← back to rankingPODD · Insulet Corporation
Medical - Devices · mkt cap $9.9B · calls: Q1 FY2026 vs Q4 FY2025
45.0 conviction · conf-adj 45
conf 3/10 Opus+GPT ✓ agree
enthusiasm:18.0 · trend:0 · quantifies:0 · impact:0 · under_radar:5 · credibility:12 · business_impact:8 · disruption:0 · commitment:-4 · confirmation:6
Enthusiasm latest 6 / prev 7 (flat)
Insulet's AI story is overwhelmingly product-side—iterative insulin-delivery algorithms, automated/fully closed-loop autonomy, Omnipod Discover insights, and commercial targeting tools—not a corporate AI platform narrative. Q4 FY2025 is the only call with an explicit AI mention (cloud data ecosystem for customer-service efficiency and cost-to-serve); Q1 FY2026 doubles down on algorithm cadence and closed-loop clinical data but drops the AI label. Management never ties AI or analytics to a dollar figure for revenue, margin, or headcount savings; clinical quantification (~5% and 68% time in range) supports differentiation credibility but not P&L attribution.
GROUNDED NEXT-FY IMPACT vs CONSENSUS
Grounded on actual base — revenue $2.7B · net income $0.2B · net margin 9.1% · diluted EPS 3.48
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: medium (model's call-read: high; verdict above is the hard-data one used for ranking) · confidence: 3/10
| Claim | Figure | Arithmetic | Next-FY Rev % | Next-FY EPS % |
|---|
~5% simulated time-in-range improvement (100 vs 120 mg/dL target option)
engagement · soft | ~5% TIR improvement (simulated) | Clinical efficacy metric (Q1 FY2026 algorithm release), NOT a $ figure. Time-in-range is a glycemic-control outcome, not revenue or cost; a 5% TIR gain strengthens Omnipod 5/6 competitiveness and retention but management gave NO attach rate, ASP, patient base, churn, or incremental-revenue dollar in any claim. Cannot apply rev_uplift_pct=100*claim_$/2,708.1M (claim_$ undefined). Mapping 5% TIR->5% revenue is rejected as conflating a clinical % with a financial % (would wrongly imply ~$135M rev -> ~$12.4M NI @9.124% -> ~3.45% vs consensus NI / 5.0% vs GAAP NI). Not estimable as next-FY uplift. | | |
68% time-in-range, type-2 fully closed-loop with no boluses (ATTD feasibility)
other · soft | 68% TIR | Feasibility topline efficacy ahead of EVOLVE pivotal; commercial launch cited 2028 — outside next-FY (FY2026) revenue recognition. 68% is an absolute TIR level, not a disclosed incremental revenue %, $ bookings, or share-of-TAM. Type-2 TAM is large but unquantified here. No claim_$ -> rev/eps undefined. Not estimable as a next-FY uplift. | | |
3 algorithm improvements over the next 3 years (2026-2029)
other · soft | 3 algos / 3 yrs | Roadmap cadence (incl. 3rd-gen algorithm + Gen-6 Omnipod in 2027), pushed OTA to installed Pods. Zero dollar content — no per-release incremental revenue, adoption lift, pricing, or margin impact. Cannot allocate consensus growth to it without inventing per-algorithm %. It is the qualitative engine behind consensus growth, not a separable, sizeable uplift. | | |
Assumptions: All claims are adopter-side (Omnipod algorithm / closed-loop product efficacy); no supplier-side AI-compute revenue. NO claim carries a dollar figure (revenue, bookings, cost saving, or headcount), so no incremental-margin/tax/phasing arithmetic was applicable. Had a $ claim existed, EPS sizing would use the consensus adjusted basis (NI ~$357.8M, EPS $4.88) — NOT GAAP NI $247.1M (EPS $3.48), which would inflate EPS% as a thin-base artifact; default incremental net margin = FY2025 9.124%, tax 21% for opex saves. Phasing: Claim 1 Q1 FY2026 (in P&L window but no $); Claim 2 launch 2028 (0% next-FY revenue); Claim 3 2026-2029 (no FY2026 $ phasing given). No bookings->revenue conversion required.
