The IT Sourcing Blog

Contractual Sledgehammers and Management Nuts

2010-08-11T23:09:00.001+01:00

There is a fundamental problem with ICT contracts in the public sector. It arises from the persistent use of contractual sledgehammers to crack management nuts.

The problem, simply described, is this. When things go wrong with IT projects in the public sector, lessons are identified. But, they aren’t learned consistently. Contractual lessons are learned well, since one of the things that the public sector does well is create templates and writing guidance. However, management lessons are not learned so well. The public sector is not too good at firing or demoting people if they don’t deliver great results (or rewarding those that do deliver with rapid promotion and additional responsibilities, or cutting out unnecessary processes and procedures, or… well, you get the idea).

So, instead of solving a management problem by changing management, there is a tendency to deal with it by changing the contract.

- Got problems with risk management? Why not make it the supplier’s contractual responsibility to notify you of potential risks and create mitigating actions?

- Can’t keep control over user requirements? Why not make it the supplier’s contractual responsibility to manage and document your change control process?

- Staff don’t commit to delivering on their obligations and miss their own deadlines? Why not make it the supplier’s contractual responsibility to identify and manage all dependencies?

In each case, the problem is one of management. But the public sector keeps on trying to solve it using the contract. The number, size and complexity of model ICT contracts has grown like topsy over the last twenty years. However, all of these changes have made no difference to the number of public sector contracts that are successful.

As any builder will tell you, you need to use the right tool for the job. Otherwise, it will cost you more and it might not work either.

Sound familiar?

Does Agile Development need a different contractual paradigm?

2010-08-06T11:07:00.000+01:00

I've been involved in some discussions with a group of agile development experts (aka the Agile Development Network) about how they can get involved in more public sector contracts. What's clear from the discussion is that there is a gulf of understanding between software development specialists and procurement people that needs to be addressed urgently.

The argument appears to polarise around two views:
1. The very nature of our work is collaborative and about understanding the requirement as we develop the solution
vs.

2. The very nature of our work is to ensure that we don't spend money unless or until we know what we want; and that we hold our suppliers to account for their performance.

In between there's a lot of confusion, ignorance and suspicion. I think that there needs to be a new way of contracting for agile development to bridge the gap.

The ICT Project Moratorium, and the Law of Unintended Consequences

2010-06-21T12:29:00.002+01:00

As part of its drive to find £6Bn in savings in this financial year, the Cabinet Office has introduced a moratorium on all new ICT contracts above £1M in value, and initiated a review of all ongoing projects to identify opportunities to cut, rescope or reduce the cost. It has also stated its intention to renegotiate existing large ICT contracts to reduce costs, and extend the involvement of smaller companies in the provision of ICT services. Between them, the top 9 ICT providers to the public sector received nearly 60% of all public sector IT spend in 2006. Tens of thousands of small and medium sized enterprises (SMEs) pick up the remainder.

Sounds great. As a very small enterprise, I should be rubbing my hands with glee at the new opportunities that will be opened up as a result.

I'm not, though. The law of unintended consequence applies to pretty much every government initiative, and this one is no exception. The main casualties of this process will be small to medium sized ICT providers, and the main beneficiaries will be larger firms that are its target. Here's why.

For the next 3 to 9 months, pretty much no new ICT contracts will be signed (beyond those where the contract is required to support a critical public service). Across the board, public sector ICT suppliers are already seeing their sales pipelines collapse, and cashflow reducing. At the same time, public sector customers are knocking on the door demanding price cuts of 30% as a starting point for negotiation, on the (usually mistaken) assumption that they have been ripped off for years, and now its payback time. Tesco's treatment of its suppliers will be nothing compared to this.

For smaller ICT providers, this combination has the potential to be fatal. A major cashflow hit is difficult to bear if your turnover is modest, and they lack the negotiating power to resist arbitrary price cuts from their biggest customer.

Larger providers are in a much better position, even if it doesn't feel like it. Their contracts are usually long term, with high termination costs, and they will generally have the financial strength to survive a short cashflow problem. They also tend to be delivering the most critical services - the ones that, for all the bluster, public sector customers simply cannot afford to put at risk.

So, when the negotiations start, it will be the smaller firms that are forced to accept blanket price cuts, reductions in scope and long delays before new orders are placed. The main effect of the moratorium will be to drive SME's away from the public sector, or out of business altogether.

