- Federal open enrollment for 2027 coverage runs November 1, 2026 to January 15, 2027, confirmed by CMS in a statement dated July 31, 2026.
- The December 15 close is still widely published and it is out of date. That is 76 days of season, not 45.
- The pool inside that window is smaller than last year, and automatic renewals fell hardest.
- Per CMS, 76% of active plan selections on the federal exchange now come through an agent or broker, up from 71%.
Quick answers:
- When does ACA open enrollment for 2027 actually end?
- Why do so many sites still say December 15?
- Does the January 15 date apply in every state?
- Does the headline enrollment number mean people covered?
- Does open enrollment change U65 off-exchange volume?
- What should a lead buyer do differently this season?
If you buy health insurance calls or leads, your Q4 plan is probably built around a December 15 close. That date is out of date, and the correction is worth 31 days. The calendar is also the least interesting thing that changed this year.
Federal open enrollment for 2027 ends January 15, not December 15
Open enrollment on HealthCare.gov runs from November 1, 2026 through January 15, 2027. CMS states it plainly in a document dated July 31, 2026: “Open Enrollment at the Federally-facilitated Exchange will begin on November 1, 2026, and end on January 15, 2027” (CMS).
The healthcare.gov dates page lists the same milestones. November 1 opens. December 15 is the deadline for coverage starting January 1, and January 15 is the last day to enroll.
A shorter season was finalized and then set aside. A 2025 rule would have closed the federal window on December 15 from the 2027 plan year onward. A court vacated that provision in June 2026, and the CMS statement above confirms where the dates landed. That is as much history as a media plan needs.
The arithmetic is worth doing once. November 1 to December 15 is 45 days. November 1 to January 15 is 76 days. The difference is 31 days, and they are not average days. They cover the run to the final deadline, when late deciders convert.
The stale date is still the common answer, and that is the real risk. Search the question today and you get broker pages, agent help centers and enrollment-platform FAQs still printing December 15 for 2027. If your plan inherited its dates from one of those, the last four weeks of the season are budgeted at zero right now.
Most plans will not surface that until January, by which point spend has already stopped and deadline traffic is still arriving.
A smaller pool, and more of it has to be sold
Sign-ups fell. Per CMS, 23.1 million people selected or were automatically re-enrolled in a marketplace plan during the 2026 open enrollment period. Every figure in the table below is on-exchange ACA, from that CMS report or from KFF’s analysis of it.
| Measure (on-exchange ACA) | 2026 season | Change vs 2025 | Source |
|---|---|---|---|
| Plan selections | 23.1 million | down 1.2 million, a 5% decline | CMS |
| Effectuated enrollment (February) | 19.2 million | down from 21.8 million | KFF |
| Effectuation rate | 83% | down from 90% | KFF |
| Automatic re-enrollees | 8.8 million | down 19% | CMS |
| New consumers | 3.6 million | down 13% | CMS |
| Agent or broker assisted share of active HealthCare.gov selections | 76% | up from 71% | CMS |
Plan selections are not covered lives, and the gap widened. Effectuated enrollment counts the people who actually paid and held the coverage, and it sits roughly four million below the selections headline. The effectuation rate slipped by seven points in a single year. Anyone quoting the selections number as the count of people with ACA coverage is overstating it.
The renewal engine is what slowed. Automatic re-enrollment fell hardest of any line in that table, and it is the part of this market that costs nobody an acquisition dollar. When two million of those stop happening by themselves, the people do not disappear. They become shopping events.
A rising share of that shopping runs through a person. Per CMS, the agent or broker assisted share of active plan selections on the federal exchange reached 76% in the 2026 season, up from 71%. That covers the 30 states on the federal platform rather than the whole country, and active selections are people who made a choice instead of being rolled over.
So the pool shrank, the automatic part shrank faster, and the share of what is left that needs a human went up. Brokered enrollment is not a proxy for call demand. It is call demand.
The cause matters, because it decides whether the decline repeats. ASPE, the federal health department’s research office, attributes a large share to clearing out enrollments that should not have been on the books, which is a one-time correction. KFF points to the expiry of the enhanced premium tax credits at the end of 2025 (CMS, footnote 10), which is a standing price increase.
The two agencies weight them differently, and nobody has cleanly separated them. If the second is doing more of the work, 2027 contracts again, and this past season is the high-water mark to pace against.
Carrier exits decide which states the volume moves in
Treat a carrier exit as a demand event rather than a news item. A member whose plan disappears cannot auto-renew, so they land back in the shopping pool and usually need a person to place them. That is the mechanism behind Medicare plan exits in 2027, and it behaves the same way here.
Cigna is leaving the individual market entirely. It will exit all 11 states where it participates, on and off exchange. It reported first-quarter on-exchange enrollment of over 350,000 people (KFF).
Molina is shrinking rather than leaving, concentrating on roughly six marketplace states against the 13 or 14 it runs in now (Becker’s). Other carriers have announced state-level exits. Check KFF’s tracker for the live list, because outlets are counting differently and the totals do not agree.
Do not buy a national number here. Participation is set county by county, so a headline count of exits tells you almost nothing about your own footprint. Check which of the states you already buy in lost an issuer for 2027. Those counties are where displaced members surface as inbound calls, and they are the ones worth bidding up.
