Google Analytics and Your Booking Engine


Several of our previous posts on analytics have dealt with the “how to” aspect of setting up various things you might want to track across your website and your booking engine. A recent online forum discussion among innkeepers about this topic prompts an overview of what things you can track, why you might want to track them, and some potential problem areas.

1. Is your site set up to use Google Analytics?

Does your booking engine support ecommerce tracking? Is it set up correctly?

Do you have Google Analytics set up on your website? It should appear on every page (any page that doesn’t have it, will not be tracked). How do you verify that it is set up?

  • Open your favorite browser and visit your website. In most browsers, you can press Ctrl-U on the keyboard to view the source code of the page you are viewing. On all you can right-click on the page and select View Source (or something very similar) to see it.
  • Look through the source (or press Ctrl-F to search, if you prefer). You are looking for code which will include “_gaq”. By the way, if you find code that includes “_gat” you are using the older, slower, form of Google Analytics, and it should be updated to use what is called the asynchronous code. If you don’t find either the “_gaq” or “_gat” sections, you don’t have Google Analytics installed.
  • Once you find the code, check to see what it looks like. Basic tracking looks like this (the XXXXX-X is your Google Analytics ID, and don’t worry if the lines don’t break at the same points):

<script type="text/javascript">

var _gaq = _gaq || [];
_gaq.push(['_setAccount', 'UA-XXXXX-X']);
_gaq.push(['_trackPageview']);

(function() {
var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true;
ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js';
var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s);
})();

</script>

Most changes we are interested in will be made between the first and second lines of the code that start with “_gaq.push”.

2. What is tracked with basic tracking, and what is not?

Basic tracking (using the code we see above) in Google Analytics will allow the default reports (which are quite extensive, by the way) to track where visitors to your site came from (the site they visited where they clicked on a link to your site), the number of visits, time on page, bounce rate, etc.

What basic tracking doesn’t do is track visitors after they leave your site. Under some circumstances, this is not a problem. However, if, like most small lodging properties, you use a booking system that requires the visitor to go to another site for the actual booking (like virtually all booking systems in the B&B world), Google Analytics will stop tracking the visitor as soon as they go to the booking engine.

3. What else could we track that would be helpful?

There are two things that are most helpful to track. You could do one or the other, but we recommend using both, as they each provide different information.

A. Ecommerce

If your booking engine supports it (some partially support it, but many do not support it at all!) you can set up Ecommerce tracking in Google Analytics. Among other things, this will allow you to see what other site referred a guest to your site when the guest completed an online booking. In addition, with the addition of Multi-Channel Funnels, you can see what other sites help to produce bookings for you.

B. Goals

If you set up a Goal in Google Analytics for a completed booking, you can track how guests proceed to the booking, where they abandon the process, and also see what sites contribute to your online bookings. Again, not all booking systems support this fully, but it requires less work on their part than Ecommerce tracking.

4. What do I need to do to track these things?

First of all, be sure your booking engine supports Google Analytics. Scott Crumpton of Whitestone Marketing has worked with many of the booking engines, and gives the following list of systems and their ability to work with Ecommerce tracking in Google Analytics:

  • AvailabilityOnline: Since no dollar amount is taken with the booking, ecommerce tracking is not available
  • RezOvation GT: Works properly.
  • RezStream: Works flawlessly though Scott says he has run into bugs from time to time.
  • SuperInn: May not be currently working
  • Webervations: To track visits and goals you must put your GA ID in a box. However, to use Ecommerce tracking you must actually insert proper GA code in another box.
  • ResNexus: We’ve seen it work, but have also had complaints of it stopping working from time to time.

It is possible that other systems work with GA’s Ecommerce tracking, but it is not a simple process, and often the booking engine will say it works when it only works to a limited degree.

In addition it is possible to set up Ecommerce tracking to capture the specific items purchased (rooms booked, etc.) via the booking engine. However, so far as we are aware, none of these booking engines have not set up their systems to provide that level of detail.

If your booking engine is one which supports ecommerce tracking, you need to enable ecommerce tracking in Google Analytics. To do this, go to Admin -> Profile Settings, and under Ecommerce Settings select “Yes”.

The second step will be, to add two lines in your GA code on each web page. These lines should be added before the line which reads

_gaq.push(['_trackPageview']);

The lines to add (changing YOURWEBSITE to your actual domain name) are:

_gaq.push(['_setDomainName', 'YOURWEBSITE.com']);
_gaq.push(['_setAllowLinker', true]);

In addition, any links to the booking engine, or forms which send information to the booking engine, must include GA code, as outlined on the GA Help site.