Top line: Not separately estimable; aggregate adopter rev_uplift_pct = null. Insulet's AI is the algorithm layer inside Omnipod 5/6 (target-setting, automated insulin delivery, type-2 closed loop). The quantified items — ~5% simulated TIR improvement, 68% type-2 TIR feasibility, 3 algos in 3 years — are clinical-efficacy and roadmap metrics with no dollar anchor, so no clean additive rev_uplift_pct is computable without fabricating an adoption model. Directionally these sustain share gains in a competitive AID market, but that benefit is already embedded in the ~+23% FY26 consensus revenue ramp (~2,708M -> ~3,331M), not an incremental layer on top.
Bottom line: Not estimable; aggregate adopter eps_uplift_pct = null. No cost saving, FTE reduction, or productivity-in-dollars was disclosed, and no incremental revenue can be flowed to EPS. The algorithm roadmap delivered OTA to the installed base is favorable for incremental margins (monetizes existing Pods at low marginal cost) but management quantified none of it. Sizing an EPS% off the GAAP base (NI 247.1M, 9.1% margin) would be misleading versus the adjusted basis consensus uses; with no anchored figure, est_eps_uplift_pct is null.
[impact n/m (all claims soft/unanchored)] Adopter-side quantified claims yield null est_rev_uplift_pct and null est_eps_uplift_pct, so no measurable gap vs a consensus AI line-item. Consensus already models robust roadmap-driven growth: revenue ~2,708M -> ~3,331M FY26 (+23%) -> ~3,958M FY27 (+18.8% YoY); adjusted EPS 4.88 -> 6.48 (+32.8%) -> 8.12 FY27 (+25.3%); implied NI ~$357.8M -> ~$465.7M -> ~$566.5M. The Gen-6 Omnipod + 3rd-gen algorithm (2027) and the algorithm cadence ARE the engine of that ramp, so the AI/algorithm story is baked into the trajectory rather than sitting above it. Management gave no incremental dollar figure to push numbers beyond this curve. The one genuinely new TAM lever — type-2 fully-closed-loop (68% TIR feasibility) — is a 2028+ launch (EVOLVE pivotal just enrolling), beyond the FY26-27 window, so it is likely only partially in outer-year estimates but not quantifiable or isolatable today.
MODEL CONSENSUS (impact)
Opus+GPT ✓ agree
Both analysts independently reached identical verdicts: all three claims clinical/cadence-only with no $ anchor, so all pcts null/soft, aggregates null, adopter-side, priced_in high. Merged the strongest bases.
| Field | Opus 4.8 | GPT-5.5 |
|---|
| Rev uplift % | – | – |
| EPS uplift % | – | – |
| Priced in | high | – |
| vs analysts | unclear | – |
| Confidence | 3 | – |
| Top line | Not separately estimable. Insulet's AI is the algorithm layer inside Omnipod 5/6 (target-setting, automated insulin delivery, type-2 closed loop). The quantified items — ~5% simulated TIR improvement, 68% type-2 TIR feasibility, 3 algos in 3 years — are clinical-efficacy and roadmap metrics with no dollar anchor, so no clean rev_uplift_pct can be computed without fabricating an adoption model. Directionally these improvements sustain share gains in a competitive AID market, but that benefit is already embedded in the +23% FY26 consensus revenue ramp (2,708M -> 3,331M), not an incremental layer on top of it. | – |
| Bottom line | Not estimable. No cost saving, FTE reduction, or productivity-in-dollars was disclosed, and no incremental revenue can be flowed to EPS. The algorithm roadmap is delivered OTA to the installed base (favorable for incremental margins as it monetizes existing Pods at low marginal cost), but management quantified none of it. Sizing an EPS% off the GAAP base (NI 247.1M, margin 9.1%) would in any case be misleading versus the adjusted basis consensus uses; with no anchored figure, est_eps_uplift_pct is null. | – |
| Reasoning | Consensus already models robust roadmap-driven growth: revenue 2,708M -> 3,331M FY26 (+23.0%) and adjusted EPS 4.88 -> 6.48 (+32.8%) -> 8.12 FY27 (+25.3%). The Gen-6 Omnipod + 3rd-gen algorithm (2027) and the algorithm cadence ARE the engine of that ramp, so the AI/algorithm story is baked into the trajectory rather than sitting above it. Management provided no incremental dollar figure that could push numbers beyond this curve. The one genuinely new TAM lever — type-2 fully-closed-loop (68% TIR feasibility) — is a 2028+ launch (EVOLVE pivotal just enrolled first participant), beyond the FY26-27 window consensus covers, so it is largely NOT yet in numbers but also not quantifiable today. | – |
Rows highlighted where the two models disagreed.
