In defence of limited govt IT purchasing


opinion Over at iTNews, seasoned enterprise IT writer Brett Winterford has put together a stinging analysis of Federal Government technology purchasing trends, with particular reference to the Department of Health and Ageing’s … less than convincing recent decision to re-sign its incumbent outsourcer IBM. Quoth iTNews:

Of the $8.35 billion spent on IT and telecoms during the term of the current ALP Government (in purchases above $80K), $3.21 billion (38 percent) of the contract value or 41.5 percent of contracts were purchased/signed via a direct sales process rather than a tender.

This means that agencies chose to purchase IT and telecommunications without going to open tender on almost as many occasions as they followed the guidelines.

What I love about this article is how it demonstrates Health’s hypocrisy in re-signing its incumbent supplier without going to tender — by asking rival companies in a blind test whether they could realistically provide the same services. Shockingly, as it turns out … they could.

Now normally I’d throw my full support behind this article … representing as it does a staunch attempt to hold the Government to account for not following its own regulations. I applaud this kind of analysis, and I think we need more of it. However, for the sake of debate and injecting a more conservative point of view, let’s play devil’s advocate.

Firstly, it should be obvious at this point that it is not the Government’s ability to follow its own tendering process which is broken, but rather the tendering process itself. If you speak to public sector chief information officers off the record regularly, which I do, you will find that most of them find the tendering process unwieldy, slow and overly bureaucratic and that they believe it hampers their every effort to improve the level of services which their IT departments are providing.

The suppliers hate them too — because the cost of responding to a tender request is prohibitive, the procedures are labyrinthine and often you never get a return from your investment.

This simple fact is the reason why Government CIOs will often do anything they can to avoid going through a formal competitive tender when they need to buy new IT equipment or services; because they have the best interests of their organisation at heart, and the tender processes do not.

It’s also the reason why Australian governments — state and federal — are right now rationalising their IT buying processes under a series of whole of government panels — because this alternative option both saves money and cuts the normal tendering process out of the equation by pre-qualifying suppliers.

Secondly, before we slam the Department of Health and Ageing too much, let’s consider the fact that in benchmarking its existing supplier IBM against the market and then signing a new contract for the same services, the much-maligned department was actually following what is pretty much standard practice in the private sector.

Westpac, for example, was recently in exactly the same situation as DOHA — and with the same supplier. After years of problematic relations with IBM that spilled over into the public arena, the bank’s tech czar Bob McKinnon put Big Blue through a series of hoops and eventually ended up retaining its services in a new contract.

It is quite rare for a private or public company to issue a public request for tender document as all government departments are required to do. Instead, the private sector will usually issue a much more limited request for information from a select group of suppliers, and then negotiate further with one or more until it gets what it wants.

This, by the way, was precisely the format followed by government-owned business enterprise NBN Co when it put billions of dollars of technology procurement on the market last year. Why did Mike Quigley not go through the formal government purchasing procedures? Because it would have taken 18 months, and the NBN Co chief simply didn’t have that time to waste.

Lastly, let’s not forget that buying a complex package of IT services over multiple years is not as simple as putting out a request for tender for a different type of desktop PC.

Changing your IT outsourcer, in a major organisation like DoHA, is extremely risky. You need only look at the Department of Defence’s experience with KAZ a few years back to see what can go wrong. After inking a $200 million desktop support contract with the then-Telstra subsidiary in November 2005, Defence then proceeded into hell. Quoth Computerworld in August 2007:

“According to a Defence employee, who requested anonymity, KAZ has dropped the ball when it comes to support levels. “It has been 10 months now since they took over and things have gotten progressively worse,” he said. “I work in Defence and I am telling you that things are absolutely abysmal when it comes to computer support; I am quite sure anyone that you talk to at Defence would agree with me.”

In the context of this risk, the incredible complexity of government purchasing systems and the lack of a real need to change suppliers of what are, essentially, commodity IT services, what option do you think DoHA — and countless other government departments before it — found more attractive? Why, to benchmark your current supplier, beat their price down and address pain points, then re-sign or extend their contract and get the Minister to exempt you from a competitive tender.

Of course, this isn’t always the right answer. But often it is. Sometimes it’s better to bend the rules a little rather than to lose the entire game.

Image credit: Zsuzsanna Kilian, royalty free


  1. I dunno mate – a lot of companies are a bit wary of panels too. ie to get on the panel you have to go thru a lot of effort (identical to a single source tender) – if you “win” you get to compete endlessly for the term of the panel against competitors. Worse, you could be the patsy which has been allowed to “win” in order to provide a price incentive to the actual preferred supplier – committing yourself to years of quote and $0 of return.

Comments are closed.