The Southwest One story (part 4) – from hubris to High Court (almost)

Here is the final part of Dave Orr’s excellent analysis of the Southwest One council outsourcing story.  Here, he looks at the fundamental reasons why things went wrong and suggests some lessons learnt. Read part 1 here, part 2 here, and part 3 here

So, what can we learn from this whole sad story?

·                Councils naturally want the most services (at reasonable quality) for the least cost, and conversely outsourcers (at a basic level) want the most income for the least cost of supply. This essential contradiction, means that expressions like “partnership” need to be mutually reconciled, before signing the contract!

·                Southwest One essentially failed because IBM did not deliver on the extravagant promises for transformation and savings, via SAP, that were made to win the business, even though the competing BT bid was substantially lower. (That failure may not be all IBM’s fault, but the savings were not delivered).

·                A good local government tailored version of SAP was NOT delivered on time in 2009 and it was plagued with faults. Continuing problems, with £4.6m of duplicate invoices and uncollected debts, running into 2011/12, fatally damaged Southwest One’s reputation. As a result of that poor reputation, no other public partner ever joined Southwest One, which substantially damaged the growth model underpinning the joint venture.

·                A joining payment by prospective public partners to Southwest One founder partners was another dis-incentive. Also, calling prospective public partners “institutional chauvinists” for not joining (a quote from the increasingly desperate former Somerset County Council CEO Alan Jones) was counter-productive!

·                IBM appears to have over-estimated the overall spend available for potential savings and then calculated that procurement in Councils was so bad, that savings of around 30% on that spend were achievable.  Given that Councils either competitively tender or access pre-tendered national frameworks, was the £192m of savings ever a reasonable estimate?

·                Large scale outsources over a long contract of 10 years or more require an ability to foresee the future that is simply not possible to capture in a fixed contract. In a 10-year contract, there will be three changes of national government and three changes of local government. That is a great deal of unpredictable change to cope with via a fixed, long-term contract.

·                Local Government will always be at a disadvantage in resources and skills to a large multi-national contractor like IBM, when it comes to negotiating, letting and managing a complex multi-service contract. Can a local Councillor (perhaps with small business experience) working with the Council’s Head “Bean Counter”, successfully negotiate a complex contract with IBM?  Or is that a local league team playing in the premier league, with inevitable results in terms of a level contract playing field?

·                What was the culture of Southwest One (75% owned by IBM)? Was it private, public or a hybrid? The management culture remained firmly IBM, yet the Councils and Police workforces were seconded and remained equally firmly public sector rooted. There is such a thing as a public service ethos.  In fact, Southwest One was run like a mini-IBM based upon global divisions, complete with IBM standard structures and processes. Southwest One seconded employees were not allowed anything like a full access to IBM internal systems, thus creating additional complexity, as “real” IBM employees relied entirely upon on-line systems.

·                There was a multi-tier workforce (without a single set of employment Ts and Cs): IBM employees + Councils staff + Police civilian staff + Southwest One direct hires that complicated HR matters greatly and made mixed teams in a single shared service much harder to amalgamate.  And this meant the IBM managers of Southwest One never really gained the sort of command & control of the multi-tier workforces that their bonus-oriented model needs to function. I doubt that IBM would ever again contemplate the seconded staff model over the TUPE transfer model.

·                Councils run a wide variety of complex services that are being delivered at that geographical administrative level, but may have no business connection e.g. maintaining roads and delivering adult social care. So, beware simplistic approaches that simply divide a Council into so called “back” and “front” (or even add “mid”) offices.

·                Somerset County Council ran with a “thin” Client management Team that, in my view, did not have sufficient expertise or enough staff resources to effectively manage this complex contract with IBM.  The Councils relied upon definitions of “partnership” that meant one thing to the Councils’ side and quite another thing to IBM!

·                In Southwest One, Somerset County Council handed their entire IT Service over lock, stock & barrel. Can you really consider IT as wholly a “back office” service? Many successful private Companies see IT as a strategic service to be kept under their own control.

·                Councils are in different places, that come with differing problems, so are meant to be “local” and different! Councils are not a set of Tesco stores to a set format that come in small, medium & large sizes!  How can a one-size-fits-all IT system like SAP transform a whole Council?

·                The real savings might have been found in optimising processes out in big departments (like Social Care, Education, Highways) that lay outside of Southwest One reach. The focus on IT rather than service processes was another flaw in the model.  In the Councils, a Corporate Service Plan is kept up to date every year and an IT Strategy written and updated annually in support of that. Technologies are identified to support the Service Plan e.g. RFID in library books for self-service automation and out of hours return.  In Southwest One, there was no such equivalent. IBM had a Target Operating Model (TOM) but that was expressed in sales terms i.e double the business in Southwest One within 3 years.

·                The lack of strategic planning in Southwest One and any real alignment with the needs of the Councils impeded the innovation and flexibility that the Councils hoped IBM would bring to the joint venture. Once the Government cuts arrived, that lack of joint alignment was painfully exposed.

So looking to the future, are large-scale multi-service outsources with a single contractor, underpinned by a complex contract and a long term, ever viable? I think not.

Is a multi-sourcing strategy with a range of suppliers preferable? I think it is.

Will we ever see the mythical hybrid combining the “best of the public sector” with the “best of the private sector”? I doubt it.

Has the failure of Southwest One had any impact in the recent Council elections or on the reputations of those Councillors involved? Sadly not, but that is another story...

Share on Procurious

Voices (5)

  1. bitter and twisted:

    My pet theory is that its impossible to transforn procurement by implenting SAP.

    You can’t know how you want SAP to be customised until you’re actually using tjhe damn thing. And then its too late.

  2. Chris C:

    Congratulations Dave on a well researched and presented investigation. I look forward to future reports since you are persistent and able to get to the crux of the matter!

    I can’t understand how a customised version of SAP (often an expensive and cumbersome exercise) run by a competitor (IBM) could be expected to work more efficiently/cheaper than many other potential options. Nothing like starting off on the wrong foot! It seems that politicians often just do not fully understand IT.

  3. Trevor Black:

    There are other serious implications in the way this project was railroaded. We should always remember that there are in many public sector organisations where ill considered barking mad schemes are driven forward, senior staff that are side lined (or worse) if they have the temerity to ask basic common sense questions on issues that have not been considered or deliberately ignored. I have also reached the conclusion that nearly all projected savings and KPIs are plucked from thin air for political reasons.

  4. Roger Conway:

    An excellent and clearly expressed summary of this disaster. Living in Somerset I concur with all of Dave’s conclusions.

    I have also been to see the ‘controlling’? committees in action and frankly David is being too kind to those elected councillors who’s job it was to protect the tax payers. As I left the first one of those meetings I was asked my impressions to which I uttered the phrase “the intellectual paucity is staggering”. Over the subsequent 4 years I have seen no evidence that would cause me to change that opinion. In fact quite the opposite.

  5. Dan:

    Thanks Dave, I’ve hugely enjoyed reading this series. Mainly because I don’t live in Somerset…

Discuss this:

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.