QUANTIFICATIONS
Time in range improvement from algorithm target setting (simulated): approximately 5% (Q1 FY2026 launch (simulated analysis cited on call), both)
“our simulated analysis switching to target glucose setting from 120 milligrams per deciliter to our new 100 milligrams per deciliter option delivered an approximately 5% improvement in time and range.”
Time in range — type 2 fully closed-loop feasibility: 68% (feasibility study presented at ATTD (ahead of EVOLVE pivotal), topline)
“We're encouraged by the results from our feasibility study that we presented at ATTD, which highlighted 68% time in range with no boluses.”
Planned algorithm release cadence: 3 algorithm improvements over the next 3 years (2026–2029 window referenced on Q1 call, topline)
“we have 3 algorithm improvements over the next 3 years.”
PAST (realized)
- Q4 FY2025 / Flavia Pease: For the year, we successfully optimized both our cost to acquire and our cost to serve, 2 key metrics we remain focused on improving as we enhance customer lifetime value.
- Q4 FY2025 / Ashley McEvoy: We completed the STRIVE study for Omnipod 6 and moved into the next phase of our EVOLUTION study supporting our fully closed-loop system for adults with type 2 diabetes.
- Q4 FY2025 / Ashley McEvoy: This weekend, we launched a limited market release with Omnipod 5 with a lower set point, advanced automated mode.
CURRENT (now)
- Q1 FY2026 / Ashley McEvoy: Beginning with this year's launch of our second-generation algorithm coupled with our Libre 3 Plus sensor integration, and the broader rollout of Omnipod Discover, our new data insights platform.
- Q1 FY2026 / Ashley McEvoy: We're deploying tools to optimize physician targeting and conversion while expanding reach and frequency.
- Q4 FY2025 / Ashley McEvoy: And with the help of AI, we are increasingly tapping into our unique cloud-based data ecosystem to enhance customer service efficiency and satisfaction, reducing our cost to serve while strengthening retention.
- Q4 FY2025 / Ashley McEvoy: And our enhanced insights and analytics capability are helping us optimize our cost to acquire and cost to serve, driving continued expansion in customer lifetime value.
FORWARD (guidance)
- Q1 FY2026 / Ashley McEvoy: We are making strong progress on our sixth generation Omnipod paired with our third-generation algorithm, which is planned to launch in 2027.
- Q1 FY2026 / Ashley McEvoy: We enrolled our first participant in EVOLVE, our pivotal study to support FDA filing next year and launch in 2028.
- Q1 FY2026 / Ashley McEvoy: We have 3 algorithm improvements over the next 3 years.
- Q4 FY2025 / Ashley McEvoy: We expect to initiate our pivotal EVOLUTION study this year, supporting a regulatory filing in 2027 and a commercial launch in 2028.
- Q4 FY2025 / Eric Benjamin: With Omnipod 6, we can always push the latest technologies directly to Pods that customers have.
TRACK RECORD — PROMISE vs DELIVERY
88/100 track record delivers 6 calls reviewed
Insulet makes few hard numeric AI targets but reliably executes dated algorithm and sensor-integration milestones (iOS/G7, international G7, Libre 3 Plus, 100 mg/dL algorithm). Multi-year Omnipod 6 (2027) and fully closed-loop type 2 (2028) timelines are progressing on schedule but not yet judgeable.
Launch iOS app integrated with Dexcom G7 in the U.S. in the first half of 2025 — promised Q4 FY2024
delivered Q2 FY2025 confirmed a full U.S. iOS+G7 launch in June after a limited release in Q1; Q1 FY2025 had cited >40% of eligible users on iOS.
Roll out Dexcom G7 internationally starting with U.K. and Netherlands in Q1 2025 — promised Q4 FY2024
delivered Q1 FY2025 reported G7 live in both the U.K. and Netherlands with additional markets planned through the year.
Full U.S. market release of iOS with G7 before end of Q2 FY2025 — promised Q1 FY2025
delivered Q2 FY2025 stated the iOS+G7 integration was fully launched in June 2025.
U.S. FreeStyle Libre 3 integration on track for first half of 2026 — promised Q3 FY2025
delivered Q1 FY2026 said Libre 3 Plus integration is launching that quarter, consistent with the H1 2026 target and paired with new algorithm updates.
Launch Omnipod 5 algorithm update with 100 mg/dL target setting in 2026 — promised Q4 FY2025
delivered Q1 FY2026 reported the enhancement launching that quarter with simulated ~5% time-in-range improvement and increased automated-mode uptime.
