← back to rankingWELL · Welltower Inc.
REIT - Healthcare Facilities · mkt cap $137.9B · calls: Q1 FY2026 vs Q4 FY2025
56.0 conviction · conf-adj 55
conf 5/10 partial
enthusiasm:27.0 · trend:8 · quantifies:5 · impact:0 · under_radar:5 · credibility:0 · business_impact:8 · disruption:0 · commitment:0 · confirmation:3
Enthusiasm latest 9 / prev 7 (rising)
Welltower’s AI thesis moved from internal data-science advantage and operational technology in Q4 FY2025 to explicit AI monetization in Q1 FY2026 through supervised and unsupervised model licensing. Management ties the platform to faster capital allocation, external fee opportunities, acquisition sourcing, and operational automation, with several concrete figures. Credibility is relatively high because management cites long-running use since 2016, realized transaction activity, signed external partnerships, and fund economics, while still admitting operating technology is early.
GROUNDED NEXT-FY IMPACT vs CONSENSUS
Grounded on actual base — revenue $10.7B · net income $0.9B · net margin 8.8% · diluted EPS 1.39
These are next-fiscal-year annual uplift estimates, not next-quarter numbers.
Aggregate next-FY est. rev uplift: 0.0026577% · next-FY EPS uplift: 0.0136179% · vs analysts: behind · priced in: medium (model's call-read: high; verdict above is the hard-data one used for ranking) · confidence: 5/10
| Claim | Figure | Arithmetic | Next-FY Rev % | Next-FY EPS % |
|---|
$80B acquisition/disposition activity supported over 10 years via data-science/ML
other | over $80B cumulative over 10 years | $80B/10yr = ~$8.0B annual gross transaction flow; this is investing/capital-deployment volume, not P&L revenue, and acquisitions add rental revenue only over time at undisclosed yields/cap rates. No revenue/EPS mapping obtainable from inputs. | | |
Capital-allocation decision cycle cut from 5-9 months to weeks
productivity · soft | 5-9 mo -> 'mere weeks' | Midpoint 7mo to ~4wk (0.923mo) = ~86.8% cycle-time reduction; velocity gain only, no dollar saving or revenue conversion disclosed. Unanchored to $. | | |
2025 screening platform: several thousand acquisitions screened -> ~1,000 communities curated
engagement · soft | ~1,000 communities | ~1,000/>=3,000 = at most ~33% funnel conversion; count metric with no per-community revenue base disclosed, so no revenue/cost base obtainable. | | |
welltower.ai-supported transactions Q1'26: 37 communities, 4,200+ units
engagement · soft | 37 communities / 4,200 units | 4,200/37 = 113.5 units/community; no rent-per-unit, cap rate, NOI yield or purchase price disclosed, so incremental rental revenue cannot be sized. Volume metric, not P&L. | | |
Fund management fees on ~$2.1B third-party capital @1.35% (attributed to data science)
revenue | $2.1B x 1.35% | $2.1B x 0.0135 = $28.35M annual fee run-rate. $28.35M/$10,667.136M rev = 0.266%. Fee income is high-margin; using averaged 45% incremental net margin (X 50% / Y 40%) -> $12.7575M NI; $12.7575M/$936.845M NI = 1.362% EPS. | 0.0026577 | 0.0136179 |
Fund economics '2x higher' than peers, attributed to data-science platform
revenue | 2x | Relative multiplier on the same fee stream, not additional revenue; the AI premium (~half of the $28.35M) is subsumed within the fund-fee row above and NOT added to the aggregate to avoid double-counting. Anchored multiplier but no standalone base. | | |
Occupancy growth 400bps vs NIC ~250bps, attributed to WBS
revenue · soft | 400 vs 250 bps (150bps outperformance) | 150bps occupancy outperformance; SHO operating-portfolio revenue base is not disclosed in inputs and bps-occupancy != bps-revenue (rate-dependent), so no obtainable base to size. Directional topline only. | | |
Assumptions: Current revenue=$10.667136B, current net income=$936.845M, net margin=8.78%. Fund-management fee revenue treated as a full next-FY annualized run-rate (fund at final close) and modeled at 45% incremental net margin — the average of X's 50% and Y's 40%, both defensible (fee income is structurally high-margin vs the blended ~8.8% company margin). The '2x fund economics' multiplier is subsumed within the fund-fee row, not added, to avoid double-counting. Transaction-volume, screening, cycle-time and occupancy-bps claims left unsized because no per-unit/per-deal revenue or segment base is disclosed in the inputs. All claims are adopter-side; no supplier-side revenue. No quantified cost-saving claim was disclosed. (Note: pct fields expressed as ratios per the method's claim_$/base definition.)