Once you have Ecommerce tracking (and/or Goals) set up, you will be able not only to see where bookings are coming from, but also to use Multi-Channel Funnels to see what sites are contributing to your online bookings.

Happy tracking!

Posted in Analytics | Tagged , , , , , , | Comments closed

Guest Satisfaction vs Revenue: Things to Think About


A recent post from Daniel Edward Craig on the 4 Hoteliers site (and re-posted in other places) has caused me to reflect a bit on the “business is business” side of things, as compared to the warmer, fuzzier, guest satisfaction side.

slide 300x199 Guest Satisfaction vs Revenue: Things to Think AboutThe article from Daniel Craig was based on a question he had received, asking if your hotel ranks at the top on TripAdvisor, are you not charging enough? Daniel surveyed a number of lodging professionals (hoteliers and consultants), and received a variety of answers. He concludes that there is no consensus, but clearly the benefits outweigh any costs.

From reading the article, it appears that by his conclusion, Daniel is saying, in essence, that being at the top on TripAdvisor is worth more than could be gained by a price increase which could cause a drop in TripAdvisor rank.

A bit of context

The question of raising rates, at the possible expense of TripAdvisor ranking, is essentially analogous to the concept of revenue management. That is, the approach used by airlines, many larger hotels, and sometimes other businesses, to use a fluctuating rate scale, rather than posted rates. Thus, when there is lots of available inventory, prices are relatively low. As inventory shrinks (fewer available rooms), rates increase. Other factors, such as holidays, may also play into the rate ultimately charged.

Using that analogy, then, the theory behind the question would be that there is a “sweet spot” where rates are increased until the TripAdvisor position slips down, just below the top group of properties. If this theory were viable, a property would be maximizing its revenue by charging more than its competitors, while not losing enough occupancy to reduce overall revenue. The profits would be gained by (a) increased room rates, and (b) reduced costs due to slightly lower occupancy.

Things to Consider

1. Is it True?

Is it true? Is there a way to estimate the number of rooms a property would lose by dropping a few spots on TripAdvisor?

Searching for this information, we found a lot of articles stating that higher rankings means more bookings, which is certainly quite a reasonable expectation. However, none of these articles actually quoted any study, or supplied data, to back up these claims. The closest we could come to real data was a study by WIHP (a hotel marketing consultancy), indicating that, for properties in the top 20% for their destination get a lot of bookings from TripAdvisor. However, this study indicates that even properties below that 20% mark get a significant share of their bookings as a result of TripAdvisor.

Thus, while there is something of an urban legend that properties which rank higher on TripAdvisor will get more bookings than the others, there is little data to indicate that this is always the case. Certainly rank makes a difference, but there may be little difference between properties ranked close together.

2. What is the Price of Lost Guest Satisfaction?

Since reviews on TripAdvisor, and particularly the ratings which are used to determine the rankings, are purely the subjective views of the reviewer, there is no way to objectively correlate one rating against another. As such, these ratings are, in essence, measurements of customer satisfaction, or popularity of the property. No doubt one review (or even a handful of reviews) ranking a property lower than do the other reviews, will be seen as an aberration. However, consistently lower reviews, especially when accompanied by descriptions of perceived shortcomings of the property, will be seen as a significant decline in customer satisfaction.

If that decline in customer satisfaction in the form of less favorable reviews causes a property to lose its ranking on TripAdvisor – to use the WIHP data, below the top 20% for the location – a significant decrease in reservations (perhaps as much as 50% of the bookings related to TripAdvisor – which could be a loss of 5-10% of overall bookings) could overcome any price increase.

3. Can You Stop the Slide?

The general sense of revenue management is that the loss of revenue due to lower occupancy is more than offset by the increased rate. There is a customer satisfaction component to this, as well, but customers do understand the law of supply and demand – as there are fewer rooms available, paying more for a room rarely causes “sticker shock.”

This is not the case if you raise rates until your TripAdvisor rating slips. There is nothing that the guest can see as an objective reason for higher rates (such as few rooms being available). In other words, there is no objective, economic, reason for the rate increase. That, in itself, leads to a negative guest experience, which may well lead to more negative TripAdvisor reviews. This could clearly spiral out of control.

In addition, the analogy to revenue management breaks down in a very significant way. With revenue management, the situation is entirely in the control of the property. If a mistake is made, and rooms are not selling, changes can be made and the damage kept to a minimum.

Not so with TripAdvisor reviews. If your high rates have caused unfavorable reviews, the property has no control over which guests, or how many of them, post reviews. You have no control over the rating they give. It is fairly well accepted that unhappy guests are more likely to be motivated to write reviews than happy guests. Is that a chance you want to take?

