Dynamic pricing at a daily-fee course means moving green fee rates with demand instead of publishing one rate card per season, and the honest version of it is narrower than the version in most vendor decks. The strongest published research on golf pricing found that demand for golf is largely price inelastic, which means the discounting half of the discipline, the half these systems are usually sold on, is the half least likely to make you money. The upside sits on the upward side of the curve, and in the inventory you are currently losing to no-shows and to peak times you have never tested.
What dynamic pricing actually means on a tee sheet
Four separate decisions get bundled under the one term, and they are not equally easy or equally valuable.
Rate movement by time of day and day of week is the layer nearly every course already runs by hand, in the form of a seasonal card with a twilight rate at the bottom. Software makes it finer rather than different in kind.
Rate movement by how full the slot already is comes next. A Wednesday afternoon sitting at 80% full two days out is a different problem from the same afternoon at 10% full, and a static card treats them identically.
Rate movement by how far ahead the booking is made is the third layer. Hotels call it advance purchase pricing. It works in golf because it pulls demand earlier into the booking window, where you still have room to move things around.
Rate movement on outside signals is the fourth: weather forecasts, a tournament two towns over, a school holiday. This is the layer that gets demonstrated most often and shipped least often, and it is worth making a vendor show you the rule firing rather than describe the capability.
What does the research actually say about discounting green fees?
The most useful study on this question is Enz and Canina (2017), published in the Journal of Revenue and Pricing Management, covering 80 golf courses and more than 2 million rounds.
Two of its findings matter to a daily-fee operator.
The first is that demand for golf is, in the authors' words, "primarily price inelastic, suggesting that a drop in price will not have a significant positive impact on demand and revenue will decrease." Read that twice, because it is the reverse of the standard pitch. Cutting a rate to fill a quiet slot does not reliably attract enough additional play to cover the revenue given up on the players who were coming anyway.
The second finding is sharper. Courses priced above their competitive set "not only reaped higher RevPATR, but also experienced higher daily course utilization." RevPATR is revenue per available tee time, the golf equivalent of a hotel's RevPAR, and it is the measure this article keeps returning to. The courses charging more were earning more per available slot and selling more of those slots.
Treat that as evidence rather than as law. It is one study, on a US sample, comparing price positions between courses rather than testing intraday rules inside a single course. Price position also travels with course condition, and a well-kept course can charge more and fill more for reasons that have nothing to do with its rate engine. What the study does establish is that the assumption sitting under most dynamic pricing conversations, that a lower price buys volume worth having, did not survive contact with two million rounds of data.
Why the discounting instinct is usually wrong
The empty 2pm Tuesday slot feels like free money. Nobody is playing it at $65, the cost of one more group looks like nothing, and $40 beats $0.
Two things break that reasoning.
The first is cannibalization. The $40 rate is not offered only to the golfer who would otherwise have stayed home. It is offered to everyone booking that window, including the retired regular who plays every Tuesday at 2pm and would have paid $65 without thinking about it. If four of every five groups you sell at the lower rate would have booked anyway, the discount has to generate a great deal of genuinely new play before it breaks even, and the research says that new play does not arrive.
The second is that the marginal cost of a round is not zero. Another group still consumes cart wear and turf recovery and a share of the daylight, and on a busy afternoon it consumes pace as well.
Filling the slot is not in doubt. A lower price will do that. What no dashboard can show you is how many of the golfers who took the lower price would have paid the higher one, because that comparison is not observable in your own booking data. Only a deliberate test gets close to it, which is the subject of a later section.
The lift number nobody can source
Dynamic pricing is sold on a lift percentage. You will see first-year ranges, and you will see a headline upper bound presented as observed in case studies. We went looking for the study behind any of them and could not find one. Not a gated one, not a small one: none. That is why no lift percentage appears anywhere in this article, including a flattering one.
That is worth stating rather than quietly correcting, because those numbers are everywhere and most of them come from the same kind of place. One figure you will meet repeatedly, a 19% average green fee revenue increase said to rest on analysis of more than 12 million rounds, is hosted on a domain that reads like a governing body. Follow it back and it is a software company's marketing about its own product, republished through a partner announcement. Two other percentages in wide circulation, an 8% figure and a 10% to 20% range, are likewise vendors describing their own results.
None of this proves dynamic pricing does not work. It means nobody has published an independent estimate of how well it works, and every percentage in circulation should be read as a marketing artifact until somebody names the sample behind it.
So there are four questions to put to any vendor quoting a lift figure. How many courses. Over what period. Measured against what comparison group. And who paid for the analysis. A number with no control group is not a measurement, it is a before-and-after across a period in which the weather also changed.
