Costco Is Selling Medicare. The Part That Should Change How You Work.
A warehouse club can put a plan in front of five million people. It cannot tell one of them whether their cardiologist is in network. That gap is your job — and this is how you do it well enough that it becomes obvious.
For independent agents & agency owners · ~9 min read · Industry Update
On August 18, 2026, Costco and SCAN Health Plan announced a partnership to sell jointly branded Medicare Advantage plans. The reaction in agent circles was predictable and mostly wrong. Half the posts said agents are finished. The other half said it was nothing.
Both miss what the announcement actually says. So let us go through what was confirmed, what was not, and then spend most of our time on the part that matters: the specific work that makes an agent worth more than a display rack.
What Was Actually Announced
These come from the SCAN Health Plan release and Wall Street Journal reporting dated August 18, 2026.
| Item | Detail | Why it matters to you |
|---|---|---|
| Structure | Jointly branded Medicare Advantage in two states, plus a Medicare Supplement plan in a third | Narrow at launch. Most books are unaffected in year one. |
| Market size | Executives say those markets hold roughly 5 million Medicare enrollees | Small footprint, large population. This is a test with real scale behind it. |
| The carrier | SCAN Health Plan — about 460,000 members across CA, AZ, NV, TX, NM, WA | Not a startup. An established nonprofit MA carrier. |
| Possible benefits | Revamped pharmacy experience, Medflex OTC benefits, vision, audiology | Benefit-rich positioning aimed squarely at the ancillary conversation. |
| Membership | No Costco membership required | Federal rules prohibit tying Medicare coverage to a retail membership. |
| Approval | Pending federal regulatory approval | Nothing is final. Do not market it. |
What Has Not Been Confirmed
This is where agents are going to get themselves in trouble over the next few weeks, so be precise:
- The states have not been named. Executives declined to identify them, citing disclosure restrictions while approval is pending.
- There is no launch date. The release says “in the coming years.” Anyone telling you 2027 is guessing.
- Final benefit design is not set. Everything listed is described as possible, pending approval.
The Line To Read Twice
Buried in the coverage is the sentence that actually determines whether this is a threat or an opportunity for you. The plans will be sold in Costco stores, online, and through insurance agents.
Agents are in the channel. In the announced model, this is not a direct-to-consumer play that routes around distribution — it adds retail and digital alongside it.
That is the good news. Here is the uncomfortable part: being in the channel is not the same as being necessary to it. If your value proposition is “I can hand you a brochure and take your application,” a warehouse aisle and a website do that faster and at a lower cost than you do. The channel will keep whoever adds something the other two options cannot.
Why Brand Trust Is the Real Threat
The competitive shift here is not about benefits. It is about the question a prospect asks themselves.
For years the question has been “which carrier should I pick?” — a question most seniors feel unqualified to answer, which is exactly why they call an agent. When a brand they already trust from twenty years of grocery runs appears on a plan card, the question quietly changes to “which name do I recognize?”
That is a question they feel perfectly qualified to answer without you.
You do not beat brand recognition with more brand recognition. You will never out-trust Costco on trust. You beat it on specificity — by being the only party in the transaction who knows this particular person’s medications, prescribers, and pharmacy, and who has checked them against the actual plan documents for the actual plan year.
Agency Engineer Playbook
Reframe your role out loud. “My job is not to sell you a plan. It is to make sure whatever plan you end up in actually covers your doctors and your prescriptions next year — including the ones that changed.” That sentence positions you as verification, not sales. Verification is the thing a store shelf structurally cannot provide.
The Five-Step Verification Protocol
This is the tactical core. Run it on every Medicare client, every year. It is not complicated. It is just work most agents skip, which is precisely why doing it is a moat.
1. Build the drug list properly, not approximately
“She takes a blood pressure pill and something for cholesterol” is not a drug list. It is a liability. What you need on every record:
- Exact drug name, and whether it is brand or generic
- Dosage and strength
- Quantity and days supply per fill
- The pharmacy they actually use — the specific location, not the chain
The fastest reliable method is to ask the client to read the labels to you, or to photograph the bottles and text them over. Do not work from memory — theirs or yours. A 20 mg versus 40 mg difference can move a drug across a tier boundary, and tier is what determines cost.
2. Verify every prescriber and facility for the plan year in question
Network status is not a permanent attribute. A physician in network in 2026 may not be in network in 2027, and provider directories are updated on the carrier’s schedule, not yours.
Capture the primary care physician, every specialist seen in the last twelve months, and the preferred hospital system. Then verify each against the plan’s directory for the year the coverage will be effective. When it matters — a specialist mid-treatment, an oncologist, a surgeon with a procedure scheduled — confirm directly with the provider’s office rather than relying solely on the directory, and note the date you confirmed and who you spoke with.