In addition, you have no control over how far your ranking may slide on TripAdvisor. You do not control (or even have a way to accurately understand) their proprietary algorithm. While you can guess how it works to some degree, there is no way that you can push rates down a bit to get better reviews and “adjust” your ranking. In fact, just to disabuse any of the notion that price-tweaking can affect the ranking with any degree of precision, here is a quote from WIHP’s study:

Cutting prices doesn’t seem to be a large factor in customer satisfaction, at least not near as large as quality of the hotel and it’s service []. A guest is more likely to remember a great experience than a great price.

One other factor that we’ve noticed over the years – the positive side of guest satisfaction. Happy guests come back, and they send friends. One thing that is very hard to measure is the value of that TripAdvisor ranking that sent a guest, who enjoyed their stay, and not only came back (perhaps again and again), but also referred friends and family.

Posted in Reviews | Tagged , , | Comments closed

BBOnline Traffic Results Confirmed


Last week’s post, examining whether or not BBOnline’s staff had been able to rectify the huge drop in referrals and bookings after their site redesign a year ago, observed that more information was desirable, reflecting the results of other properties.

In the comments, Sarah provided some information, indicating a similar drop of 61.4% in traffic, noting that others had increased, while some had dropped slightly, but BBOnline’s drop was the most significant – by a lot!

We also were contacted by a consultant who works with several inns. He shared with us some statistics from several of his clients, and gave us permission to post them without identifying the properties. He also noted that several of his clients have dropped BBOnline’s directory, so these only reflect clients who are still listed on BBOnline.

These figures are all for Sept. 1, 2011 through August 31, 2012, compared with Sept. 1, 2010 through August 31, 2011, so should provide a good comparison to the figures in our previous post.

Property 2011 Visits 2012 Visits Pct Change
Property A 588 256 -56.46%
Property B 1,105 350 -68.33%
Property C 883 343 -61.16%
Property D 693 178 -74.31%
Property E 937 451 -51.87%
Property F 892 277 -68.95%

The average drop year-over-year is about 63.5%.

Our statistics indicate that, for B&B directories, such as BBOnline, every 100-150 hits yields a booking. If those statistics are even roughly accurate, they indicate that these properties have dropped from getting 6-10 bookings a year from BBOnline to getting only about 1-3 bookings per year. This is consistent with the reports that have come from other innkeepers.

How can you use this information?

As our previous post observes, it is important to use tools like Intell-a-Keeper or Google Analytics’ Multi-Channel Funnels, to make sure that credit is being given for all bookings that a directory may have helped to promote. That said, the big question, is what is the ROI (return on investment) of the directory?

If you know exactly which bookings came from BBOnline, add up your total revenue from those bookings (not including taxes). If you don’t know the specific bookings, then take your average length of stay in days and multiply it by your average daily rate (ADR). The result is the average value of a booking. Multiply that number by the number of bookings from BBOnline to get the approximate revenue from BBOnline.

Take the resulting figure (revenue from BBOnline) and subtract the cost of your BBOnline subscription. The result is the raw profit or loss from BBOnline. Divide that number by the cost of your BBOnline subscription to get the ROI.

There are no “hard and fast” rules about what your ROI should be. If a directory provides value other than bookings, that should be added to the figure for revenue from the directory. In any case, you would certainly expect the ROI to be more than 2 or 3. Some say it should be more like 8-10. Our view is that it depends on the benefits the directory brings you. A directory that gets you media coverage, for example, has brought you incalculable value, yet may not directly produce high numbers of bookings.

Posted in Directories | Tagged , , , , , , | Comments closed

BBOnline – Bouncing Back? Or Not?


Last April we first posted our concerns about the sudden drop in traffic from BBOnline to our Bed & Breakfast in Freeport Maine. Our traffic had dropped 95% for the first several months of 2012 compared to the same time frame in 2011. Other innkeepers reported similar drops.

We followed up that article with another, detailing concerns about an email response to our inquiries about the traffic drops, then a third article about a more encouraging telephone response from BBOnline staff.

Is BBOnline bouncing back?

Now that our busy season is three-quarters over, we wanted to see where things stand with BBOnline. Are they bouncing back and returning to their former spot as one of our top B&B directories? Or do they continue to languish at the bottom, sending us only a trickle of the former traffic?

Bouncing Back?

