Tour & activity API architecture for distribution is more important than it appears

February 1st, 2018

by Alex Bainbridge

As I was reading back through the tour & activity sector overview white paper I wrote end of 2013 and published last week I came across a few paragraphs outlining the two approaches that could be taken by tour & activity retailers (e.g. the leading online travel agents) when connecting to supplier systems via API to enable connected distribution.

Why connected distribution?

Connected distribution is important because this gives live tour availability data to retailers.

Live availability data is vital when customers are booking on the day or booking low availability tours as otherwise customers book on sold out tours causing significant customer service issues and hassle for the customer, the supplier and the retailer.

Additionally it is efficient to transfer bookings from the retailer to the supplier via API so that availability can be immediately reduced on the supplier system (without human input) and so that the ticket or voucher can be activated and used immediately.

The industry wide design decision was on HOW to achieve this connected distribution, not whether to connect retailers to suppliers using an API. Everyone wanted connectivity

How could connected distribution be achieved?

The two options (as I outlined to retailers 2013, using the text from that time):

[A] Publish an incoming API data standard
Growing in popularity is the concept of an online travel agency publishing an incoming API format and letting suppliers develop against that. Initially this looks ideal to all parties but there is a potential side effect:

The data standard tends to be lowest common denominator. i.e. better data such as provided by TourCMS will exist in API format that will not be fully consumed by the incoming API. Many fields will not be provided.

Startups and OTAs with a desire for the best data will therefore go straight to the supplier source for their data. They will create better services and the company who created the reduced capability incoming data API service will just look sadly lacking in data by comparison.

[B] Use the suppliers outbound API
Companies such as TourCMS can provide extensive data via API using a uniform data standard for hundreds of companies.

The winning OTAs will therefore be those who are the most innovative in their use of the data, just like innovation on data has created winners in the hotel and the airline sectors.

What was decided and why?

All retailers independently chose to create an incoming API format that all suppliers wanting to connect had to use [A].

There wasn’t really any discussion about these two approaches – the supplier systems were just told how it was going to work 🙂

I expect this was simply a resource decision (available developer time).

When you compare the technical work required to build and maintain one API and then let 3rd party suppliers develop against that….. and compare that level of work to connecting to potentially hundreds of supplier systems all with slightly different API data formats – you can see why retailers took the route they did.

Why is this interesting in 2018 when the work has already happened and the decision was made years ago?

It is interesting because some of the differentiation challenges that retailers and supplier reservation systems are struggling with TODAY stem from this simple retailer made architectural decision many years ago.

Revisiting the options in 2018

[A] Publish and incoming API data standard

A very simple set of data fields – the lowest common denominator approach

  • Easy for any new reservation system entrant to connect to the online travel agents / retailers
  • Suppliers think they can build their own reservation & distribution technology “with a couple of developers” as connectivity looks simple

As such, there has been a proliferation of supplier reservation systems and technologies. Is this a good thing?

I suggest that at least for the retailers, this proliferation of supplier reservation systems is a long term problem. Technically yes they are easy to manage as the retailers only have to manage a single incoming API so supplier system proliferation is not hurting there.

However because there are so many supplier reservation systems they are then forced to differentiate between themselves in ways that are unhelpful to retailers:

  • Non retailer distribution capability – e.g. helping tour suppliers trade directly with consumers (i.e. without suppliers paying retailers any commission)
  • Connecting to more and more retailers – a fragmented retail landscape is good business for reservation system providers as helps makes the case to suppliers that they need a reservation system to manage all their connected distribution

i.e. supplier reservation systems are not inherently incentivised to help specific retailers.

Additionally retailers themselves are struggling to differentiate between themselves. They all have the same basic availability data from suppliers (and extranets!)

i.e. the industry is effectively neutralised with no actor incentivised to help another

[B] Use the system suppliers outbound API

Resource heavy development project for retailers

Going this route would have been a big call. The friction was high but the reward was also high for retailers.

Fundamentally, the question is, how does a retailer win the tour & activity market?