The inventory problem underneath the pricing problem
The National Golf Foundation reported in January 2025 a 9% no-show rate, drawn from more than 500 US courses and 10 million rounds. No pricing rule addresses that. Deposits on peak inventory, a card held on file and a reminder sequence that actually reaches people are what move a no-show rate.
Work the size of it out on your own sheet rather than on ours. NGF's 18-hole facility profiles put combined green fee and cart revenue between $793,500 and $825,100, so take the round $800,000 for the arithmetic. If that 9% rate applied evenly to revenue and none of it were recovered, it would stand for about $72,000 of sold inventory that produced nothing. In practice some of those rounds are prepaid, some get backfilled, and some were never going to resell at short notice, so treat $72,000 as an outer bound rather than an estimate. Even so, recovering a quarter of it would be worth about $18,000, which is more than most courses could plausibly capture by discounting Tuesday afternoons for a year.
There is a second inventory problem, and it is about information. NGF also reported in August 2025 that only 40% of golfers book tee times exclusively or mostly online, against 80% to 90% in airline and hotel bookings. If most of your reservations still arrive by phone, your demand curve is being read from a partial sample, and the rate a caller is quoted depends on whether the person at the counter reads the screen or the laminated card taped beside it. Rules that bind only on the website are not pricing rules. They are website rules.
Capacity is a pricing lever, and a safer one
Riccio (2012), in the International Journal of Golf Science, modeled the effect of a wave-up policy on a par 3 and found the capacity of the course rising from 6 to 6.67 groups per hour, which the paper puts at a 10% increase. That is inventory created at full rate rather than inventory sold at a discount, and it points in the direction the Enz and Canina result favors.
The obvious shortcut in the same direction backfires. Kimes and Schruben, writing in the Journal of Revenue and Pricing Management in 2002, simulated what happens when you compress tee time intervals to fit more starts into the day, and reported that "interval reductions may actually lead to decreased revenue". The added rounds are paid for in pace, and a five and a half hour round is a rate cut you never get to book as one.
If you are going to discount, fence it
None of this makes discounting always wrong. It makes unfenced discounting wrong.
A rate fence is a condition attached to a lower price that a full-rate buyer would not accept. Walking only. Weekday before 11am. Booked fourteen days out and non-refundable. Residents holding a card. The discount is not the point of the exercise, the condition is, because the condition is what stops the lower price reaching the golfer who was going to pay the higher one.
The test is simple and slightly uncomfortable. Would your Saturday 8am foursome accept the conditions in order to get the price? If they would, you have not built a fence. You have cut your rate and put a label on it.
This is also the real reason rate compression fails. A course with a floor of $55 and a ceiling of $65 has neither a fence nor a spread; a $10 difference will not move anyone's tee time, so it adds variance to the rate card without changing behavior. We are not going to tell you what the correct spread is as a percentage of your median rate, because no defensible figure for that exists. The workable test is behavioral. The gap has to be wide enough that a real golfer at your course would change their plans to capture it, and the way you learn that is by trying it on one day-part.
What regulars notice, and what to do about it
The research is not on your side here either, and it is better to know that before the first complaint than after. Kimes and Wirtz, in the Journal of Revenue and Pricing Management in 2003, found that US golfers rated varying price levels unfair, and rated pricing that varies by time of booking as neutral to slightly unfair. Time of booking is precisely the mechanism most golf systems implement. Two things soften the finding without cancelling it. The study is from 2003, before a generation grew up buying travel and ride-hailing at prices that move. And Wirtz and Kimes found in 2007 that familiarity with a pricing practice raises how fair people judge it to be, which makes the 2003 result a starting position rather than a verdict.
So expect the reaction rather than being surprised by it. Rollouts go wrong in the first three weeks, when a regular sees his usual Tuesday round costing $8 more than last week and asks the starter why. What the starter says next decides how the rest of it goes.
Courses that handle this well tell people before it happens rather than after, lead with what got cheaper ("off-peak rates dropped to $40") rather than with what got dearer, and give staff a straight answer to "why is it more today?" that does not sound defensive. The straight answer is usually the true one. The times most people want now cost more than the times most people do not, and the quiet times cost less than they used to.
There is a longer-run risk that matters more than the complaints, and it is the strongest practical argument for restraint on the discounting side. If your rates fall predictably as a date approaches, golfers learn it. A course that reliably drops its Saturday morning rate on a Thursday night has taught its best customers to book on Thursday nights. Rate integrity is the discipline of never making waiting the smarter move, and it is much easier to hold if your rules mostly move prices up on scarce times rather than down on quiet ones.