Initiate pivotal EVOLUTION study in 2026; target FDA filing 2027 and commercial fully closed-loop type 2 launch in 2028 — promised Q4 FY2025
too-early Q1 FY2026 confirmed first EVOLVE pivotal participant enrolled and reiterated 2027 filing / 2028 launch; feasibility data showed 68% time-in-range without boluses.
PRICED-IN (REFINED)
MEDIUMEst. revisions falling · Fwd P/E 29.2 · EV/Sales 3.6x
AI claim maps to International Omnipod, UNITED STATES
Estimate momentum is weak: buy/strong-buy counts drifted down and holds rose since early 2026, and price targets stair-step lower (lastMonthAvg $231 vs lastQuarter $242 vs lastYear $319), so the street is not actively raising the AI/growth bar. Valuation is still moderately rich at ~29x next-FY EPS and ~3.6x EV/Sales with consensus already embedding fast growth (FY26–27 revenue ~+24%/+19%, EPS ~+33%/+25%), so much of the core growth story is in the multiple even as revisions soften. AI-driven upside for Omnipod would most plausibly flow through the Omnipod product line and the U.S. geography, not the tiny Drug Delivery line. Mixed signals—falling revisions but not cheap—support a medium priced-in verdict: incremental AI upside is not clearly being revised up, but the stock is not a clear low-multiple catch-up name either.
COVERAGE — ENTHUSIASM TRAJECTORY + CATALYSTS
6Q4 FY20244Q1 FY20255Q2 FY20255Q3 FY20255Q4 FY20256Q1 FY2026
AI enthusiasm across 6 calls — trend → flat
Steady algo and data-platform story; CEO transition diluted specificity; no explicit AI push despite Discover and closed-loop depth.
RECENT AI CATALYSTS & NEWS
BUSINESS IMPACT - QUALITATIVE MATERIALITY
8/10 qualitative impact material medium-term · mixed evidence
Where AI matters: AID control algorithms, closed-loop product, data insights
ML/control algorithms are the core Omnipod 5/6 value proposition—OTA cadence, Libre integration, Discover insights, and type-2 fully closed-loop roadmap drive share, retention, and a future TAM lever—but management cites only clinical metrics (~5% TIR, 68% type-2 feasibility) with no incremental revenue or margin dollars, and much of the roadmap is already in the growth curve.
Caveats: No management P&L attribution—upside is clinical/roadmap narrative, not separable EPS; Competitive AID algorithm arms race (Tandem, Medtronic, Abbott) can neutralize differentiation; Type-2 fully closed-loop revenue is 2028+ and still pivotal-stage; Algorithm gains may already be embedded in ~23% FY26 consensus revenue ramp
AI DISRUPTION / CANNIBALIZATION RISK tailwind · 2/10
Insulet sells physical disposable pods and regulated automated delivery; better algorithms reinforce rather than replace that hardware-consumables model, and there is no plausible near-term path for GenAI to automate away insulin infusion or collapse pump pricing the way it deflates billable services.
OPTIONS / MARKET STRUCTURE
option liquidity: fair
proxy inputs — dollar-ADV $181M · beta 1.197 · px $142.43
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
Mixed — insiders buying, institutions adding, management language 1/10 hedged.
INSIDERS buying 3 open-market buy(s) vs 1 sell(s) — net accumulation
INSTITUTIONS (13F) adding as of 2026-03-31: 86 new / 139 closed positions; 411 increased / 270 reduced; institutional ownership -0.76pp; -62 net 13F holders
MGMT LANGUAGE 1/10 hedged No explicit AI/ML; AID means insulin delivery, not artificial intelligence.
VERBATIM AI QUOTES
“Beginning with this year's launch of our second-generation algorithm coupled with our Libre 3 Plus sensor integration, and the broader rollout of Omnipod Discover, our new data insights platform.”
— Ashley McEvoy, Q1 FY2026
“And our simulated analysis switching to target glucose setting from 120 milligrams per deciliter to our new 100 milligrams per deciliter option delivered an approximately 5% improvement in time and range.”
— Ashley McEvoy, Q1 FY2026
“Additionally, we improved the algorithm performance, so it now increases the amount of time users spend in automated mode with fewer interruptions during extended high glucose events.”
— Ashley McEvoy, Q1 FY2026
“Algorithm innovation will continue to be a key R&D focus.”
