UA Make vs. Buy in PSS System

UA justified the decision to implement SHARES company-wide by saying they owned SHARES and that would give UA the flexibility to enhance and modify the system to serve UA's unique needs.

Often in IT projects there are major tradeoffs in make vs. buy of critical software. The problem of having a unique system is that the full costs of all enhancements and updates to the system have to be borne by a single customer forever. In most cases buying software products whose development costs can be spread over many customers is more economical. In the airline industry, where there are hundreds if not thousands of international airlines, it would seem more economical for 3 or 4 independent companies to create the major components of passenger service and loyalty system software, and for individual airlines to focus on implementation of the systems and perhaps limited customization.

It is interesting to note that over at AA, Maya Leibman, who is highly regarded, has reversed a major buy vs. make decision made by her predecessor in favor of buying software components when possible. The Computerworld article is here http://www.computerworld.com/s/arti..._course_midflight_?taxonomyId=18&pageNumber=1

I wonder if UA will be revisiting its make vs. buy decision at some point in the future.

I'll bite. Despite what I read on IBBs by people that have no idea what they are talking about - why would they?

Make vs. Buy in software products has been an issue for 40 years and isn't unique to airlines. Some of us have experience in areas like software. Is UA an airline or a software company? It's a legitimate issue. It's interesting and relevant that AA, the original developer of SABRE, is making the buy choice.

Are you insinuating that UA doesn't do a recurring cost benefit analysis on all facets of their operation? And that because another airline, in BK, with an entirely different set of variables went in one direction UA should follow?

They've said that Shares gives them the operational and business flexibility they need to best manage their business. I have no idea whether or not this is right or wrong as I haven't seen the data set.

But I have no reason to doubt them.

You also have no reason or data to support them - you simply have no data, nor do I.

Do you really think that the business needs of American and United are so dramatically different that for United the answer was clearly make and for American is it is clearly buy? Or perhaps one of them is not making the correct decision?

The only truly hard number in the decision is the software licensing cost. Everything else is estimates - estimates for how much you will spend on internal development, estimates for potential incremental revenue generated, estimates for losses due to missing features. Many organizations have an internal bias toward internal development and control which ends up not being the right choice in the fullness of time.

If you assume that American's management are equally bright as United's, and that the needs and costs are similar, it is in fact interesting that they have reached a different conclusion. That's the new data that I presented.

Right. Which is why I don't start threads about topics I have no true information on.

AA and UA are in fundamentally different places in their business. Ever go through a BK reorg? Near term cash gen and increasing of free op cash flow often trumps long term business decisions.

Yes, they are both airlines. But if there was a one size fits all solution that was best for every airline then they'd all be running it. But that's simply not the case.

For some reason it seems like folks go on a witch hunt against Shares. I just don't get it.

One thing to take into consideration is that companies often make decisions on their future based on their ability to enable the change. Considering that UA & Continental were merging, the decision in the shorter term tends to be one of selecting a system that only part of the organization has to migrate to in order to decrease risk in the merger.

We'll have to look a few years down the line to see if UA still feels that supporting their in house solution still makes sense for them.

Also look at the recent announcement that UA was going to be "owning" the inflight wifi system, developing their own software instead of using gogo... when you think about the reasons given, this may be one of the smartest decisions they've made. If they want to give their best customers free access... they can. If they want to play with pricing... they can. There are pros and cons on both sides of the develop versus buy discussion. And others are right... different companies have different needs and means.

I recall reading somewhere that LH had developed some software or another, and other carriers were integrating with it as well... I'm short on specifics, but certain that this practice does indeed happen...

I believe LH uses Amadeus. I think they are a part owner of the company that owns, develops and licenses the system. Amadeus was one of the three systems that AA says it is considering.

Cost, while important, is not the only factor. Sometimes the flexibility to do whatever you want is worth an increase in cost. It's a business decision that gets made based on data none of us have access to, thus making this thread somewhat pointless.

A new version of Amadeus was in progress and United was signed on as a partner. They actually bailed on it before the merger was announced IIRC.