Larger providers will sit out the storm, knowing that this time next year many of their smaller, more nimble competitors will no longer be there.

All quiet on the IT procurement front

2010-03-16T17:07:00.001Z

Things are still in the world of public sector procurement at the moment. The end of the financial year usually heralds a sudden surge in expenditure, as government departments try to work off excess budget allocations before they lose them altogether. But this year, things seem to be different. A quick search on SIMAP reveals that the total number of contract notices was down by 10% in February 2010 as compared with Februar 2009 (1,007 to 909)

It is the opposite of a perfect storm: everything is contriving to put the brakes on market activity.

The election (of course) is the biggest factor in people’s thinking. Although the dates haven’t been announced, everyone knows that it will happen on or about May 6^th.

This, on its own, is usually sufficient to cause a delay in making any major purchasing or project initiation decisions. But this is being exacerbated by genuine uncertainty about the outcome. A year ago, Gordon Brown was a lame duck prime minister who couldn’t get a price on an election victory. Now, whilst he’s hardly the favourite, there is a real possibility that the electorate may choose to continue with him (or, even worse, give us the first hung parliament for over a century).

In some areas, such as the National Programme for IT, the future of the programme itself hangs on the outcome of the public vote. This would be a great shame since, having spent all this money to build the infrastructure for National Care Records, it flies in the face of reason to cancel the part of the budget that will actually start delivering real benefits for patients. Faced with this, it is almost impossible for public sector managers to set priorities for any investment beyond the essentials required to keep public services going.

Finally, there is the uncertainty over future funding for government programmes, regardless of who the government turns out to be. I say uncertainty, but of course I mean the opposite. Everyone is perfectly certain about future funding: there will be a lot less of it.

So, what’s going to happen to public expenditure this summer? I think that things will stay quiet for quite a while after the election. But things are likely to pick up in the winter, particularly in the IT sector, for a number of reasons:

To deliver real efficiencies in a business, you almost always need to invest in improved systems and processes.
Pressure on IT expenditure has been depressed for a number of months now. There is pent up demand that will need to be satisfied
Whatever is said during the course of the election, the reality is that real, actual cuts, are impossibly difficult things for politicians to deliver. Quite apart from the unpopularity of cuts in services, most politicians didn’t choose their career in order to do less. Instinctively ,they want to do more, and that means more projects and more procurement
The green shoots of recovery, albeit slow growing, will start to do what Gordon Brown cannot, and drive down the public sector deficit

Tories promise to learn from government IT disasters - How Exactly?

2010-03-09T14:05:00.000Z

http://www.computerweekly.com/Articles/2010/03/03/240492/Tories-promise-to-learn-from-government-IT-disasters.htm

I was fascinated to read the Computer Weekly article on a recent interview with Stephen O’Brien, the shadow health secretary. One quote in particular caught my eye:

"..I'm very confident the one thing we can do is learn the lessons from experience… I am confident that our process will minimise the chances for what are design errors."

It all sounded terribly familiar, since the same line has been taken on government IT projects by the current administration and the one before it. The Office of Government Commerce published a fascinating report entitled “The Common Causes of Project Failure” back in 2005, where it set out a series of checks that should be implemented to maximise the chances of successful projects. It was part of a long trend of reviewing past failings and adding further controls and quality standards to try to prevent them re-occuring.

You would think that after all these lessons being learned, government would have started to get quite good at IT projects, wouldn’t you?

However, the problem runs too deep to be fixed by documenting lessons and adding another layer of bureaucracy onto future projects. For that matter, one of the problems that is now emerging is that projects are drowning in a sea of paperwork – too busy ticking boxes to actually deliver anything meaningful.

The problem is one of incentives. Ask a senior civil servant what is most likely to further their career and the list of factors will include things like: “a broad range of skills”; “experience of working in different government departments”; “good networking and relationships”; “a reputation for sound judgement”; “experience in developing policy”.

Delivering IT projects on time and to budget doesn’t figure very highly on the list.

In fact, it is rare even today for the SRO (Senior Responsible Owner) to be in post for the entire duration of a project. Project success or failure doesn’t blight a career in the same way that it would in the private sector. Until it does, you can write all the lessons learned reports that you want; the results will still be the same.

