Advisory: Airfare offers are coming up. Be prepared!

In the beginning of this month, we wrote a post on an experimental estimation method for predicting the timing of airfare offers. According to the post, we mentioned that there wouldn’t be an airfare offer in the month of September. And this has been proven right so far.

When is the airfare offer expected?

We had also estimated that the sale period would start around 3rd October 2017. However, given that Dussehra falls on 30th September this year, there is a fair chance that the sale will be declared a bit earlier than what we had previously expected. We are now estimating the earliest start date of the sale to be around 26th September (coming Tuesday). There will definitely be an offer available before 6th October.

What are the travel dates for the offer?

As mentioned in the previous post, the upcoming sale may have very little to do with flights taking off in the month of October. Flights for almost all of the days of November until mid December are the ones expected to be going on sale in the upcoming offer.

How to get the airfare offer details?

As has always been the case, we will send out an email notification to all the users of Unfare as soon as we find an offer available in the market. So if you have not yet created a free account, do create one here so that you don’t miss out on these airfare offer notifications.

Predicting the dates for the next airfare offer (no sale in September?)

Predicting when exactly a airline would declare an airfare offer is like predicting the rain. Its complicated. But, we are finding that it can indeed be predicted to a reasonable extent. Earlier this year, we were able to establish a threshold, when crossed, increases the expectation of an airfare sale and we called it the UnfareScore. However, it does not tell us the exact time frame when the sale is going to happen.

With the huge amount of airfare data that we now have at Unfare (http://unfare.in), we are now able to build certain logical models that can reasonably predict the time frame of an airfare sale. While this is still work in progress, we wanted to publish our estimates in advance.

No fare sale expected in September:

Airfare sales happen around a month before a period of low demand. The sale is meant to stimulate the demand. The month of October is a high demand period, with many festivals (Durga Puja, Diwali, etc.) and hence there is no incentive for the airlines to discount the fares. Air tickets will get sold by itself thanks to the high demand in the market. Hence, we do not expect any meaningful airfare sale to happen in the month of September.

Fare sale expected to start in October:

Airfare sales are expected to be back right from the first week of October. Why is it so? Just like we identified October (high demand period starts 22 September) as a high demand period, the month of November is a low demand period and airlines will have to discount the fares in order to sell the tickets. Currently, we have data only till the end of November and hence December demand estimates can only be made at a later date.

Looking forward:

While the prediction of the airfare sale period is still a new thing for us, we are confident that we will be able to quickly learn from our predictions and update it back to make the model better. Starting next year, we will also be able to cross check the previous year’s sale period and hence further fine tune the sale predictions.

There are two important questions that we are trying to find answers for:

1. When is the fare sale expected?
2. Which dates will get discounted fares during that fare sale?

Our estimates:

1. Fare sale is expected around the first week of October (3rd or 4th Oct mostly)
2. At the moment we can estimate that flights for 7-9 November are surely going to get discounted (obviously, there are more, but we need a few more days to be sure of that). Generally, flights for the month of November could get discounted.

Big disclaimer: This is the first time we are doing such an estimation. Unless we get this right a few times, we will not make a recommendation based on this model. Once we validate it, we can publish such estimates ahead so that you can plan your ticket purchases accordingly.

Just like predicting the rains, this is one prediction, if made right, is going to benefit a lot of people. We are excited!