The Booking Window: Reading Lead Time to Decide When to Discount and When to Hold

Most independent hoteliers I meet are sitting on a forecasting tool they never open. It is not a fancy revenue-management system. It is the booking window: the gap between when a guest books and when they show up. Read it right and it tells you, weeks ahead, whether a date is going to be strong or soft, whether you should hold your rate or sweeten an offer, and which guests to talk to first. Read it wrong, or not at all, and you do what almost everyone does. You panic-discount too early, hand margin to the OTAs, and call it “the market being slow.”

Let me walk you through how I actually read this stuff for the hotels I work with, because it is less intimidating than the revenue-management priesthood makes it sound.

What the booking window actually is

Every reservation has a lead time: book on June 1, arrive June 20, that is a 19-day lead time. Stack up all the reservations for a given arrival date and you get a shape, a curve, that shows how bookings accumulate as the date approaches. Average that across a season or a segment and you have your booking curve.

The curve answers one question that matters more than almost any other in this business: on any given day, am I where I should be? If on the 45-days-out mark a typical October Saturday is usually 40 percent booked, and this year’s October Saturday is sitting at 55 percent, you are pacing ahead. That is not a date to discount. That is a date to hold, or quietly push rate.

The reverse is the trap. A date pacing behind makes you nervous, and nervous hoteliers reach for the discount lever way too soon, before the bulk of that date’s demand has even shown up.

Why averaging everything together hides the answer

Here is the first mistake I see: hoteliers look at one blended booking curve for the whole property and treat it as gospel. The problem is that a single average is a smoothie of wildly different behaviors.

Think about who actually books your rooms:

• Direct and returning guests tend to book the furthest out. They know your property, they planned the trip, they came to your site on purpose.
• OTA bookers skew shorter and more price-reactive. Many are comparison-shopping in the last couple of weeks.
• Last-minute mobile is its own animal, often inside seven days, frequently same-day.
• Group, wedding, and event blocks have a long, lumpy curve that looks nothing like transient demand.

Blend those together and the average lies to you. A date might look “behind” only because your long-lead direct segment is light this year, while your last-minute segment, which always fills the final two weeks, has not even started. If you discount on day 45 to fix a “problem” that the last-minute segment was going to solve anyway, you just gave away rate for nothing.

So the first real move is to segment the curve. At minimum, split it three ways: direct, OTA, and everything-else (group, corporate, wholesale). You do not need a data scientist. You need your PMS to export reservations with a booking date, an arrival date, and a channel tag, and you need a spreadsheet.

Building your curve without a fancy system

Pull a year of reservation-level data. For each booking, compute lead time = arrival date minus booking date. Then bucket arrivals by something meaningful (month, or day-of-week, or your own season labels) and by channel. For each bucket, look at how cumulative bookings build across the days-out timeline.

What you are building is a reference: “for a typical shoulder-season Tuesday, here is the percent of final bookings we usually have in hand at 60, 45, 30, 14, and 7 days out.” That reference is your pace benchmark. Everything else is comparing today’s live pace against it.

Here is an illustrative version of what one of these tables looks like once you have it. These numbers are made up to show the shape, not a benchmark to copy:

Notice how different those two columns are. The peak Saturday is essentially “sold by 14 days out, hold your rate.” The shoulder Tuesday does most of its work inside the last two weeks, so a soft reading at 30 days out is normal, not an emergency. Same property, opposite playbooks. One curve would have hidden both.

When to hold

You hold rate when the data says demand is coming and you have time for it to arrive. Concretely:

• You are pacing ahead of your benchmark at the same days-out. The date is doing better than a typical year. Discounting here is lighting money on fire.
• The date sits inside a segment that historically fills late. Your shoulder Tuesday is supposed to look empty at 30 days. Trust the curve.
• Your compression dates are near. If a citywide event, festival, or peak weekend is approaching, last-minute and overflow demand props up the surrounding nights. Hold and watch.
• OTA visibility is doing the heavy lifting you do not want it to. If a soft date is “filling” only because OTAs are quietly discounting your inventory through their own promo programs, the fix is not for you to cut rate too. It is to capture more of that demand direct so you keep the margin. This is exactly the dynamic I broke down in how OTAs steal your search traffic.

Holding is the harder discipline because doing nothing feels like negligence. But the booking window gives you permission to do nothing when nothing is the right call.