Top line: Only one AI claim maps to a revenue line: the ~$2.1B third-party fund at 1.35% blended fees = $28.35M annual fee run-rate, just 0.266% of $10.67B revenue. The larger AI narrative ($80B/10yr transaction volume, thousands of deals screened into ~1,000 communities, 37 communities/4,200 units via welltower.ai, 400 vs 250bps occupancy) is real and is the engine behind WELL's acquisition-led growth, but none carries a disclosed per-unit/per-deal revenue base, so it can't be sized discretely; its value shows up indirectly in the acquisition-driven revenue ramp consensus already models.
Bottom line: Discretely quantifiable AI EPS contribution is the fund-fee flow-through: $28.35M x 45% avg incremental margin = $12.76M, or +1.36% of $936.8M net income. Cycle-time compression and the 2x fund economics are genuine bottom-line levers but unanchored to a dollar figure, so left null. A small, real, single-digit-percent EPS tailwind vs a consensus that already assumes EPS roughly doubles.
Consensus revenue rises ~$10.32B (FY25) -> $13.76B (FY26), ~+29-33%; net income $1.24B -> $1.93B (+~106%); EPS roughly doubles. The hard AI uplift here is only ~0.27% of revenue and ~1.36% of net income — an order of magnitude smaller than the consensus growth step-up. That growth is acquisition-driven, the exact channel the AI platform enables, so the AI value is largely embedded in consensus rather than additive. Discrete AI delivery sits well below the priced-in growth bar; verdict conservatively 'behind'.
MODEL CONSENSUS (impact)
partial
Adopted Y's no-double-count treatment of the 2x claim, averaged the margin, standardized pct fields to ratios, and took the more conservative verdict where the two genuinely tied.
Conflicts reconciled
- est_rev_uplift_pct: X=0.003987 (incl. 2x premium) vs Y=0.266(%) -> used 0.0026577 ratio, fund-fee only, dropping the 2x premium per Y to avoid double-counting
- est_eps_uplift_pct: X=0.022693 (50% margin, incl premium) vs Y=1.21(%, 40% margin) -> used 0.0136179 ratio at averaged 45% margin, fund-fee only
- fund incremental margin: X=50% vs Y=40% -> averaged 45% (both defensible)
- vs_analyst_expectations: X=behind vs Y=inline -> used behind (conservative tie-break) and lowered confidence to 5
| Field | Opus 4.8 | GPT-5.5 |
|---|
| Rev uplift % | 0.266 | 0.003986538172290147 |
| EPS uplift % | 1.21 | 0.022692651392996574 |
| Priced in | high | high |
| vs analysts | inline | behind |
| Confidence | 6 | 5 |
| Top line | Only one AI claim maps to a revenue line: the $2.1B third-party fund at 1.35% blended fees = $28.35M annual fee run-rate, or just 0.27% of $10.67B revenue. The larger AI narrative (driving $80B of 10yr transaction volume, screening thousands of deals into ~1,000 communities, 37 communities/4,200 units via welltower.ai, 400 vs 250bps occupancy) is real and is the engine behind WELL's acquisition-led growth — but none of it carries a disclosed per-unit/per-deal revenue base, so it can't be sized discretely. Its value shows up indirectly in the acquisition-driven revenue ramp consensus already models (2026E +29%). | Hard adopter-side revenue math is limited to fund fees: $28.35M full fee revenue plus $14.175M AI-attributed fee-premium revenue equals $42.525M, or 0.399% of current revenue. Other claims show scale or operational velocity but do not disclose a revenue conversion base. |
| Bottom line | The discretely quantifiable AI EPS contribution is the fund-fee flow-through: $28.35M x 40% incr margin = $11.34M, or +1.21% of $936.8M net income. Cycle-time compression (5-9mo -> weeks) and the 2x fund economics are genuine bottom-line levers but unanchored to a dollar saving, so left null. Net: a small, real, single-digit-percent EPS tailwind vs a consensus already assuming EPS roughly doubles. | Using 50% incremental margin for fund fees, hard AI-linked net income impact is $21.2625M, or 2.269% of current net income. No quantified cost-saving claim was provided. |