  • having the most supply?
  • having the most detailed tour description content?
  • the ability to best match a tour with a customer during the research phase leading to winning consumer trust and loyalty?
  • enable the supplier to best explain their product and differentiate from other suppliers in the same destination leading to greater supplier support?

OK so which is more important is not settled. However all four strategies are easier to execute with deep data from supplier reservation systems, not just the basic availability data that is currently connected.

Key points:

  • There will be winning supplier reservation systems (and losing supplier reservation systems) as it will be more obvious which systems are more capable at helping their suppliers with great connected distribution
  • Innovation will happen at the supplier reservation system level that will directly help their suppliers work better with the retailers. With option [A] there is no distribution innovation
  • Supplier reservation systems will have business exit opportunities to retailers as they will be able to show they are better for that retailer than other systems, rather than being just another system on a long list all who create the same data (as currently with option A)

i.e. this route is a riskier, winner takes all, outcome for both retailers and supplier reservation systems. It would have locked in the early mover advantage that the early retailers and the early reservation system providers enjoyed a few years ago.

A new option [C] combining the benefits of [A] & [B]

Opportunity here for a third option that I didn’t consider a few years ago when outlining the two original options.

Continue with the incoming API concept [A] as this is easier to manage for the retailers and we are all collectively several years into this industry wide wiring project now. However massively increase the number of fields that these APIs support and their general capability. Move away from lowest common denominator and move to total content connectivity and functionality that enables tour personalisation.

This will deliver the innovation platform benefits of using the supplier APIs [B] as well as the efficiency of using the retailer incoming APIs [A]. It will reduce the proliferation of supplier reservation systems and incentivise supplier systems to innovate on content provision to the retailers.

More about what I would do as a retailer in the next post next week. Are you subscribed? 🙂

Photo: data (scrabble) by justgrimes (CC BY-SA 2.0) (Source)

This content is protected by copyright. Link sharing is encouraged but duplication and redistribution is illegal

Comments

5 responses to “Tour & activity API architecture for distribution is more important than it appears”

  1. Interest reading, thanks Alex.

    Touches on a question that I keep asking: are we sure that the drive for ecommerce / connected distribution is necessarily a positive trend for the entire T&A segment?

    Single day / in-destination, yes – of course.

    But how about with multi-day and custom itineraries? There’s an important human component to that customer experience that connected distribution is commoditising out of the picture.

    The multi-day operators I work with are resistant to OTAs, as the human interaction is their entire raison d’etre. But perhaps they’re swimming against the tide of inevitability? Is the market just changing over time?

    • Alex, great read and probably the most interesting debate in terms of where the bottleneck stands for future developments of the industry as a whole.

      Matthew, for Multi-day: I would say that in that case, the APIs should revolve around the communication. Platforms like Evaneos are essentially a messaging + payment platforms.

      Actually when you look at the full service provided by the OTAs, most of the reservations systems have been working around the content and process the booking part, however there is a part of the service where you communicate directly with the client (in the tailor-making process), there hasn’t been an “message aggregation” API out there yet.

      • Pieterjan Behaeghe says:

        @Thomas : You mention that Evaneos is essentially a messaging / payment platform.
        But do you have insights on the DMC’s selling their ‘products’ on Evaneos, ho they manage their bookings and customer relationships ? Do they have other channels where bookings are coming in or are do they only have one channel, Evaneos ?
        Thank you for your insights.

    • Alex Bainbridge says:

      For multi-day, this is a big question. Retailers could be very helpful to the sector and enable a move away from PPC advertising etc – but the challenge is sales volume of these hard to sell tours – a single human travel agent who sells a particular tour operator once every month won’t get any real knowledge of the tour company – and will therefore pass on the harder customer questions to the supplier…… and still take the commission. Suppliers don’t really see that as a step forward!

      Digital distribution could work – if I was doing this I would be video heavy as a retailer ensuring the supplier creates that human connection to the customer that way!

  2. Great read. Thanks for this !