DARM Series: Setting the Bar with RevPAR

There were hurricanes, fires, pandemics, and run-of-the-mill technical difficulties … and we’re talking about DARM, not the apocalypse. While many others took rigorous notes to create sweeping summaries of the conference and its sessions, RevPARTY took note of where we could add value. In this series, we cover topics that may have been missed, concepts that are hard to cover so briefly, and discuss the ideas we wished we had but time did not allow.

Where We Left It At DARM

As happens more often than we’d like, we ran out of time during the RevPAR Discussion Panel at this year’s Vacation Rental Data and Revenue Management Conference. We were able to cover many of the advantages and minutia of using RevPAR as a key metric in our industry:

• It can be customized to many needs and perspectives (adjusted, bedrooms, etc.)

• It can account for fluctuating inventory levels

• It captures the tradeoff between ADR and occupancy

• It lends itself to analysis because it treats the collinearity/positive correlation between ADR and occupancy (i.e., in peak season ADR and occupancy both raise at the same time)

• Economically it represents the expected value of any room night which is a simple way to make decision making easier (the probability a room night is booked is equivalent to occupancy and the expected reward is ADR, making RevPAR the expected value)

But … I think there’s something you should know about RevPAR that wasn’t covered.

It takes data and revenue management to the next level. It supports other functions that never touch data platforms, pricing models, or the financials. It not only makes for a better revenue manager; it makes for a better vacation rental manager.

It’s the space between RevPAR and Net RevPAR or the distance between actual RevPAR and RevPAR targets.

The Problem with RevPAR Alone

Most commonly RevPAR is utilized by the revenue management team and decision makers to gauge top line revenue, year over year growth, and portfolio performance compared to market. Its great at doing all of those things. However, especially as a portfolio and team grow, it gets more and more difficult to make such generalized metrics useful.

Net RevPAR helps get at this issue but only at a surface level. With Net RevPAR you can evaluate a property or portfolio’s profitability. At the end of the day, this is the number any top-notch vacation rental manager cares about the most. They’re running a business and the goal of the business is to be profitable.

But with only those two numbers, how do vacation rental managers know what to do to get better? Are their costs too high? Do they need more revenue? The answer to both is always “yes.” But how do they do it?

Line Items on a PAR Basis

We all know that the difference between RevPAR and Net RevPAR is all of the associated costs and expenses with renting that room night to a guest. Many, especially larger portfolios, will break down their expenses to a PAR (per available room) Basis to report on and track unit economics for owners and investors. It’s essentially their financial statement divided by room night count (rooms in inventory x nights in period) so that year over year changes are inventory agnostic.

By simply breaking out line items on a PAR Basis, it helps contextualize the impact each item has on Net RevPAR. If MaintenancePAR, UtilityPAR, RentPAR, or any other cost metric is relatively high then it’s an indicator that that component of the business could use a little work; the same as a well-done vacation rental Profit and Loss Statement.

But the more amazing part is the scene it sets as a vacation rental business plays out on a larger stage.

How Did We Miss Targets & How Do We Fix It?

While calculating costs on a PAR basis helps you identify and treat challenges to Net RevPAR and profitability, it doesn’t diagnose challenges to revenue. As a portfolio grows, as a team grows, as there are more markets to manage, and as strategies become more complex and solutions more sophisticated; it gets harder and harder to pinpoint useful and impactful insights. Most short-term rental managers want to know:

• Why and how did we miss our targets?

• Who was responsible for the miss?

• Could it have been prevented?

• How do we fix it?

• How do we prevent it from happening in the future?

In order to show how anybody could answer these questions and drive results in their business, we’ll breakdown an example with simplified components step by step (I tried to do a full financial breakdown but it’s a bit heavy for a blog post).

Unit Count

First, there needs to be a consistent approach to unit and room night count. Most vacation rental managers simply remove “blocked” (owner blocked, comped stays, property hasn’t onboarded yet, etc.) room nights from their PAR calculations. This gives a really fair and apples to apples view of efficiency and effectiveness. The downside is it doesn’t always explain the financials in totality.

When looking at top line company revenue to analyze cashflows, ROI for investors, or budgeting the absolute amount of money matters. Any discrepancy in actual unit count to forecasted unit count has to be accounted for because if it isn’t there will be missing money from the bank account. Moreover, if you have a property onboarding/launch team or an owner sales team the unit count variance is an indicator of their ability to convert the property pipeline and meet deadlines.

Market Performance vs. Property Performance

Everyone in the vacation rental industry benchmarks their portfolio RevPAR to the market RevPAR. It’s an easy to comprehend metric, its how we benchmark stocks and mutual funds, and it feels good to say, “Our portfolio beat the market by 20%.” But beyond knowing if the portfolio did “good” or “bad” it doesn’t generate any actionable insight.