| Reasoning | Quantifiable AI uplift = +0.27% revenue / +1.21% EPS. Consensus already projects 2026E revenue $13.76B vs $10.67B (+29%) and EPS $2.85 vs $1.39 (+105%), netinc $1.93B vs $0.94B (+106%). That ~29%/105% growth is acquisition-driven — the exact channel the AI platform enables — so the AI value is embedded in consensus, not additive on top. The standalone AI figures (sub-1.3% on both lines) are an order of magnitude smaller than the consensus growth gap; nothing in the claims points clearly above what analysts already assume. | Consensus revenue rises from $10.321B in FY2025 to $13.760B in FY2026, growth of $3.438B or 33.31%; the hard AI revenue uplift calculated here is only $42.525M, 0.399% of current revenue and about 1.24% of the FY2026 consensus revenue step-up. Consensus net income rises from $1.241B to $1.930B, growth of $688.8M or 55.51%; the hard AI net income uplift is $21.263M, about 3.09% of that consensus net-income step-up. |
Rows highlighted where the two models disagreed.
QUANTIFICATIONS
acquisition and disposition activity supported by data science and machine learning: over $80 billion (last 10 years, both)
“This was instrumental in driving over $80 billion of acquisition and disposition activity over the last 10 years.”
capital allocation decision cycle time: from 5 to 9 months to mere weeks (Q1 FY2026 external partnership disclosure, both)
“These models enabling the real-world application of AI by accelerating capital allocation decisions from 5 to 9 months to mere weeks and significantly increasing velocity to market.”
acquisition screening using data science platform: several thousand acquisitions; roughly 1,000 communities (2025, both)
“In 2025, we targeted and evaluated several thousand acquisitions using our data science platform, and from those, curated a portfolio of roughly 1,000 communities that we engaged and transacted with sellers mostly off-market.”
transactions supported by welltower.ai: 37 communities and over 4,200 units (Q1 FY2026, both)
“The majority of our acquisitions activity was highly granular, single-asset transactions where our teams operated as local sharpshooters supported by insights from our data science and machine learning platform, welltower.ai. These transactions added 37 communities and over 4,200 units to our seniors housing portfolio.”
fund management fees attributed to data science capabilities: 1.35% (U.S. Seniors Housing Fund I final close, topline)
“The fund was significantly oversubscribed, which we believe is a reflection of our data science capabilities and capital allocation track record. And the fund includes approximately $2.1 billion of third-party capital with blended management fees of 1.35%”
relative fund economics attributed to data science platform: 2x higher (Q4 FY2025 discussion of fund economics, topline)
“the economics that we received on our fund business that suggests to you, which is significantly higher, many would claim 2x higher than many others have received in the fund management business is a pure function of our capability of the data science platform”
occupancy outperformance attributed to WBS: 400 vs 250 (last year, discussed on Q4 FY2025 call, topline)
“You know what sort of our peers or NIC data or others have sort of reported, I believe, for NIC 99 occupancy growth last year was something 250 or something like that. And we did 400, sort of that gives you the answer to the first question -- last question.”
PAST (realized)
- Q1 FY2026, Shankh Mitra: "since 2016, through the efforts of multidisciplinary team of PHD computer scientists, engineers, statistician and mathematicians, we have pioneered the application of data science and machine learning in real estate investing."
- Q1 FY2026, Shankh Mitra: "This was instrumental in driving over $80 billion of acquisition and disposition activity over the last 10 years."
- Q4 FY2025, Shankh Mitra: "In 2025, we targeted and evaluated several thousand acquisitions using our data science platform, and from those, curated a portfolio of roughly 1,000 communities that we engaged and transacted with sellers mostly off-market."
- Q4 FY2025, Nikhil Chaudhri: "The fund was significantly oversubscribed, which we believe is a reflection of our data science capabilities and capital allocation track record."