The three markets, and how well each one is counted
Under-65 health is three separate markets, and every figure above belongs to only one of them: on-exchange ACA. That is where most coverage of the season goes wrong. A writer pulls a marketplace enrollment number, and a paragraph later it is standing in for private off-exchange volume it never described. We keep the difference between a U65 lead and an ACA lead sharp for the same reason.
| On-exchange ACA | Off-exchange ACA-compliant | U65 off-exchange, non-ACA | |
|---|---|---|---|
| When demand exists | Nov 1 to Jan 15, plus qualifying life events | Same plan-year logic, no subsidy | All year, no enrollment window |
| What moved into 2027 | 23.1M selections, down 5%; 19.2M effectuated (CMS, KFF) | Fell from 3.5M in 2019 to 2.2M in 2024 (ASPE) | 2024 rule unenforced since Aug 2025; sold in 36 states, prohibited in 5 (KFF) |
| How well it is counted | Monthly, to the individual | Trend published, no current level | Not counted at all (KFF) |
| What that means for pacing | Plan precisely, compete hardest | Do not size it from ACA figures | Steady baseline, unverifiable ceiling |
| Who it suits, and how to hedge | Buyers who can concentrate spend into a fixed window and want the deepest pool | Rarely a standalone buy; it moves with on-exchange pricing | Buyers who need volume between seasons and can price without a published benchmark |
The three markets are not counted to the same standard. On short-term and other non-ACA under-65 coverage, KFF puts it flatly: “There is no current or comprehensive data on the number of consumers enrolled in an STLD” (KFF), meaning a short-term, limited-duration plan. The last hard count anyone cites is from 2022. It excluded policies sold through associations, which the federal rulemaking says is where most of them are sold.
So the year-round market you would naturally hedge into is the one nobody can size. The absence of a number is a data problem rather than a demand problem. Your own delivered volume is the only reliable read on off-exchange U65 buying, so use your historical fill rate as the planning input and stop waiting for a market report to arrive.
What is still genuinely unsettled
The state exchanges have not converged, and the spread is six weeks wide. January 15 is the HealthCare.gov date. States running their own exchanges set their own:
| State exchange | Published 2027 open enrollment window |
|---|---|
| New Jersey | Nov 1 to Jan 31 |
| Minnesota | Nov 1 to Jan 15 |
| Colorado | Nov 1 to Jan 15 |
| Washington | Nov 1 to Dec 31 |
| Connecticut | Oct 23 to Dec 23 |
| Vermont | Nov 1 to Dec 15 |
Connecticut is the one to notice. It published in July 2026, after the federal window was restored, and still chose a December 23 close. Several states have published nothing. Massachusetts has published a start date and no end date. If you buy by state, inherit nothing from the federal calendar.
The longer federal window is not guaranteed to survive the season. The decision that restored it is under appeal. The government has asked for a ruling by December 31, 2026, which falls inside the window itself. The dates are operative today and nothing is paused. It is still the wrong year to sign a non-cancellable January commitment.
Premiums are proposed, not final. Insurers have proposed a median increase of 15% for 2027 across 276 insurers (KFF, updated August 3, 2026). Last year, according to the same KFF analysis, a proposed median of 18% finalized at 20%. The revision tends to run one way, so watch it before setting a bid ceiling.
Movement between the three markets is modeled, not observed. Projections exist for how many people shift to other coverage when subsidies lapse. Nobody is measuring where they land. Treat any claim that a specific number “moved to short-term plans” as an estimate wearing a fact’s clothing.
Frequently Asked Questions
When does ACA open enrollment for 2027 actually end?
January 15, 2027 on HealthCare.gov, confirmed by CMS in a statement dated July 31, 2026. December 15 still matters, but only as the cutoff for coverage beginning January 1. Enrolling between December 16 and January 15 gives coverage starting February 1. States running their own exchanges set their own dates.
Why do so many sites still say December 15?
A 2025 rule would have ended the federal window on December 15 from the 2027 plan year. Many pages were written while that was the expected outcome. A court vacated the provision in June 2026, before it took effect. Those pages were not wrong when published and most have never been revisited, which is how the old date keeps reaching media plans.
Does the January 15 date apply in every state?
No, and assuming it does is the expensive version of this mistake. January 15 is the HealthCare.gov date. Published 2027 end dates currently range from December 15 in Vermont to January 31 in New Jersey. Connecticut set a December 23 close after the federal window was restored, so a later federal date does not imply a later state one.
Does the headline enrollment number mean people covered?
No. Per CMS, 23.1 million is plan selections, counting everyone who picked a plan or was rolled into one. According to KFF, effectuated enrollment, meaning people who paid and held coverage, was 19.2 million in February 2026. The gap is roughly four million people, and according to KFF the effectuation rate fell from 90% to 83% over the same year.
Does open enrollment change U65 off-exchange volume?
Not directly, and this is the distinction most coverage misses. Non-ACA under-65 products such as short-term and fixed indemnity plans have no enrollment window and sell year round. The season changes attention, not eligibility. There is no published data on how many shoppers land in a non-ACA plan, so treat any specific figure with suspicion.
What should a lead buyer do differently this season?
Extend the flight to January 15 rather than December 15, and treat December 16 to January 15 as its own offer, since coverage bought then starts February 1. Re-check carrier participation in the states you actually buy in. And price the season against brokered enrollment rather than total enrollment, because the brokered share is the part that is growing.
Where this leaves your plan
You will plan this season against on-exchange ACA, because it is the market that publishes the best data. That is a good reason to trust the numbers, and not a reason to weight the budget there. On-exchange gets counted closely because it is federally administered, not because it is where the most winnable demand sits.
Three moves follow from all of this. Extend the flight to January 15, and treat December 16 to January 15 as its own offer, because coverage bought in that stretch starts February 1. Re-check carrier participation in the states you actually buy in, then bid up the counties that lost an issuer. And price the season against brokered enrollment rather than total enrollment, because the brokered share is the part that is growing.
Weight the plan by where the buying happens, not by where the counting happens. If you want a second read on how that splits across your states, our U65 private health lead generation team can look at it with you.