Vacation rental portfolios have to properly forecast in order to generate actionable insight and begin to understand any miss against targets. There are three key components to RevPAR, ADR, and Occupancy forecasting that are absolutely necessary to diagnose challenges and find solutions to maximizing vacation rental revenue:

• Forecasted/expected market performance (Market RevPAR)

• Forecasted/expected portfolio performance compared to market (RevPAR Pen)

• Forecasted/expected portfolio performance (RevPAR)

The forecast can estimate portfolio performance by assuming it performs at X% compared to market or it can estimate performance compared to market by assuming a certain portfolio performance, but all three components must be included.

By forecasting this way, property managers can determine if their lack of or excess revenue was due to changes to the market as a whole or due to the performance of the property/portfolio against the market. Any change in Market RevPAR represents changes to the market as a whole while changes to RevPAR Pen signify individual property performance issues.

For example, if the market underperformed by 10% compared to forecast that change would correlate to a drop in actual property RevPAR. Even if the property performed just as well against the market as forecasted, the property RevPAR would be lower because the whole market under-performed.

Additionally, if the property performed 10% worse against the market than expected, even with the market under-performing, if the property had performed as forecasted it would’ve had a higher RevPAR. But, because the property under-performed compared to market in a market that under-performed the negative effects to RevPAR are cumulative. The change in RevPAR Pen accounts for a further drop in the property’s actual RevPAR.

All told, the example property missed its RevPAR projections. Normally, most property managers would say something like, “the market was down,” or, “we messed up.” Unfortunately, those aren’t useful in diagnosing problems and finding solutions. If the vacation rental manager forecasted and compared actuals using the method above, they could say, “We lost $11 in RevPAR to market underperformance which wasn’t preventable … but our property also performed 10% worse than expected so instead of recouping the market loss we gave up another $9 in RevPAR. We need to fix that.”

Diving Deeper and Diagnosing

Forecasting RevPAR and analyzing the results using this method is a great way to understand a portfolio’s overall performance but diving deeper really helps you diagnose underlying challenges.

The first and most obvious step is to break RevPAR into its components: ADR and occupancy. Using the exact same method as above, ADR and occupancy changes can be attributed to market performance or portfolio performance.  It’s a crucial step in determining how targets were missed and where your portfolio or market’s strengths and weaknesses are.  

To take any analysis further, it gets a lot more complicated. There are so many factors that contribute to ADR, occupancy, and RevPAR that only with careful metric tracking and an experienced hand can true insights be extracted. Many of the various components also have correlations and interactions which influence one another and cause endogeneity. However, if you can deep dive on these components at the smallest level then their impact can be large.                                                                        

RevPARTY Takes Your RevPAR to the Next Level

RevPARTY, as the name implies, has a strong handle on RevPAR. When you break down expenses to a PAR basis, we know what a normal breakdown looks like because we’ve worked with so many portfolios across so many markets. Not only can we help you with the RevPAR, ADR, and occupancy calculations above, but we can push the analysis further to identify and solve your revenue challenges. Whether you’re a DIY kind of manager and just want some additional information from our Vacation Rental Guides or you want a bespoke engagement/revenue management services, RevPARTY is here to help drive revenue and drive results.

RevPARTY Monte Carlo Simulation: For Property Underwriting and Financial Projections

RevPARTY has recently introduced a new component to its financial projection modeling for property underwriting that is a game changer for any company evaluating the risk involved in a new property or looking to benchmark their property’s performance against expectations. RevPARTY’s Monte Carlo Simulation now provides value to companies that other products simply cannot. Rather than just spitting out a number, RevPARTY’s Monte Carlo Simulation provides a range of possibilities and the probability that each outcome will occur. As opposed to relying on assumptions that may be too aggressive or too conservative and ignoring interconnected components of a model, now its possible to accurately asses risk in a variety of future possibilities and make better decisions now. RevPARTY’s Monte Carlo Simulation allows property managers and property owners to better understand ROI, evaluate best and worst case scenarios, avoid potential mistakes, and act on better opportunities.

What Is A Monte Carlo Simulation?

Monte Carlo Simulations are algorithmic models that rely on random sampling to produce a range of results. The repeated iterations of the same model with different randomized inputs produces a range of outcomes as opposed to a single, deterministic, expected value.

For example, we know that if we flip a fair coin four times its expected that (or the most probable outcome is that) two times the coin will land on heads and two times the coin will land on tails. However, we also know that this won’t happen every time. A deterministic model, like standard property underwriting, would only output the expected result of two heads and two tails. A Monte Carlo Simulation would simulate the four coin flips thousands of times and show the range of possibilities: how likely it is that heads or tails appears only once, how often can we expect the result to be all heads, and so on.

This sort of analysis adds value when different results have different consequences. For instance, if the above example is a fair game and a you when the coin lands on heads you might expect to break even … but you would also want to know how often you will lose money; or even further, how often will you lose all your money. This sort of risk assessment becomes even more valuable when a company is deciding between multiple opportunities or have different appetites for risk. Applying this to vacation rentals and hotels is especially important with how many variables affect each other and drastically alter profitability from one scenario to the next.