CURRENT (now)
- Q1 FY2026, Nikhil Chaudhri: "The majority of our acquisitions activity was highly granular, single-asset transactions where our teams operated as local sharpshooters supported by insights from our data science and machine learning platform, welltower.ai."
- Q1 FY2026, Shankh Mitra: "Given the modular and portable nature of the platform, we launched our first external partnership during the first quarter, licensing bespoke, supervised and unsupervised models to public storage and a leading global private equity firm."
- Q1 FY2026, Shankh Mitra: "We are thrilled with the progress made by Tech Quad in reimagining our technology ecosystem to improve the resident and site level employee experience."
- Q4 FY2025, John Burkart: "We are methodically removing time-consuming administrative burdens that employees contend with on a daily basis, freeing them to focus on what they signed up for, taking care of residents."
FORWARD (guidance)
- Q1 FY2026, Shankh Mitra: "what we remain most excited about and our most meaningful opportunity to drive bottom line growth is through the expanded role that technology, data and innovation will play in our business with the ultimate goal of improving the experience of our customers and site level employees."
- Q1 FY2026, Shankh Mitra: "More to come on this front in months and quarters ahead, but we have been incredibly busy since the announcement in March as many highly respected real estate, non-real estate and sovereign wealth funds have reached out to us to explore similar partnerships."
- Q1 FY2026, Shankh Mitra: "Our newest addition to our team will only accelerate these efforts."
- Q4 FY2025, John Burkart: "technology, leverage technology to improve the customer and employee experience, automating processes and providing personalized experiences."
- Q4 FY2025, Shankh Mitra: "we continue to double down on that. And I think it is a long sort of a journey in front of us."
TRACK RECORD — PROMISE vs DELIVERY
—/100 (no quantified promises) no-quantified-promises 6 calls reviewed
Welltower talks heavily about its data-science/machine-learning platform (welltower.ai), the Welltower Business System (WBS), and a 'Tech Quad', describing 10M+ micro-markets, 8,000 employees trained on WBS, utilities down 2.1%, and continued NOI/margin gains. But every AI/technology claim is qualitative ('drive margins meaningfully higher', 'digital transformation') with no numeric target AND future timeframe ever attached to the AI capability; reported tech metrics are backward-looking actuals, not prior quantified promises, so there is no judgeable AI promise-vs-delivery track record.
PRICED-IN (REFINED)
MEDIUMEst. revisions flat · Fwd P/E 108.9 · EV/Sales 13.2x
AI claim maps to Senior Housing - Operating, Outpatient Medical
Estimate revisions look flat: rating counts have been largely unchanged, and last-month price targets are equal to the last-quarter average, though above the last-year average. Consensus revenue and EPS forecasts already embed strong forward growth, but the data does not show a clear recent upward migration in estimates. Valuation is rich at about 108.9x forward EPS and 13.2x EV/Sales, so AI upside tied mainly to Senior Housing - Operating and Outpatient Medical is partly reflected despite flat revisions. That combination points to medium priced-in risk rather than low, because a rich valuation means the market is already paying for substantial upside.
COVERAGE — ENTHUSIASM TRAJECTORY + CATALYSTS
6Q4 FY20246Q1 FY20256Q2 FY20257Q3 FY20257Q4 FY20258Q1 FY2026
AI enthusiasm across 6 calls — trend ↗ rising
The story progressed from data-platform advantages to a broader technology-led operating model tied to margins, employee experience, and growth duration.
RECENT AI CATALYSTS & NEWS
BUSINESS IMPACT - QUALITATIVE MATERIALITY
7/10 qualitative impact material medium-term · mixed evidence
Where AI matters: capital allocation, acquisitions, fund fees, SHOP operations
Welltower’s AI/data-science platform appears materially useful to its own business: it supports granular acquisition sourcing, faster underwriting, fund economics, and some operating automation in senior housing. The hard monetization is still modest versus total revenue, but the platform is tied to large capital-deployment workflows and new external model-licensing opportunities rather than generic productivity talk.
