In Defense of the Portal

Everybody who writes about points will tell you the same thing. Do not redeem through your bank's travel portal. Transfer to a partner. The portal is what they give you so you stop asking questions.

For flights, that is correct and I follow it.

For hotels it is mostly wrong, and I can show you why.

Two sets of numbers do the work here, and they are not both mine. The transfer ratios are ours, checked against multiple trade sources and dated. The point valuations are The Points Guy's, used as a starting default because a public, documented baseline beats one I invented. What is mine is the arithmetic that puts the two together, which as far as I can tell nobody else publishes.

Chase changed how portal redemptions work in 2025, so the comparison is no longer a single number. It is a range, and the conclusion holds across the whole of it. Of the twenty-two bank-to-hotel transfers we track, thirteen return less than the portal's one cent floor. At its two cent ceiling, all twenty-two of them lose.

Why hotel transfers are so bad

A transfer ratio and a valuation have to be read together, and almost nobody does it. Amex transfers to Hilton at 1:2, which sounds like a gift. Two Hilton points for every Amex point. Who would turn that down?

A Hilton point is worth about four tenths of a cent. So two of them are worth eight tenths of a cent, and you just turned an Amex point into less than a penny.

It does not improve much from there. Marriott is three quarters of a cent at 1:1. IHG is six tenths. Wyndham is seven tenths. Bilt into Hilton at 1:1 is four tenths, which is the worst transfer on the site by a distance.

Meanwhile a Chase point spent in the portal is worth somewhere between one and two cents depending on the booking, and asks nothing of you but a click. The unglamorous option beats the clever one most of the time.

What Chase changed, and why the number is a range now

For years a Chase point had one price in the portal. 1.5 cents on a Sapphire Reserve, 1.25 on a Sapphire Preferred, and that was that, on every hotel, every flight, every car.

In June 2025 Chase replaced that with Points Boost, and the shape of the deal inverted. The default is now one cent a point on every card. On top of that, Chase tags a rotating selection of hotels and flights with a boosted rate, up to 1.5 cents on the Preferred tier and up to 2 cents on the Reserve tier. No tag, no boost, one cent.

Read that carefully, because it is the whole change in one line. The old system had a floor. The new one has a ceiling. There is no published minimum above a cent, and the boosted values people report finding have already drifted down since launch. The Edit, Chase's luxury hotel collection, launched with a guaranteed 2x and quietly became "up to 2x" in December 2025.

The old fixed rate is not gone, though, and this is the part most write-ups get wrong. If you held an eligible card before Points Boost launched, the points you earned up to late October 2025 keep the legacy 1.5 or 1.25 cents until late October 2027, and Chase gives you whichever is better. A lot of people reading this are still sitting on a grandfathered balance and have not noticed anything change yet.

I have done better than 1.25 cents with a boost. I have also opened the portal and found nothing boosted at all. That is the honest description: it is a lottery with a known floor, not a rate. And it is why every figure on this page is dated, because this feature has been devalued twice in fifteen months and neither time was announced.

Every hotel transfer worth making

This is the complete list of bank-to-hotel transfers that clear 1.25 cents, which is roughly the middle of the portal's current range and was its old fixed rate. Six of them, out of twenty-two. Raise the bar to a boosted two cent redemption and the list is empty.

World of Hyatt is the one everybody knows about, and it earns it. Against the chains most people actually stay at, it is not a close contest. A point sent to Hyatt returns more than twice what the same point returns at Marriott, more than twice what it returns at Hilton, and nearly three times what it returns at IHG. It is the only mainstream hotel program where transferring is clearly the right move.

The reason is the award chart. A published chart means a property selling for six hundred dollars can cost the same points as one selling for two hundred, and that mismatch is where outsized redemptions come from. Hyatt is the last big program that still works this way, and it is why Hyatt is the first thing I check and usually the last.

It got more expensive in 2026 and I am not going to pretend otherwise. On May 20 Hyatt went from three pricing bands to five, moved more than a hundred properties up a category and twenty-four down, and stretched the top of the range hard: a category 8 peak night went from 45,000 points to 75,000. Hyatt says it remains committed to publishing a chart, and it does still publish one, but the bands are wide enough now to behave a lot more like dynamic pricing than they used to.

It is still comfortably the best hotel transfer partner there is. The most telling detail is that the 1.65 cent valuation in the table above was set in September 2026, after the devaluation, and it went up rather than down. A program can get more expensive and still be the best thing on the board by a distance, and that is exactly where Hyatt sits. Check the award price against the cash price on your specific stay, which was always the right advice, and go.

The Citi line carries a catch worth knowing. That 1:1.5 is the rate on the premium Citi cards. Points earned on a no-annual-fee Citi card transfer at seven tenths of the listed figures, which drops the same move to about nine tenths of a cent and puts it back behind the portal.

And I will be honest about the top line, because I did not know it until I ran these numbers. Wells Fargo into Choice at 1:2 comes out ahead of Chase into Hyatt, and I have never made that transfer in my life.

The other way to use a portal: pay cash, earn points

Everything above is about spending points through a portal. There is a second move that runs the other direction, and it is the one I use most often.

Book the hotel through the portal with cash, and collect the elevated earning rate. Several cards pay far more on travel booked through their own portal than on travel booked anywhere else, and the best of them are paying rates you will not find in any other category.