SHARES is not a built-from-scratch product that Continental created back in the day. It is a full-featured, commercial PSS built by a 3rd party company just like Amadeus, Apollo, ITA or the others.

The reason United wants to keep SHARES, in addition to the lower recurring costs to operate it, is that the ability to program against it is far faster and cheaper than having to contract out to the 3rd party which created the PSS. Things like adding BoB or bag fees or ToDs or whatever else are faster and easier to implement. At least that's the claim they have made. And either they are lying or they know what they're talking about. Feel free to choose your side there.

Then you've probably made an ass out of u and me.

POTD. :-:

Amadeus came from System One, which is the backbone of SHARES. SHARES works.

"Thousands of international airlines"?

There probably are lots that use a Google Docs spreadsheet and Quicken Home and Business Edition to run their airline

When I was a project manager implementing/designing IT systems every company thought their operations were absolutely unique needing highly-tailored solutions. Truth of the matter was they weren't. Every company has the same basic generic functions as others in the same industry. Once had an art appraisal company that found it hard to believe they needed a simple inventory/pricing system with some minor tweaks for additional pieces of data. I was always of the mind that the SHARES decision was based on what made sense given the merger. Whether to make or buy going forward opens up the internal struggles for turf in the company masquerading as "easier to modify" arguments. YMMV.

I have seen lots of videos on "Border Wars" of those crop dusters that fly drugs over the US/Mexico border....

The "new" Amadeus system (Altea) is a standalone development by Amadeus. Outside the USA, a majority of the major non-LCC players are either already using Altea, have plans to migrate to Altea or are considering Altea for their future needs. It is the future of IT for legacy carriers, but it doesn't come cheap.

I can certainly buy the argument that UA, as the world's largest airline by some measures, can afford to build and develop its own system and that its scale makes doing unique things with it possible.

But can it afford not to develop it and sell it to other airlines? If it's such a good system they could earn substantial income from it because others want to buy it. If no one wants to buy it, then perhaps UA also would do better with a bought system.

As a (very) small business owner with an off-the-shelf EPOS system, I spend my time cursing its limitations. But I keep it because a) I know and understand it with all its limitations and b) I'm fully aware that anything else will have just as many, but different, limitations.

Selling the system to other airlines requires business development, marketing, sales department, and that's just to sell the darn thing. Then they actually need to support it and respond to the other airlines issues as they arise.

And that's assuming that they don't feel that the system provides a strategic advantage and that they are fine revealing some of their business processes / tools to the other airlines.

You are correct. If you have a strategic advantage then do not sell it to your competitors. The cost to market and support may be greater then the income you could get.

Except for Virgin Austrailia and Etihad who (according to linkedin) are going to Sabre. For a while Navitaire was the in for LCC and some of them (Jetblue) have migrated to Sabre.

Navitaire's NewSkies system is still the system of choice for most "pure" LCC operators who can generate high segment volumes and rely heavily on ancillaries to generate revenue. NewSkies is a "ticketless" system using a single database and a "Super PNR". This is ideal for airlines that do not sell outbound interlines, sell overwhelmingly via the web, do not require type-A GDS distribution beyond the major players, have minimal codeshares, etc.. It's major customers include Ryanair, Air Asia, Cebu Pacific, GOL, etc...

Jetblue used the old Navitaire product (OpenSkies) but as their model moved more towards interlining, they migrated to Sabre. Sabre is still a "legacy" system in that it is e-ticket driven rather than ticketless - that has its disadvantages for creative ancillary generation, but allows more flexibility in interlining, codesharing, etc.. Sabre is usually cheaper than Altea on a segment basis and doesn't have the multi-year waiting lists that Altea has for implementation, but its development has been less dynamic than Altea (in my personal experience at least).

In addition, there are a number of smaller/regional players in both the legacy side (SITA, Travelport, Travelsky, Mercator, etc..) and the LCC side (Radixx, AirKiosk, Videcom, TTG, etc..) but none have the combination of volume, features and development that the big players do.