Monte Carlo Simulations are especially helpful when random variables follow some type of distribution or range (like a normal distribution), when multiple random variables interact (how ADR and occupancy combined determine revenue or how occupancy directly drives housekeeping fees), previous static states effect future states (as is the case when one month’s bookings and reviews influence the next month’s bookings), or when multiple variables are random but correlated (such as how ADR and occupancy have a negative correlation to one another but also share a positive correlation to market RevPAR). All considered, Monte Carlo Simulations are the perfect fit for property underwriting and vacation rental financial projections!

How Does It Benefit Property Financial Projections?

First, a Monte Carlo Simulation more accurately predicts outcomes when a scenario depends on sequential states. This is why it’s often used in planning project schedules, but it also applies to vacation rental underwriting and hotel underwriting. When the success of one month is dependent on the success of the previous months, this causes a ripple effect through the remainder of the projection and a Monte Carlo Simulation more accurately accounts for the randomness of when these events will happen.

A great example of this is the number and quality of reviews on a property’s listing. The quality and number of reviews largely influence the total bookings a listing receives through both occupancy and rate. The more reviews and the better the reviews the easier it is to get bookings; and thus, the easier it is to get even more reviews. While we could estimate the number of reviews during a certain time period there is some randomness as to when exactly that will actually occur. If these bookings and reviews occur earlier than expected, then the following months will likely see revenue results better than projected; but these bookings and reviews could also happen later than planned and have a negative impact on the months to follow. Monte Carlo Simulation allows anyone to play out both scenarios and have the distribution of risk, including the second order effects later in the timeline, displayed in the model outputs.

Second, a Monte Carlo Simulation more accurately models the outcomes of binary events like legal reform or exogenous market shocks. In a normal underwriting model, one would have to model the effect of new laws effecting short term rentals as a risk discount or maybe as two different scenarios (one where the law passes and one where the law fails). However, with a Monte Carlo Simulation we can aggregate both of these scenarios into the same output and it is represented in the range of outcomes. It also allows a business to model the effects of not only IF the law passes, but WHEN the law passes. This leads to a much more realistic range of outcomes and a more accurate reflection of risk.

Third, the simulation more accurately models the range of possibilities and variability which also means it can more accurately model the effects of portfolio size. Everyone remembers the “law of large numbers” and its always at work in the hospitality industry. The same reason that hotels can more easily revenue manage is the same force that causes huge financial swings in vacation rentals, sample size and variability.

For every unit, every night, there are only two possible outcomes that couldn’t be more different: booked and empty. Its 0% or 100% occupancy. Its part of the difficulty with data for vacation rentals. However, with ten units there are now possible occupancies of 0%, 10%, 20%, 30%, etc. Moreover, the increased sample size will cut the outcome variability by more than half. A larger portfolio has less variation and outcomes are easier to predict. Without a probability driven model (i.e., standard underwriting) a portfolio with identical units would always yield the same results regardless of portfolio size. One unit is the same as 100 units. However, with a Monte Carlo Simulation portfolio size, and its increased or decreased risk, is represented in the model outcomes and paints a more accurate picture for any property manager. It directly enables the modeling of scale and diversification.

How Would I Use It?

RevPARTY’s Monte Carlo Simulation is the best tool for property evaluation; whether that’s underwriting a property not currently in one’s portfolio or trying to successfully forecast financials and benchmark performance. The model accounts for the most variables and generates the most reliable outcomes. The repetitive modeling directly tackles many of the data obstacles vacation rentals face. While it can do everything a standard underwriting or forecast model can do, a Monte Carlo Simulation adds the most value by providing solutions in these instances:

Investment Opportunity:

-          When expected value isn’t enough and a fuller picture of probable outcomes and risk is needed

-          When the number of units and the subsequent price is being negotiated but needs to be weighed against the benefits of having an additional unit

-          When there is a possible exogenous shock (like policy, enforcement, development, market trend, etc.) at some point during the underwriting period

-          When trying to obtain financing (especially if trying to demonstrate a low risk of default)

-          When trying to make the best decisions for portfolio selection

Target Setting:

-          To show not just how much a property over/under preformed compared to the target, but to quantify how much the property did so taking everything into account and not just a single metric

-          When setting more accurate targets with a confidence interval

-          When trying to attribute outcomes to or determine the effects of underlying causes (occ, ADR, market performance, a shock, etc.) or trying to drive organizational accountability

Where Can I Get This For My Business?

RevPARTY is thrilled to offer Underwriting with a Monte Carlo Simulation model to all of our clients seeking to underwrite future deals or create targets for future performance. We remove a lot of the leg work and lead time out of the process by developing our model with several clients to standardly incorporate all possible inputs and outputs saving you time and money. Moreover, we customize each simulation to answer the questions your company needs to be successful.

With a model developed by industry experts, tested in the field repeatedly with clients, and analytic sophistication that includes 10,000 simulations you can rest assured that RevPARTY will help you make the best decision possible.