These are the standouts. The earning rates come from the issuers' own pages and terms. The last column converts them using The Points Guy's valuation for that currency, which is what tells you whether the multiplier is actually worth anything.

Nineteen percent back on a hotel stay you were paying cash for anyway is not a rounding error. It is better than any category bonus I have on any card, and it is available on a ninety-five dollar card.

Below those, 5x is the common floor. Chase Sapphire Preferred, the Freedom cards, the no-fee Citi Strata, Citi Double Cash, and most of the Capital One lineup all sit there. Do not assume your card is one of the good ones. Capital One pays 10x on the Venture X and 5x on nearly everything else, which is the trap in reading a headline rate and thinking it applies to the whole issuer.

Two things surprised me when I went and checked rather than trusting what I remembered:

• The Sapphire Reserve was cut from 10x to 8x during 2025. New applicants from June, existing cardholders on renewal from late October. If you still think it is 10x, it is not.
• Amex Gold went from 2x to 5x on prepaid portal hotels in April 2026, with no fee increase. On that card hotels now out-earn flights, which are 3x. That is backwards from every other card here.

Two issuers are worth knowing about for the opposite reason. Every Amex portal rate is prepaid only, so pay-at-hotel earns you nothing extra. And Wells Fargo's portal is not a bonus channel at all, because Wells Fargo defines its categories by merchant code rather than by booking channel, and a portal booking codes as a travel agency rather than a hotel.

The fine print that decides whether it works

These come from the issuers' own terms, and the same rules turn up at Chase and at Citi independently, which tells you it is how the model is built rather than one bank being difficult.

• Compare the portal price against booking direct, first, every time. Ten times points on a rate twenty percent higher is not ten times anything. This is the step people skip and it is the one that decides the whole question.
• The portion covered by a hotel credit earns nothing. Chase and Citi both say so outright. So you cannot stack your card's hotel credit against the multiplier and collect on both, which is exactly the move you would reach for.
• A points-and-cash booking only earns on the cash part.
• Cancel and the points reverse. Package rates, air plus hotel, usually do not qualify at all.
• You still give up elite credit. A portal booking is a third-party booking whether you paid cash or points. No elite nights, usually no base hotel points, and benefits at the property's discretion.

That last one is the real cost, and it is why none of this is a blanket recommendation. If you are chasing status at a chain, the elite nights are worth more than nineteen percent back.

And there is one more, which does not show up in any terms document and which I learned by standing outside a locked door.

The exception: a portal that does not cost you elite credit

There is one that breaks the rule, and it is worth knowing about precisely because it breaks it.

Rove is not a credit card. There is no co-branded card at the middle of it. It is a booking platform with its own transferable currency, and you pay with whatever card you already carry. It offers two rates on the same room: a standard rate that pays more of its own miles and forfeits your hotel loyalty, and a loyalty-eligible rate that pays fewer miles but leaves your hotel points, your elite nights and your status benefits intact.

Take the second one and the charge comes from the hotel rather than from the platform, so your credit card's own hotel category still fires. Hotel points, elite nights, a transferable currency, and your card's earning, on one stay. That is four streams where a bank portal gives you one and takes the elite credit away.

I am being deliberately vague about the rate because Rove will not commit to one. Its own wording is "up to 25x miles per dollar," which floats by property and by demand, and I am not going to print a number the company itself will not stand behind. Treat it as worth pricing against the portals on the specific stay you are booking, rather than as a fixed figure you can plan around.

What Rove miles reach, and what they are worth →

Flights are the opposite, and it is not close

None of this transfers to airline miles, and I want to be clear about that because the two get discussed as one subject.

Airline award charts still contain genuine mispricings. A business class seat that sells for six thousand dollars can cost the same miles as one that sells for two. A portal, by design, ignores all of that and pays you a flat rate against the cash price. Every bit of the upside is the part the portal throws away.

I have made this mistake myself and paid for it. On a trip in 2024 I booked three seats home through a portal at exactly 1.00 cent a mile because it was the button I understood at the time. That trip is written up here, hotels included, and the flight is the one redemption in it I would take back.

Four ways this argument can mislead you

I would rather hand you the objections than let you find them later.

• Cents per point is a ratio, not an amount. A superb rate on an eighty dollar room is still an eighty dollar room. Choice and Wyndham score well here partly because their properties are cheap, and a big number over a small number is still a small number.
• Portal bookings usually do not earn hotel status. Most chains treat a portal stay as a third-party booking, which means no elite night credit, no base points, and sometimes no elite benefits when you arrive. If you are chasing status this whole argument changes, and it is the strongest case against my approach.
• Low-cent valuations are the least reliable ones. The defaults below are borrowed, and on a program worth well under a cent a tenth either way swings the conclusion. Treat the Choice and Wyndham lines with more suspicion than the Hyatt one.
• A great ratio at a hotel you do not want is not a good redemption. The arithmetic cannot tell you whether you want to stay somewhere. Availability where you are actually going beats every number on this page.

Where these numbers come from

The transfer ratios are ours, re-verified against multiple trade sources and dated on the page. The full partner list is here.

The valuations are not ours. They are The Points Guy's published figures, used as a starting default because a public, documented baseline beats one we invented. Change them to what you actually achieve and the conclusions move with you. Every rate we use is listed here, and the methodology page shows the arithmetic.

The one thing I would ask you to take from this: run the ratio and the valuation together before you transfer anything, because the ratio on its own is designed to look generous. The transfer calculator does that arithmetic for you.