When to discount (or really, when to act)

You act when the curve tells you demand genuinely is not going to arrive in time at the current rate. The honest signals:

1. You are persistently behind your benchmark across several check-ins, not just one nervous Monday morning glance.
2. You are inside the segment’s natural fill window and still short. A shoulder Tuesday that is light at 30 days is fine. The same Tuesday light at 7 days, when it is normally 62 percent booked, is a real gap.
3. There is no compression event coming to bail you out.
4. Demand signals upstream are weak too. Site traffic, search interest, and metasearch impressions for those dates are down, not just bookings.

And even then, “act” does not have to mean “slash the public rate.” A blanket discount is the bluntest, most margin-destructive tool available. Better moves, roughly in order of how much I like them:

• Add value instead of cutting price. A package with parking, breakfast, a late checkout, or a local experience holds your headline rate while still nudging the fence-sitter. It also reads better in search and on your booking engine.
• Target the offer to a segment. Email your past guests and loyalty list a member rate for the soft dates. That is a direct, high-margin booking you control.
• Use length-of-stay or day-of-week tactics to move demand onto the weak nights instead of discounting the whole stay.
• Open or widen availability to channels you had restricted, before you touch rate at all.
• Only then, if the gap is real and close, consider a genuine rate move, and ideally one your direct channel can match or beat so you are not just feeding the OTA machine.

The part nobody connects: lead time and search visibility

Here is where my world, SEO and AI search, plugs directly into your revenue strategy, and why I care about your booking curve at all.

Your longest-lead-time guests, the high-intent, high-margin direct bookers, are researching weeks before they book. They are typing your town plus “boutique hotel” into Google. Increasingly they are asking ChatGPT and other AI assistants “where should I stay in [your town] for an anniversary weekend?” That research happens at the wide, early part of the booking window, exactly the demand you most want to capture direct.

If you are invisible during that window, one of two things happens. Either the guest never finds you, or they find you through an OTA that did show up in that early search, and now you are paying 15 to 25 percent commission on a guest who was looking for a hotel like yours anyway. The booking curve told you the demand was out there. Your search presence decided whether you captured it or rented it back from Booking and Expedia.

This is why I think about pace and visibility together:

• Long-lead direct demand is won with strong organic hotel SEO and AI-search visibility so you are in the consideration set when planning starts. The search volume is real and growing, “aeo” alone runs about 27,100 US searches a month, “generative engine optimization” around 5,400. People are learning how to be found in AI answers. Your guests are using those AI answers to plan trips. If you are not in the answer, you are not in the trip. I dug into this in is your hotel invisible to ChatGPT.
• Mid-window comparison demand is where your Google Business Profile and metasearch presence decide whether the direct rate even gets seen next to the OTA rate.
• Short-window, ready-to-book demand is won or lost on your booking engine and direct-booking experience. When someone has already decided, friction at checkout sends them back to the OTA tab.

When your search visibility is healthy across all three, your direct booking curve starts shifting earlier and fatter, which is exactly what gives you the confidence to hold rate. Weak visibility compresses your direct demand toward the last minute, which is precisely when you feel most pressure to discount. The two problems feed each other.

A simple weekly rhythm

You do not need to live in spreadsheets. Here is the cadence I recommend to the independents I work with:

• Once a week, compare live pace against your benchmark for the next 60 to 90 days, by segment, not blended.
• Flag only the dates that are genuinely behind inside their natural fill window. Ignore the noise of dates that are supposed to look empty right now.
• For flagged dates, check upstream demand (site traffic, metasearch impressions, search interest) before you decide. Weak bookings plus weak upstream demand is a real gap. Weak bookings plus healthy upstream demand often just means the conversion or visibility piece needs work, not the rate.
• Act with the least blunt tool that closes the gap. Value-add and targeted segment offers before public rate cuts.

The goal is not to never discount. The goal is to stop discounting reflexively, out of fear, on dates that were always going to fill late, and to redirect that energy toward capturing more high-margin direct demand earlier in the window.

Do this for a season and two things change. Your average rate stops leaking through premature discounts, and you start to see, with a few weeks of warning, where the real soft spots are, so you can fix them with marketing instead of markdowns.

None of this is about beating the OTAs or pretending you can fire them. They are part of a healthy mix and a useful demand source. It is about understanding your own demand well enough that you stop handing them the bookings you could have kept, and stop discounting the ones you were always going to get.

Want help reading your curve?

If you want a second set of eyes on your booking window, and on whether your search and AI visibility are letting you capture demand early enough to hold your rate, that is exactly the kind of thing I do. Book a free intro call and bring a year of reservation data. We will pull your curve apart by segment and find the dates where you have been leaving money on the table. If you would rather start with the visibility side, my hotel SEO service is built to get you into that early research window where the high-margin direct bookings live.