Caveats: External AI licensing may remain small relative to REIT rental revenue; Management attributes broad acquisition success to AI without fully isolating causality; Competitors can adopt similar underwriting and operating analytics; Operational technology in SHOP is still early and may face implementation friction
AI DISRUPTION / CANNIBALIZATION RISK tailwind · 1/10
AI does not plausibly commoditize Welltower’s core model of owning and operating healthcare real estate; physical assets, zoning, care operations, capital access, and local market relationships remain durable. AI may make competitors better at underwriting, but it does not automate away senior housing demand or rent/NOI collection.
OPTIONS / MARKET STRUCTURE
option liquidity: good
proxy inputs — dollar-ADV $622M · beta 0.825 · px $195.35
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
Confirming — insiders neutral, institutions adding, management language 5/10 measured.
INSIDERS neutral no open-market buys/sells in last 6mo (routine: 36 awards, 1 tax-withholding)
INSTITUTIONS (13F) adding as of 2026-03-31: 176 new / 128 closed positions; 825 increased / 453 reduced; institutional ownership -4.14pp; +49 net 13F holders
MGMT LANGUAGE 5/10 measured AI/ML is barely discussed; management owns welltower.ai usage, but most language is broad, future-oriented, and qualified.
commit “our most meaningful opportunity to drive bottom line growth is through the expanded role that technology, data and innovation will play in our business”
commit “supported by insights from our data science and machine learning platform, welltower.ai”
hedge “This digital transformation, which we are striving for”
VERBATIM AI QUOTES
“what we remain most excited about and our most meaningful opportunity to drive bottom line growth is through the expanded role that technology, data and innovation will play in our business with the ultimate goal of improving the experience of our customers and site level employees.”
— Shankh Mitra, Q1 FY2026
“The majority of our acquisitions activity was highly granular, single-asset transactions where our teams operated as local sharpshooters supported by insights from our data science and machine learning platform, welltower.ai.”
— Nikhil Chaudhri, Q1 FY2026
“since 2016, through the efforts of multidisciplinary team of PHD computer scientists, engineers, statistician and mathematicians, we have pioneered the application of data science and machine learning in real estate investing.”
— Shankh Mitra, Q1 FY2026
“This was instrumental in driving over $80 billion of acquisition and disposition activity over the last 10 years.”
— Shankh Mitra, Q1 FY2026
“Given the modular and portable nature of the platform, we launched our first external partnership during the first quarter, licensing bespoke, supervised and unsupervised models to public storage and a leading global private equity firm.”
— Shankh Mitra, Q1 FY2026
“These models enabling the real-world application of AI by accelerating capital allocation decisions from 5 to 9 months to mere weeks and significantly increasing velocity to market.”
— Shankh Mitra, Q1 FY2026
“Our relentless and manacle focus on the digital transformation of the business and dramatically improving customer and site level employee satisfaction will be the force multiplier on the attractive beta of our business.”
— Shankh Mitra, Q1 FY2026
“We launched our private funds management business, overhauled our internal and external incentive structure, made substantial progress on Welltower Business System initiatives and created our Tech Quad to take our technology journey to the next level.”
— Shankh Mitra, Q4 FY2025
“As I think about the next few years and beyond, our focus is simple. People, optimizing the human interaction, provided delightful experience; processes, remove bottlenecks and streamline flow; data, provide our operating partners with robust objective data to drive positive outcomes; and technology, leverage technology to improve the customer and employee experience, automating processes and providing personalized experiences.”
— John Burkart, Q4 FY2025
“We are methodically removing time-consuming administrative burdens that employees contend with on a daily basis, freeing them to focus on what they signed up for, taking care of residents.”
— John Burkart, Q4 FY2025
“The fund was significantly oversubscribed, which we believe is a reflection of our data science capabilities and capital allocation track record.”
— Nikhil Chaudhri, Q4 FY2025
“In 2025, we targeted and evaluated several thousand acquisitions using our data science platform, and from those, curated a portfolio of roughly 1,000 communities that we engaged and transacted with sellers mostly off-market.”
— Shankh Mitra, Q4 FY2025
“One is our data science platform, which is mature, but there's a lot of work to do.”
— Shankh Mitra, Q4 FY2025
“Now if your question is on the operational side, operational technology side, I wouldn't even call us mediocre. I will call us mediocre-minus.”
