Improving the ‘Journey before the Journey’ using Deep Learning

4 MIN READ (distilled from a dozen hours of research)

Deep Learning is simply the evolution of AI and it can help drive revenue, not just drive autonomous vehicles.

Travel planning is a very time consuming and often confusing ‘Journey before the Journey’ for most aspiring travelers. This is true now, more than ever, as we all try to navigate our path forward through this global health and economic crises. When we can safely travel again, it will be our responsibility to make that experience, both pre-trip and during-trip, as good as it can be for your customers.

Online search is the first step for a majority of travelers, with some consumers visiting up to 38 sites before booking a ticket/room/rental property/ cruise/vacation package. The number of digital travel touch-points grows rapidly, as travelers look for better offers via search engines, booking apps, online travel agencies and deal sites. Beyond that, as the graph depicts, the frequency of travel planning is also increasing in great part due to the ‘discovery’ work that is done on our mobile devices.

However, 39% of leisure travelers and 45% of business travelers believe that they use too many websites to find what they are after. Travel Marketers can reduce this overload by serving personalized offers at precisely the right moment. This moment is where AI/Deep Learning can change the game for digital marketers in the travel industry.

Raise your hand if you think all this AI/ML/Deep Learning (DL) stuff is ‘BS’!

You have been hearing/reading/podcasting a LOT about Artificial Intelligence (AI) and Machine Learning (ML) and now, Deep Learning (DL). How do you separate the truth from fiction? Why should you care?

Deep Learning is simply an innovative branch (the latest and most evolved) of AI, a subset of ML, that closely imitates the work of the human brain in processing data and recognizing patterns for decision making.

But wait, AI/DL is not just for autonomous cars and Netflix recommendations - it is able to drive better Marketing outcomes at no additional cost vs. ML tech (which is what your current retargeting partner is using).

A lot of reporting on AI focuses on taking cost and humans out of the equation. A lot more attention should be put on the idea that it can be used to increase revenue and grow your business.

By adding a complex layer of AI/Deep Learning on top of a retargeting engine, users and publishers and product recommendations displayed in banners are selected more accurately which in turn increases the effectiveness of client campaigns by up to 40% as a combination of increases in conversion rates (CR), average booking value (ABV), and click thru rates (CTR).

And all of this is available despite the fact that using DL does NOT increase what it will cost you to run a campaign vs. using ML technology (which is what your current vendor is using).

OK, but why does DL outperform ML?

More data, processed faster, combined with a decision algorithm that is not static and is constantly improving. Don’t make the mistake of presuming that the decision algorithm doesn’t need to change constantly - remember, we are doing 1:1 user-level targeting.

That translates into; 

• Greater reach - on average 50% greater because RTB House chooses a different audience than other solutions (not just because we may have different publisher inventory) and your campaign can participate in more auctions (internal & external)
• Fewer impressions - on average 50% fewer (because we don’t need/want to waste impressions even though we don’t charge for impressions)
• Higher CR, ABV, and CTR - as much as 40% cumulative increase in performance

Why should I believe you are going to outperform my current retargeting partners?

You shouldn’t - you are going to need to test and see the results. And further, we are going to encourage you to keep your current retargeting partners running at the same time. 

Here is why:

This may seem counterintuitive, but we have an enormous amount of data over thousands of campaigns to prove it is true.

Further, we are not driving up the cost of your other retargeting campaigns because the overlap in the Users pursued is relatively small:

Summary:

The ‘Journey before the Journey’ is long and complex and loaded with data points your current display retargeting solution is not collecting or using.

Deep Learning is simply the evolution of AI and it can help drive revenue, not just drive autonomous vehicles.

Deep Learning powered Display Retargeting will improve; Reach, CTR, CR, and ABV (and we have an enormous amount of data from thousands of campaigns to back up our claims).

Deep Learning powered Display Retargeting with RTB House will NOT cost you more to run than your current ML powered solution because we will always honor your goal/limit on CPB or ROI or ROAS or COS.

Deep Learning powered Display Retargeting with RTB House is a very effective complement to your current display retargeting strategy.

As stated at the outset - we are here to build your business, adding RTB House to your marketing strategy is a great way to do just that!