The customer has a right to be unreasonable, but not unfair

2010-02-26T13:14:00.001Z

Recently, I was working on a project where the customer chose to extend the deadline for receipt of bids. The reason that was cited was a period of bad weather during the tender preparation period, combined with a recognition that the original timetable was extremely tight. The extension occurred very late in the tender process (with only a week to spare), and triggered a debate within the bid team about whether to put in a letter of objection.

For the bid team, this was anything but good news. From a personal perspective, they had put in some herculean efforts in order to be ready to meet the original deadline. You could tell just by looking at their faces that the idea of another week of long nights and weekend working was unwelcome news. From a competitive perspective it did them no favours - they believed that the competition were more likely to benefit from extra time than they would.

Now, regardless of EU procurement rules, most customers are unreasonable in the way that they construct their procurement processes. I have a vivid recollection of developing a procurement plan on behalf of a customer - initially it implied that the main bid evaluation work would be conducted over the Christmas break. One short review later and guess what? The plan was adjusted so that it would be the bid teams who had their holidays disrupted.

This is normal - most bid managers will tell you that the runup to Christmas and Easter is marked by a flood of requests for proposals.

But there is a difference between unreasonable and unfair. When bidders are put under extreme pressure to deliver to a short deadline, and pull out all the stops to do so, it just isn't right to change the rules on them.

How many staff are really in scope?

2010-02-17T10:38:00.001Z

Its become traditional in most government departments to fill vacancies in the organisation with interim staff and contractors. So traditional, in fact, that some "temporary" workers have been in place for years and years and are indistinguishable from the civil servants that they sit next to.
Most contractors have no desire to be treated as employees, largely for tax reasons but also for reasons of independence and control. The businesses that use them have no desire to offer them the benefits that a full time employee would receive. But, no matter how much the business, or the contractors themselves, seek to avoid it, employment status is not a matter of choice. Its a matter of fact that, rather annoyingly, can only be established by an Employment Tribunal.
This becomes a problem when defining the scope of an outsourcing contract. TUPE regulations provide a measure of protection for employees who are in-scope. The same protection does not apply to third party suppliers of goods and services, such as contractors. But where you have contractors who have become part and parcel of the service being outsourced, the dividing line between the two can be difficult or impossible to draw. Organisations can find themselves with liabilities for employees that they didn't even know they had. Suppliers will, understandably, want to seek indemnities against any employment claims arising from contractors who feel aggrieved at the outcome of the outsourcing.
So, if you are working on an outsourcing or retendering project, are you sure you know how many staff are actually in scope?
(TUPE=Transfer of Undertakings Protection of Employment)

Writing a Winning Tender : Rule 3

2010-02-08T13:57:00.000Z

Rule 3: Don't Lie

This rule was brought into sharp focus by the recent court case between BSkyB and EDS, where the judge set aside the limits of liability in the contract on the basis that EDS had fraudulently misrepresented themselves when tendering for the work.

The temptation to lie (euphemisms: "emphasise the positive", "gild the lily", "gloss over that area", "be creative"), is inevitable when bidding. Every bid manager with a sales target and a mortgage to pay has the urge to push the limits of the truth a little bit in order to make their pitch that little bit more enticing. Surely, that's just about putting your best foot forward?

Well, maybe, but what's the difference between putting your best foot forward and lying? The embarrassment test: if you would be personally embarrassed by having to explain in detail what you did and why you did it, then its a lie. I am put in mind of certain Members of Parliament who would have benefitted from applying the embarrassment test to their expenses claims. They may not technically have been lying, but once the full facts were available, they may as well have been.

People lie because they think that the truth is unsatisfactory - and in bid terms that means that there is a weakness in their offer that they don't want the customer to know about. If its a real weakness, then it will come out in the end: through failures to deliver on the contract, cost over-runs that erode the profit margin, damage to the relationship or all three. Business won on the back of deceit is always bad business.

Writing a Winning Tender : Rule 1

2010-01-26T13:43:00.002Z

I’ve been doing quite a bit of work in the last year helping organisations to prepare tender responses, to the extent that I accidentally ended up being perceived as a bit of an expert in tender writing by a client.

This was a great example of the way in which other people can have a totally different perception than the one you expected. I thought of myself as a contracts and commercial expert with a good line in strategy development. They thought my biggest asset was an ability to write complete sentences. Such is life….