— Shankh Mitra, Q4 FY2025
ANALYST QUESTIONS ON AI
Q (Q1 FY2026, William John Kilichowski): Shankh, you kind of hit on this at the end of your opening remarks, but could you talk more about the growth of the talent density and the data science platform given what you described as a halo sector? And how much this has accelerated the growth outlook of the business in your mind? And then if you could also maybe just talk to how investors should be thinking about the medium-term potential for earnings contribution from this business?
A: Shankh Mitra: "we built this data science capability, machine learning capability over the last 10-plus years to deploy capital on our balance sheet, on our books." He added: "We're in the building mode of this business, where there's something substantial come out or not, we will see in the future" and "Our phones have been ringing up the hook."
Q (Q1 FY2026, Michael Carroll): Shankh, I know that the WBS model continues to evolve, I mean, how beneficial are these new partnerships that you're creating with PSA and others to take WPS to the next level. I mean, I'm assuming that Welltower is getting access to more new data that they didn't have access to Bulfor. I guess, how beneficial could that be as you kind of refine those systems?
A: Shankh Mitra: "think about in our SHOP technology in 2 different -- completely different segment... one is our data science platform, which is focused on allocation of capital and finding granular opportunity and changing the velocity that exists in this business from months to days right?" He clarified: "Welltower Business System, which is the operational side of the business, is not something that we are collaborating with public storage."
Q (Q1 FY2026, James Kammert): Shankh and team, is there a way to leverage the data science into other geographies beyond your core U.K., U.S. and Canada? Or are those markets structurally don't have private pay or other cultural issues that leave you unlikely to pursue in terms of external growth?
A: Shankh Mitra: "The short answer is, yes, it can be" and "it is absolutely scalable across geographies and product types and beyond real estate product that I just mentioned."
Q (Q1 FY2026, Richard Anderson): So Shankh, you talked about doing the hard things, not the easy things and making decisions with that mindset. And I'm thinking as you're talking about data analytics and all these sort of tangential opportunities that sort of spin out of senior housing platform... Are you thinking about senior housing as sort of a bed from which you grow other businesses outside of data centers or data analytics if you get my point, right, you become like a diversified vehicle. Is that kind of in your mind today?
A: Shankh Mitra: "We are not trying to go from senior living to other asset classes in real estate." He added: "we have built capabilities, right, such as this data business that we talked about could potentially become more than a platform that we use for an internal application, we'll see where we get to."
Q (Q4 FY2025, William John Kilichowski): My question is on the Tech Quad. You've already made such progress with Welltower Business System and with your data science platform. I'm just curious, what challenges are there left in the senior housing space to tackle? You're still hiring significantly. And then maybe an extension of that would be, are you building something that would eventually be monetizable?
A: Shankh Mitra: "There are 2 ways I think about our technology platform. One is our data science platform, which is mature, but there's a lot of work to do." He said fund economics show monetization: "the economics that we received on our fund business... is a pure function of our capability of the data science platform." On operating software: "we will never see us sell our operating software to someone else so that they can compete with us."
Q (Q4 FY2025, Ronald Kamdem): Just wanted to double click on some of the occupancy performance, both last year and sort of the guidance into -- going into '26. I'd love some updated thoughts in terms of how you guys drive sort of move-ins versus move-outs? And how much is WBS contributing to the outperformance versus the industry?
A: Shankh Mitra: "You know what sort of our peers or NIC data or others have sort of reported, I believe, for NIC 99 occupancy growth last year was something 250 or something like that. And we did 400, sort of that gives you the answer to the first question -- last question." John Burkart added that occupancy is driven by "the customer experience, the speed to lead, how we're answering the calls" and "optimize at each site."
Q (Q4 FY2025, Seth Bergey): I guess just going back to the funds business, you announced the debt funds, you've deployed some of the equity funds, kind of -- and you've talked a little bit about the funds business as a way to kind of monetize the Welltower Business System and the successes you've had with the data science platform. How -- kind of how do you see the trajectory of that funds business? And should we expect that to be kind of a larger piece of the story overtime?
A: Nikhil Chaudhri: "it's opportunistic and tactical, right? We're not asset gatherers." Shankh Mitra added: "We will take capital only if we think we can make a significant return on it. Otherwise, we won't."