A few years ago, after implementing a powerful online booking system for a rather relevant multi-destination incoming operator, I posed a simple question to the company’s owner, Mister J: which location do you think gets the more quotation requests? He replied, very confidently, that the most requested quotation by travel agents was, no doubt, his native best-selling island. After a simple analysis over the last six months’ data, turned out it wasn’t: by far, the most sought after destination by his clients was the island next door, whereas his own island came in third place.
Earlier this week, a mid-sized and prestigious tour operator’s owner, Mister D, complained to me that there was a big slump in business from summer’s end.
“Can you quantify that slump, compared to last season?” I did ask him.
“Nope”, he replied.
“Well, do you have any idea what caused such slump?” I enquired then.
“Zilch”, he said.
“At least can you tell which segments did shrink in the past few months?” I insisted…
“Nah”, he muttered, knowing by then I will start one of my data analysis tirades.
Reporting has been available for centuries, most of the time in tabular form. Rows and columns of numbers clustered by totals tell a part of the story (usually, the financial part) but nobody is able to see the big picture from a set of numerical tables. Graphs and charts (called data visualizations) helped business owners get better insights for a long time; now cheaper and readily available business intelligence tools assist managers even on forecasting outcomes. More and more data sources are accessible to all kind of companies, as well as solutions to easily visualize the vast amount of data and get actionable insights.
How come, then, so many travel entrepreneurs rely almost exclusively on hunches to conduct business? I mean, if you have three clients and one of them stops buying from you, I suppose you would be able to establish the reason for that, without requiring deep analysis. If you get a couple of bookings twice a month, there is not much to analyze, is there? If you have at least a few dozen bookings coming in each week from several clients/markets, tough, I wonder how you can be so sure on hindsight what’s going on, without analyzing your data. The less transactions you have, the closer to the truth your self-sustained opinion will be; the higher the bookings volume, the more arbitrary (and mistaken) your judgement will be.
The less transactions you have, the closer to the truth your self-sustained opinion will be; the higher the bookings volume, the more arbitrary (and mistaken) your judgement will be.
I see it all the time: owners and managers boasting a sure-footed attitude towards their trade knowledge, being somewhat disappointed or incredulous whenever they face the cold, hard facts in the form of a properly conducted data analysis. My assumption is that most small to medium business owners care almost exclusively about the financial statements, being too busy with the operational side of things. Perhaps analytics sound too hostile, an unpractical concept that’s better left to academics or IT staff…
If that’s your position, I have bad news and good news for you. The bad news is that you are leaving money on the table, allowing your rivals to catch it with the competitive advantage granted by data analytics put to work. The good news: there are new and affordable tools to assist you with data-driven decisions that require no expertise at all on anything outside your usual tasks.
Repeat after me: data is money.
Now, before trying to sell you such tools, I need you to fully understand why they would be beneficial to your business. Firstly, you have to see the value of collected data: even if you still use paper and pencil in your organization, you’d have surely put together a considerable amount of data (be it from bookings and financials, at least). Repeat after me: data is money. That’s because you can transform data into meaningful information, which will help you take informed decisions and act better strategies to obtain more revenue, detect inefficiencies, reduce costs. In two words, you’d be conducting business intelligence, as opposed as simply acting on hunches.
Allow me to break down the types of data your organization might process: each department would generate a specific data mart, on what’s called an online transaction processing (OLTP). Each OLTP would serve department managers to decide what the best course of action is. Think, for instance, the dispatch officer assigning vehicles and drivers to each day’s arrival and departures: he’ll be using information coming from the daily handling of bookings. However, for that very reason, the dispatcher can’t see the big picture; the general manager instead should be able to access an online analytical processing (OLAP), which puts together all data marts across all departments to get insights from. As an example, the GM could detect that the fuel consumption and general maintenance costs of certain type of vehicle on his fleet is better suited for longer distances tours rather than airport service though city center, thus advising the dispatcher accordingly. How long would it take to see that problem? At most, a couple of hours. How much would you save in dollars or euros? Thousands! And that’s just a small corner of the big picture.
If data analysis is dismissed, only unstructured decisions will be taken; that’s hunch-driven operating, definitely not sustainable in the long run.
Adding OLTP with OLAP, we have a structured data pool, which –after processing- will help managers take structured decisions: that is exactly business intelligence. On the other hand, if data analysis is dismissed, only unstructured decisions will be taken; that’s hunch-driven operating, definitely not sustainable in the long run.
It is utterly important to consider that in today’s hyper connected world, complete and accurate analysis must include unstructured data pools, that is, data coming from website logs, social media, mobile usage, etc. In other words, the so-called “big data”: most operators relying on OLTP+OLAP are taking decisions based on about 10% of total available data (usually generated internally by booking systems, CRMs and ERPs); that is better than 0%, but unstructured data makes up for the remaining 90%!
All right, let’s see what aspects of business in general can be improved by analytics:
- Descriptive: historical data helps you see trends and correlations. For instance, my friend Mister D would have been able to detect why and how there was a slump in sales during a certain period, acting accordingly in time to correct course.
- Predictive: to find correlations and trends across datasets will help establish forecasts, future trends. Mister D’s slump could have been foretold and avoided early on!
- Decisional: builds on both the precedent items, but especially on predictive. Knowing what/why is going to happen, will derive recommended actions. For instance, the DMC owner Mister T was able to launch a successful special offer for the most sought after destination in the relevant markets.
In short, business analytics will help you – among other things- to gain competitive advantage, predict future market behavior and detect operational inefficiencies. Hence, it will allow you to be active and steer course ahead of events, instead of being reactive and follow (too late, usually) the market’s trends.
Judgement, hunches, intuitions… that’s the domain of artists, poets and movie makers. Business is science, actually it’s data science (see article here >> ). But, as I stated above, you don’t have to be a scientist: you just need an easy to use tool that provide a dashboard from where command of your business will be a breeze. Such dashboard must present clearly the information you need to pilot the company’s course with confidence, as a driver on his car. Would you drive through the city center without paying attention to the speedometer? The consequences of doing so in business are equivalent: either you end up paying a large fine, or worse. So unless your organization is a non-profit financed by the mafia, you’d better quit deciding on hunches and start analyzing your data with tools like REVVA.
Thanks for reading, sharing or commenting!
Marcello Bresin