3. Price the year, not the premium
The premium is the number clients fixate on and the least useful one in the comparison. Total annual cost is what determines whether they are better off. Model it:
- Annual premium
- Medical and drug deductibles
- Expected copays and coinsurance at their actual utilization
- Drug costs by tier across the full plan year
- Maximum out-of-pocket — the number that defines their worst realistic year
A plan that is $30 a month cheaper and puts one maintenance drug on a higher tier can cost more by March. Showing a client that arithmetic is the single most persuasive thing you will do all season, and no display rack can do it.
4. Check pharmacy network tier, not just formulary status
This is the most commonly missed step in the business. A drug can be on the formulary at a reasonable tier and still cost meaningfully more because the client fills it at a standard rather than preferred pharmacy. Preferred status changes year to year, and it changes by location.
Check where they actually fill, check that pharmacy’s status for the coming plan year, and if there is a cheaper preferred option nearby, tell them what the difference is and let them decide. Mail order is worth pricing on maintenance medications, though it is not automatically better for everyone.
5. Document what you checked, when, and against what
Every verification gets a date, a source, and a note in the CRM. This does three things: it protects you if a client’s recollection differs from what happened, it makes next year’s review dramatically faster, and it turns your book into an asset with real institutional knowledge instead of a list of names.
The Conversation That Beats a Brochure
The Agency Engineer Medicare script already collects the right inputs in Stage 2, Pre-Appointment Discovery: zip code, Medicare effective date, primary doctors, prescription list, and current plan. That is not paperwork. Those five fields are the raw material for everything above.
What changes in a world where a trusted retail brand is also in the market is how you frame the value of collecting them. Try the contrast directly and without disparaging anyone:
Notice what that does. It does not attack the competition — which would be both bad practice and a compliance problem. It describes a service the alternative structurally cannot provide, and it ends on a low-friction ask.
Then deliver. Come back with the actual findings: two drugs on tier 2, one moved to tier 3, cardiologist in network, the pharmacy they use is standard but there is a preferred one a mile away, total modeled annual cost is X versus Y. That is a fundamentally different experience from being handed a pamphlet, and clients know it immediately.
What the Last Retail Wave Got Wrong, and Why This One Is Different
Some perspective, because the panic is not proportional to the history.
Walmart shut down all 51 of its health centers along with its virtual care service. Walgreens closed roughly 160 VillageMD clinics — about half its footprint. Retail scale has repeatedly failed to translate into healthcare delivery, because delivering care requires clinicians, and clinicians are expensive, scarce, and do not scale like inventory.
But be honest about the distinction: Costco is not delivering care here. It is lending brand and retail reach to an established licensed carrier that already runs the plans. That is a materially lighter lift than staffing clinics, and it is a reasonable bet that this model survives where the clinic model did not.
So do not dismiss it. Plan for it. The correct posture is neither panic nor denial — it is raising your floor so that when a client compares you to a shelf, the comparison is not close.
The Rest of the Stack
The protocol above is the part you can implement this week on your own. The systems that make it repeatable across a whole book during a 54-day enrollment window are what our contracted agents run:
- The complete Medicare and Life sales scripts — every stage written out, discovery frameworks, and both Top-10 objection lists
- CRM build and automation templates — the actual fields, sequences, and triggers that make five-step verification a workflow instead of a good intention
- Done-for-you presentation decks for the Zoom appointment
- Ad and lead-flow systems that keep the calendar full while you are doing the deep work
- The full training library and agent resource hub
Those come with the platform. We do not sell them separately and we do not give them away — they are what contracting with Agency Engineer gets you, along with top comp and keeping your own book.
Built by Agents. Engineered for Freedom.
The anti-corporate FMO: independence, AI tools, ad strategies, done-for-you presentations, a full training library, and top comp — while you keep your book. If retail entrants are going to compete on brand, compete on expertise. We will give you the systems to do it at scale.
FAQ
Should I be telling clients about the Costco plans right now?
Only factually, and only if they ask. A partnership has been announced; the states, timing, and final benefits are pending federal approval. You cannot market an unapproved plan, and you should not describe benefits that are not final.
Does this mean agents are getting cut out of Medicare distribution?
Not in the announced model — reporting states the plans will be sold in stores, online, and through insurance agents. The longer-term risk is not exclusion from the channel; it is being interchangeable within it.
My book is not in the SCAN states. Does this matter to me?
Not operationally this year. Strategically, yes — if a co-branded retail model works in a few markets, it gets copied. The verification discipline that protects you is worth building before it reaches your market, not after.
What is the single highest-value change I can make this week?
Complete drug lists on every Medicare client record — exact name, dose, quantity, and the specific pharmacy. Everything else in the protocol depends on it, and most agencies are missing it on a large share of the book.
Is this just fear-mongering about a partnership that may not launch for years?
The opposite. The tactical work above is worth doing whether or not this particular partnership ever reaches your market. That is the point — it is not a reaction, it is a standard.