However, I have concluded that there are, in fact, only 2 rules that you really need. Both are simple, easy to remember, and rarely followed.

Rule 1: Answer the question

Rule 2: Always remember Rule 1

It is, of course, possible to add dozens of further rules to the list, relating win themes, hot buttons, credibility indicators and pretty much any sales buzzword you care to mention. But, the awful truth is that when I receive tenders from bidders, despite some wonderful graphics, excellent quality paper and outstanding use of the word “transformation”; the majority of them do not answer the question.

To win, you do not need to be perfect, you just need to be better than the competition. And most of the competition do not seem to understand Rule 1.

(Reader’s of Terry Pratchett’s excellent book Thief of Time will no doubt spot where I got the idea for this piece from – an attribution I am happy to acknowledge)

Are you going to have a meat free 2010?

2010-01-24T22:56:00.000Z

I had the pleasure of reading the new OGC guidance on the Remedies Directive over New Year (which says more than I would like to reveal about how much fun I was having).

Surprisingly, it turned out to be pretty interesting. Of most interest to me is the increased levels of disclosure that are required for unsuccessful bidders. The guidance makes clear that unsuccessful bidders should be provided with characteristics and information about the successful tender, along with scores.

I think that one of the effects of this will be to push procurement teams away from most economically advantageous tender ( what other kind of meat is there?) and back into the arms of the lowest compliant offer. The reason? Under MEAT evaluation, a customer who selects a higher priced offer would need to justify it on the basis of the additional value being delivered. This, inevitably, requires a financial value to be ascribed to the subjective differences between the bid. That's always been the case, but the difference now is that unsuccessful bidders will have access to far more information on which to decide whether the customer's judgement is reasonable.

And about time too. People are notoriously bad at making comparisons between subjective criteria and cash. At times, one also gets the impression that MEAT is used to pick the "people we like" even if it costs the taxpayer more.

So, we may be looking forward to a different kind of vegetarianism this year....

The great government asset sale approaches

2010-01-05T16:47:00.002Z

Roll up, roll up, the great government asset sale approaches. Across the public sector, departments, agencies, councils and every public organisation in between is doing the equivalent of searching down of the back of the sofa hoping to find an old building that they can sell off to raise a bit of money.
You will know the reason why, of course. The public finances aren't in great shape, and aren't going to get any better. So, in anticipation of spending cuts to come, smart departments are reviewing their asset base to identify anything that they could sell off to raise a bit of cash.
This pattern extends to IT - two years ago every project I was involved in was about building new systems and capabilities. Now I'm noticing more and more interest being expressed in projects that involve selling off IT systems and IPR to the highest bidder. I sense a new trend in commercialisation coming over the horizon.

SNOMED is the most high profile example of this trend, but its by no means an isolated case. I think that you can expect a whole raft of initiatives to sell off parts of the government IT estate next year and the year after.
It will certainly make a change from the government trying to buy big systems from the private sector.

Feedback and Experience of the Competitive Dialogue Process

2010-01-03T21:10:00.002Z

One of the hotter topics in public sector procurement at the moment is the use of the Competitive Dialogue (CD) procedure to place contracts for complex services like IT. For those who don't know: Competitive Dialogue was introduced at the end of 2006 by the EU as a new procedure to control the way in which public bodies procure goods and services.

Now that its been around for a few years, there's enough experience building up to enable us to compare CD with its predecessor, the Negotiated Procedure.

I recently helped a major service provider bid under Competitive Dialogue, and captured these lessons learned from their experience:

- is the structure right?

The general structure of competitive dialogue (a draft specification, followed by dialogue rounds, followed by an updated specification and invitation to bid) is helpful, and consistent with the established practice that was developed for IT procurements under the negotiated procedure. However, customers and suppliers would benefit from more specific guidance (from OGC and other industry practitioners) on the circumstances under which an outline proposal stage should be incorporated into the dialogue process. This will help to enable bid/no-bid decisions to be taken earlier in the process where a bidder’s technical solution is likely to be substantially behind the competition.

The use of plenary sessions with all bidders present for the customer to present periodic updates and broadcast information was helpful, and should be encouraged as good practice.

- is the level of resource required appropriate?