Where member rates sit in this
Member rates stay fixed. Rate certainty is a large part of what a membership actually is, and a member who watches their own rate float has been sold something other than what they bought. Dynamic rates apply to public green fees.
The gap between the two then becomes a membership argument in its own right. "Your membership saved you $185 this month based on the rates non-members paid" is a better renewal conversation than any brochure, and it is a report rather than a claim, provided the system holds both numbers in one place and can difference them without anyone exporting a spreadsheet.
How do we test dynamic pricing at our own course?
Since nobody can give you a credible expected lift, the only figure worth having is your own. Getting it is more attainable than it sounds, and it takes a season rather than a quarter.
Pick one day-part rather than the whole sheet, and move price in one direction at a time. Leave a comparable day-part untouched as a control, so that when the weather turns you have something to compare against other than last year.
Measure RevPATR rather than rounds. Everything else rests on this. Rounds almost always rise when you cut a price, and revenue can fall in the same week. If your reporting cannot show revenue per available tee time broken down by day-part, fix the reporting before you touch a single rate.
Test upward first. It is the direction the research supports, it is the direction nobody sells you on, and it is cheap to reverse. Raise your Saturday 8am to 10am rate by a visible amount and watch fill rate and RevPATR together. If both hold, you were underpriced, and you have found revenue without giving anything away to the golfers who were already coming.
Run any discount as a fenced offer rather than as a rate cut, so that if it works you know what worked and can repeat it.
Take no-shows out of the result before you read it. A slot filled at a discount by someone who never arrives is worse than an empty slot, because you also turned away the enquiry behind it.
Then be honest about the counterfactual. You cannot see the golfer who would have paid full price and paid less instead. The control day-part is the closest approximation available to you, and the difference between the two is the closest thing to an answer you are going to get without a research budget.
When does dynamic pricing pay for itself?
The break-even is arithmetic, and it is the one number in this article you can compute exactly, because both inputs belong to you.
Take the annual upcharge your vendor quotes for dynamic pricing over a plain tee sheet subscription, then divide it by your annual green fee and cart revenue. The result is the percentage lift you need before the capability has paid for itself.
Two illustrative runs, using the round $800,000 of combined green fee and cart revenue that sits inside NGF's 18-hole facility range. An upcharge of $3,000 a year breaks even at a lift of about 0.4%. An upcharge of $9,000 breaks even at about 1.1%. Both upcharge figures are placeholders chosen to show the shape of the calculation; use the quote actually in front of you.
You will see the round version of this stated as a rule: a 1% lift covers the cost. At the top of a quoted range it does not, which is a small but useful argument for computing a break-even rather than accepting one.
Two things the arithmetic leaves out. Revenue break-even is not profit break-even: a lift produced by discounting arrives with variable cost attached, while a lift produced by raising peak rates arrives with almost none. And the larger cost of dynamic pricing is not the license fee. It is somebody's time, every month, reviewing which rules fired and which have stopped making sense. Courses that set their rules at install and never return to them are the ones concluding a year later that dynamic pricing does not work.
What this platform does, and what it does not
Links Meridian's pricing engine conditions rates on day of week, time window, date range, how far ahead the booking is made in hours, and how full the slot already is as a percentage. Rules carry a priority order, and each can carry a cap on how far it is allowed to move a rate. That cap matters more than it sounds; it is what stops a badly written rule from quoting somebody an absurd number on a Sunday morning.
What the engine does not do is price on weather. There is no weather trigger in it. If a vendor tells you their system discounts automatically on a poor forecast, ask to watch the rule fire, and ask which forecast source and which threshold, because the distance between a real trigger and a manual flash sale somebody runs when it looks like rain is the distance between a feature and a habit.
We attach no lift percentage to this engine, for exactly the reason we told you to distrust everybody else's: nobody has published a defensible one, and inventing ours would make this article an example of its own complaint. What we will do instead is let you drive the rules yourself in a demonstration and watch them fire against your own rate card. A capability you can test beats a percentage you cannot.
Where this leaves a daily-fee operator
Test upward before you test downward. The instinct runs the other way, and the instinct is the expensive one.
Deal with your no-show rate before you touch your rate card, because the published figure for that problem is larger and better evidenced than anything published about rate rules.
Build the RevPATR report before you build the pricing rules, since without it you cannot tell a revenue gain from a volume gain that cost you money.
Fence every discount, or accept that you are cutting your rate for people who were already coming.
And treat every lift percentage anyone puts in front of you, including any that arrives with our name on it, as a claim with a named sample attached or as no claim at all.