 BBOnline   Bouncing Back? Or Not?We began our inquiry with Compete.com, which provides traffic analysis. This type of analysis only works well for sites with heavy traffic – smaller sites usually have too little traffic for Compete to be of use. As the accompanying graphic shows, BBOnline’s traffic plummeted in November, 2011, but didn’t hit bottom until January 2012, but seems to have recovered well in March and April, then dropped again from May through August of 2012. This coincides with our observation in our third article, in June 2012, that there were some signs at that time of a BBOnline recovery, but it was too early to draw any conclusions.

However, since the summer months are usually the highest web traffic months for B&B’s (even when you take into account different seasonal traffic in areas where the weather patterns are opposite), we really need to see how this “bounce” relates to other B&B directories.

Looking at Compete.com for BedAndBreakfast.com, BnBFinder.com and ILoveInns.com (the other directories have too little traffic to be a useful comparison), we see that BedAndBreakfast.com bottomed out for the season in December, then began to climb, while BnBFinder and ILoveInns.com hit bottom in November and began their climb in December. BedAndBreakfast.com hit its peak in July, while BnBFinder peaks in June, then remains relatively flat, and ILoveInns.com peaks in February, then again in June.

What stands out about this pattern for BBOnline? In essence, their traffic, when compared to the traffic leader, BedAndBreakfast.com, shows their extended low point that prompted us to write our earlier articles, but they seem to have bounced back to be following approximately the same traffic patterns as the other directories.

What about the bottom line?

First a couple of words of caution: (a) statistics from only one source (our B&B) are not necessarily a valid yardstick by which to generalize, and (b) sources of information that measure only the last source that sent traffic to your booking site are missing valuable information (as we pointed out recently).

That said, we measure effectiveness of our referral source with three separate pieces of information. We make no effort to combine them, as in some cases they would overlap, and in others they would be completely separate. We use:

  • Guest statements about where they located us
  • Google Analytics
    • Ecommerce and Goal tracking by referral source
    • Multi-Channel Funnels
  • Intell-A-Keeper for mulitple source referrals

According to our guest information, BBOnline, previously a quite significant source of revenue for our Bed & Breakfast, dropped by about 40% from 2010 to 2011, but in 2012 dropped again, to about one-third of the 2011 levels.

According to Google Analytics, our 2012 year-to-date figures, compared with the same portion of 2011, shows that our referrals are still down – though not by the 95% we had found in the early part of 2012. Now the year to date comparison shows a drop in referral traffic of “only” 63%. Clearly that is better, but still a far cry from the 2011 levels, which are, in turn, much lower than the 2009 and 2010 levels (since 2009 was a recession year, to be lower than that is very significant). In addition, the “Bounce Rate” for visitors coming from BBOnline – the percentage of visitors who view only one page before leaving the site – has increased dramatically over 2011.

In the same time comparison, GA also shows “last click” referrals accounting for only a tiny amount of Ecommerce revenue – about 5% as much as the guests report for 2011, and none, at all, in 2012.

Checking the Multi-Channel Funnels of Google Analytics, we see again a tiny amount for assisted Ecommerce conversions in 2011, and none in 2012.

So far, our Google Analytics and our guest reports show about the same thing – very little revenue from BBOnline.

Next we checked Intell-A-Keeper’s stats for 2010, 2011 and 2012. We found that there was a drop of about 60% in bookings from 2010 to 2011, but, again, no bookings in 2012.

Summing Up

All our tracking sources agree that there is a very significant drop in traffic from BBOnline to our website, and that the revenue attributable to BBOnline is only a small fraction of its prior levels. While the Compete.com statistics (which carry their own margins of error) seem to indicate improvement in traffic to the BBOnline site, our own records indicate that this has yielded only a small improvement in traffic to our site. Indeed, with the increased bounce rate figures, it looks like BBOnline is sending fewer visitors, who are less likely to be interested in booking.

Have you seen different results? Please post your observations in the comments!

Posted in Directories | Tagged , , , , | Comments closed

Comparing Directory Performance at a Glance


Last week we talked about how goals in analytics are used in Multi-Channel Funnels to show you how referral sources can be sending value (including bookings), that doesn’t always show up in the standard analytics reports. This week we’ll take a different approach, and look at creating a Dashboard designed to give you, with only a quick glance, an easy way of comparing online directory listing performance.

From the start, we want to emphasize a word of caution – we are not suggesting that this dashboard will replace a more in-depth comparison of online directory performance, or that it should be used instead of the Multi-Channel Funnel approach from last week’s post. This dashboard is intended to provide a quick overview, so you can get a sense of relative performance. It should not be used for important decisions, but more as an easy way to “keep your finger on the pulse” of performance.

