q2 FY2026
“When we look at these AI releases, some of which have negative impacts on our stocks in our industry. I look at it as these are releases that could be extremely helpful to us and to our advisers to help provide more bespoke advice to a larger number of clients. And so that's really what we are here to do is help advisers better help their clients.”
— Paul Shoukry, Q2 FY2026
ANALYST QUESTIONS ON AI
Q (Q1 FY2026, Michael Cyprys (Morgan Stanley)): Follow-up on AI: aspirations for automating processes and the AI operations agent Ray; how usage/adoption ramps vs. today; anticipated ROI; scope for additional agents and an agented workforce at Raymond James.
A: Paul Shoukry: "Really, we don't know yet. It's early. It's first inning of opportunities here and deployment. We already have over 10,000 associates that are using AI on a regular basis in 1 way, shape or form... over 3 million lines of code are written a month using AI with oversight from the technologist in the group... we are using AI to a pretty significant extent already, but I still think it's early innings. And the opportunities to expand that as these tools get smarter and more efficient is significant."
Q (Q1 FY2026, Unknown Analyst (Michael, for Alex Blostein, Goldman Sachs)): Elaborate on what specifically is driving non-comp growth in 2026, including tech investments.
A: Paul Shoukry cited cybersecurity, "the growth of AI investments and the development that we're doing with applications across all of our businesses," and infrastructure; described Ray as a generative-AI natural-language Q&A tool for advisers that creates efficiencies and frees service staff for higher-value work; said long-term AI/automation efficiencies are expected but not yet dimensioned or timed.
Q (Q2 FY2026, Devin Ryan (Citizens Bank)): Why management sees AI as a net positive vs. risk; thoughts on implications of a potential agentic AI cash-sweep optimization tool and whether RJF would evolve monetization (e.g., platform fee).
A: On cash optimization: called the tool conceptual; argued sweep-cash decline already happened without AI and sees limited incremental threat to adviser-led wealth vs. e-brokers; sweep balances have stabilized. On AI broadly: personal adviser relationships "trump any kind of technology or AI bot"; AI "should help down the road" by delivering bespoke insights, saving admin time, and freeing advisers for client relationships.
Q (Q2 FY2026, Michael Cho (JPMorgan)): Unpack $1.1B tech spend priorities; how RJF gauges success of AI initiatives including the operational chatbot; how the next step evolves as capabilities roll out.
A: Paul Shoukry: vast majority of spend focused on Private Client Group/adviser tools; development guided by a Technology Advisory Council of financial advisers providing real-time feedback; success measured via adviser feedback, recruiting visits, and technology awards — no AI-specific KPIs or ROI metrics cited.
Q (Q2 FY2026, Alexander Blostein (Goldman Sachs)): Longer-term profitability: benefits of AI and related initiatives on margins over time, given ~20% pretax margin today.
A: Paul Shoukry: LLM productivity benefits via Ray pilot ("extremely positive" early feedback); next phase is Agentic AI for process/cost-curve improvement; also using AI in cybersecurity; "hard to dimension the actual margin impact at this juncture" and margin-benefit estimates today would be "arbitrary and too preliminary."
Q (Q2 FY2026, Steven Chubak (Wolfe Research)): Follow-up on Agentic AI impact on cash levels; if competitors pivot to fee-based models to reduce cash-economics reliance, is RJF amenable given omnichannel affiliation structure?
A: Paul Shoukry: would be "flexible and open to evolving" pricing if competitive/client dynamics shift; ongoing review of all pricing/fees/payout — not an AI deployment answer.
Q (Q2 FY2026, Mike Cyprys (Morgan Stanley)): How RJF will deepen adviser relationships and evolve offerings in a quickly evolving world.
A: Paul Shoukry: culture plus "$1.1 billion on technology and AI"; advisers expected to deliver more holistic/bespoke advice; "that's where AI can actually help" — industry AI releases could help advisers provide bespoke advice to more clients.