If well managed, then the resource load is not dissimilar to a negotiated procedure. In particular, the opportunity to develop and test elements of the solution prior to submitting a proposal enables resources to be allocated efficiently by both bidder and customer. However, there is a tendency among some customers to allow “Dialogue Drift”, moving the topics under discussion away from the immediate scope of the proposed contract and onto other potential areas of business. Whilst superficially attractive, this has the effect of confusing the market about the customers priorities and the potential value of the business being offered.

- do suppliers get value out of the process?

Yes, with the proviso that, for understandable reasons, the customer is usually concerned to maintain control over the agenda and timetable for dialogue. As a result, it was sometimes difficult for bidders to undertake a truly meaningful discussion on the topics that formed the heart of their solutions.

- do clients provide bidders with the right materials / data?

Yes, but often to the wrong timetable. There is a fundamental misunderstanding amongst customers about the time it takes for bid teams to sort, analyse and interpret data in order to develop their solution. Repeatedly, information will be provided on a Friday with the expectation that bid teams will be able to present both a cogent analysis of the data, and their proposed solution against it, in the following week. In addition, data is often drip-fed to bidders, rather than presented as a single coherent package. As a result, there is a high probability that data will be misinterpreted, or that bid teams will struggle to identify the dependencies between different data sets that have been released.

Customer organisations would benefit from more guidance on the time periods that should be allowed for bid teams to digest the information provided to them, prior to the commencement of dialogue. In addition, where an updated bid documentation set is to be issued prior to submission of tenders, then sufficient time should be provided for bidders to assess the changes that have been made prior to submitting their offers.

- do bidders resource the dialogue process appropriately?

Not always. Bidders often underestimate the overall resource load that the dialogue represented, and as a result become distracted from resourcing the development of the solution itself.

Primacy and Recency Effects in Evaluation

2009-12-30T20:26:00.005Z

I regularly get involved in helping with tender evaluation and downselection decisions, attending interviews and supplier presentations and moderating the evaluation scores from the customer team.

Tender evaluation processes need to be managed against a consistent set of scoring criteria, and have to be applied in the same way to every bidder. These are basic rules of fairness that you would expect in any competitive process, and in my experience customer teams go to great lengths to keep to them.

However, one of the areas that receives little attention is the importance of the sequence of the evaluation of bids. This can have a major impact on the scores that different bidders receive.

The reason for this is a psychological effect called primacy and recency. When we process information, for example when evaluating a tender, we tend to remember things more easily if they are presented to us earlier in the sequence than if they are presented later. This is called the Primacy effect.

In a similar way, we also tend to remember information that is presented later in the sequence than if they are presented in the middle. This is called the Recency effect, and reflects the fact that we find it easier to remember the most recent event than one that happened longer ago.

You can try this for yourself using this simple experiment. Read the following list of objects once, then look away and try to write them down.

Hat, Chimney, Cow, Monitor, Pavement, Duvet, Fingernail, Clock, Potplant, Football, Sherbet

Chances are that you found it much easier to remember objects at the start of the list, or at the end, than you did for objects in the middle.

Primacy and recency effects are important in tender evaluation because they are universal (everyone is subject to them) and enduring (so, they apply whether the evaluation process lasts an hour or a week). They mean that the evaluation team will tend to remember details of the first and last bidders proposals more easily than the proposals that they saw in between. As a result, all other things being equal, a bidder who is seen first or last is likely to be scored higher than those in between.

In practice, of course, there are many other factors that affect the scoring, not least of which is the quality of the bids themselves! However, there are a few things that you can do to mitigate the effects of primacy and recency in the way you manage the evaluation process:

For written submissions, rotate the order in which different evaluators read and score the responses, so that on average no one supplier is read first or last. Most people's default approach is to read responses in alphabetical order, which gives an inherent advantage to companies whose name starts with an "A"!
For presentations, ensure that evaluation scoring is done during or immediately after each presentation, rather than at the end. This reduces the risk that the team will forget details about the bidders in the middle of the sequence.
When moderating the scores, consider starting the review with bidders who presented in the middle of the sequence. This forces the team to recall details about the middle bidders first, and reduces the risk of interference from the primacy and recency effects.

For a bit more information on primacy and recency, you can see this link http://en.wikipedia.org/wiki/Serial_position_effect#Primacy_effect.