We had previously written about simplifying your Google Analytics usage by using dashboards. This time we’ll just create a straightforward dashboard to gather information about online directories for comparison on one screen.

If you aren’t interested in how it works, and just want to install it as a custom dashboard in your own GA setup, just be sure you’re logged in to GA, then click here to install it.

Whether you just click and install the directory performance dashboard, or you read through the discussion below, you’ll want to customize one line in each of the six widgets on the dashboard. See those notes below.

Create a New Dashboard

To create a new dashboard, click on Home, then under Dashboards, click +New Dashboard, select Blank Canvas, and give your dashboard a name, such as Paid Directories, or something similar.

Adding Widgets

We’ll add six widgets (you can skip some if you don’t think they are helpful to you, or add others, of course). The first will be a widget to show the total number of visits from paid directories.

Click Add Widget. The popup has Metric selected by default, and that is fine for our purposes. Where the blue entry says “Add a metric” click and select Visits. Next click Add a filter, and click the green Dimension link and select Source. Change the dropdown to Regular Expression, and paste in the following line (this is the part you’ll need to edit for your directory listings – it is shown on multiple lines to make it fit here, but should be all on one line in your widgets):

bedandbbreakfast.com|bbonline|iloveinns|bnbfinder|
lanierbb|selectregistry|tripadvisor|bescover

Every one of our widgets needs this filter, so they will only show statistics relating to the online directories whose performance we want to compare. The “pipe” character (“|”) means “OR” so the filter simply searches for each item in order. Using the list above the filter will first add all visitors where the source was bedandbreakfast.com, then all where the source includes bbonline, then iloveinns, etc. Be sure there is no pipe character at the beginning or end, but there should be one between each entry.

Note to those who just installed from the link above: The dashboard widgets use two special sources, “bbcom” and “ta” which are unique to my setup. You will have to change them to “bedandbreakfast.com” and “tripadvisor”.

If you don’t use one of the directories, take it out, but don’t leave two pipe characters in a row (like “||”). If you have other directories to include, add them (you don’t need the full domain name – just the part that makes them unique) separating them from the others with a pipe. Use the default name for the widget, or change it if you prefer, and Save the widget.

 Comparing Directory Performance at a Glance

Visits by Source

The second widget will show us the number of visits by source. Add a new widget, and make this a Pie Chart. Set the blue metric to Visits and the green dimension to Source. Make sure the dropdown indicates it will show 6 slices. Add a filter on the Source, with a Regular Expression, just like the one above. Save the widget.

 Comparing Directory Performance at a Glance

Visit Duration and Bounce Rate

Our third widget is a table, where the metric is Source, the first dimension is Avg. Visit Duration, and the second dimension is Bounce Rate. Create a filter on the Source with the same Regular Expression. Name it and Save it.

This widget will allow us to compare online directory listings and see how well visitors from each source engage with our site – how long they stay, or how many “bounce” (leave after viewing only one page).

 Comparing Directory Performance at a Glance

Conversion Rates by Source

This will show us how each online directory listing compares for ecommerce conversions and other goal conversions. If you use Ecommerce, you can follow along. If not, you may want to consider implementing it (for more reasons than this dashboard!), but meanwhile, you can use other conversion dimensions for this widget.

Add another Table widget, with the metric as Source, and the first dimension Ecommerce Conversion Rate and the second dimension Goal Conversion Rate. Add the same filter Regular Expression, name and save the widget.

 Comparing Directory Performance at a Glance

Transactions by Source

To see how our bookings compare by online directory listing (using only last click comparisons, of course), we’ll build a widget to measure transactions by source. This widget is a Pie Chart widget, with the blue metric being Transactions and the green dimension being Source. Add our standard Regular Expression filter, name and save the widget.

 Comparing Directory Performance at a Glance

Visits and Transactions by Source

Our last widget will allow us to compare the number of visits and transactions by source. We’ll create a Table widget, where the metric is Source, and the dimensions are Visits and Transactions, respectively. Add our usual Regular Expression filter, and Save the widget.

 Comparing Directory Performance at a Glance

Step back and admire your handiwork. You may want to re-order your widgets (just drag them where you want them). For example we put the Visits from Paid Listings first, as something of an overview, with the Visits by Source pie chart below it. In the center column, we have Duration and Bounce Rate by Source table on top, with the Ecommerce and Goal Conversion Rate table below. In the right column we have the Transactions by Source pie chart, with the Visits and Transactions by Source table below it.

Now you have a tool to give you quick insight for comparing the performance of online directory listings at a glance.

Posted in Analytics | Tagged , , , | Comments closed