— Ashley McEvoy, Q1 FY2026
“We are making strong progress on our sixth generation Omnipod paired with our third-generation algorithm, which is planned to launch in 2027.”
— Ashley McEvoy, Q1 FY2026
“We are sharing data from STRIVE, our Omnipod 6 pivotal study at ADA in June, which will demonstrate continued improvement in automation and clinical outcomes.”
— Ashley McEvoy, Q1 FY2026
“We are making progress on what we believe will be the first of its kind, truly fully closed loop system for people with type 2 diabetes.”
— Ashley McEvoy, Q1 FY2026
“We're encouraged by the results from our feasibility study that we presented at ATTD, which highlighted 68% time and range with no boluses.”
— Ashley McEvoy, Q1 FY2026
“Commercially, we are upskilling our sales force to strengthen our messaging our clinical performance in the field, and we're deploying tools to optimize physician targeting and conversion while expanding reach and frequency.”
— Ashley McEvoy, Q1 FY2026
“Algorithms. David, we were slow out of the gate continuously to improve our algorithms. We've addressed that now, and we have 3 algorithm improvements over the next 3 years.”
— Ashley McEvoy, Q1 FY2026
“This new algorithm is going to have greater automation, it's going to have less bolusing, it's going to have a reduced user interaction, it was designed exactly to bring more people into the category.”
— Ashley McEvoy, Q1 FY2026
“And our enhanced insights and analytics capability are helping us optimize our cost to acquire and cost to serve, driving continued expansion in customer lifetime value.”
— Ashley McEvoy, Q4 FY2025
“Omnipod Discover is a new data platform that delivers clear streamlined insights to support efficient health care professional review of Omnipod 5 data and enable more confident prescribing. Discover provides users with actionable guidance and reassurance, strengthening engagement and adherence.”
— Ashley McEvoy, Q4 FY2025
“This includes algorithm updates that enable a 100 set point target for tighter glycemic control, increased time in automated mode and improve responsiveness to enhance both the user experience and clinical outcomes.”
— Ashley McEvoy, Q4 FY2025
“It will feature a smarter algorithm to further personalize insulin delivery with pivotal data to be presented at ADA in June.”
— Ashley McEvoy, Q4 FY2025
“It is a system that delivers therapy effortlessly, adapting automatically without any user intervention. No dosing, no mealtime actions and no required adjustments while the Pod is worn.”
— Ashley McEvoy, Q4 FY2025
“And with the help of AI, we are increasingly tapping into our unique cloud-based data ecosystem to enhance customer service efficiency and satisfaction, reducing our cost to serve while strengthening retention.”
— Ashley McEvoy, Q4 FY2025
“For the year, we successfully optimized both our cost to acquire and our cost to serve, 2 key metrics we remain focused on improving as we enhance customer lifetime value.”
— Flavia Pease, Q4 FY2025
“it's part of why we're designing one Pod that can be updated in market for faster innovation so that with Omnipod 6, we can always push the latest technologies directly to Pods that customers have.”
— Eric Benjamin, Q4 FY2025
ANALYST QUESTIONS ON AI
Q (Q4 FY2025, Joanne Wuensch): ADA is going to be here before we know it. Is there anything in particular that we should look forward to there? And I'm also trying to key in on when are we going to get a line of sight on some of the clinical steps for Omnipod 6?
A: Ashley/Eric: ADA will include pivotal STRIVE results for Omnipod 6 (third-generation algorithm); ATTD/feasibility data on EVOLUTION fully closed loop for type 2 (no bolus, no user intervention, no clinician-defined settings, self-initiated training); additional independent AID comparison studies.
Q (Q1 FY2026, Shagun Singh Chadha): You've called out three algorithm launches in the 3 years, how meaningful are those upgrades and the U.S. sales force expansion, any way to think about the pace of that?
A: Ashley: Commercial focus on selling optimized settings and time in range (new 100 mg/dL set point, more automated mode), Libre 3 Plus integration, expanded clinical evidence on the website, peer-to-peer education on clinical performance, plus sales force expansion and targeting/reach.
Q (Q1 FY2026, Matthew Taylor): On the tailwinds in '27, specifically on Omnipod 6 — do you expect that launch to drive increased share gains and customer starts, or price/mix benefits as well?
A: Ashley: Omnipod 6 extends leadership and MDI conversion; third algorithm improvement with greater automation, less bolusing, reduced user interaction; STRIVE data at ADA; improved on-body placement via over-the-air updates; single-pod chassis simplifying prescribing and supply chain.