In my last post I outlined interim results from the Cost Management Pilot Scheme currently underway in the Technology and Construction Court and Mercantile Courts.  The pilot scheme is due to run until 30 September 2012.

As part of the pilot the parties’ legal representatives (in practice, solicitors) must complete a detailed cost budget form – Form HB – and exchange with the other side when filling initial case management information.  Parties, actually the legal professionals not the litigants themselves, are required to discuss their budget with the other side during the budget building process and continue discussing budget estimates before significant stages of a case (ie Case Management Conferences, Cost Management Hearings, Pre-Trial Review and Trial).

The Courts will have regard to any cost budgets filed when considering whether to make a Cost Management Order (CMO).  If a CMO is made then revised budget estimates must be filed and clients informed in writing of the CMO.  Legal professionals must thereafter use the CMO budget as the baseline, and be prepared to explain to the Court reasons for any significant changes to the CMO baseline budget.

During the pilot scheme to date, CMOs have been made in the vast majority of cases.  Of 31 reported cases 28 had CMOs made, a “CMO rate’” of just over 90%.  The interim pilot report does not say whether this ‘CMO rate’ is in line with CMOs made during the litigation process generally, but I suspect it is considerably higher than the norm.  Some solicitors expressed concern about providing cost budget estimates early on during the litigation process which may then be later used by judges to make CMOs possibly to the detriment of their clients.  One solicitor suggested that Form HB should have a “range of figures”  in respect of the categories of work required to reach trial, as this would then make some lawyers feel less worried about giving figures which may appear to be more precise and accurate than they actually are.  Another solicitor also noted that Form HB does not in its current form allow the reference to the issues to be considered during trial.  Listing cost estimates only does not help to convey the complexity of the factors needing to be weighed up when considering estimates.  It was felt that a brief summary of the issues would flag up that the estimates provided were subject to a considerable degree of uncertainty.

Providing estimates is a vital part of any properly managed project.  As any experienced project manager will attest, collecting and presenting meaningful estimates is always a challenge.  The people best placed to provide the estimates are those who will actually be doing the work and who have some experience of working on similar projects before.  So in a software development project for example when estimating the length of time to code the software the project manager should get meaningful estimates from the software developers, not the Business Analysts, Technical Authors or anyone else involved in the design and delivery of software.  I note that as part of the Costs Pilot while most HB Forms were completed by solicitors of more than 8 years’ experience, in some instances the estimation work seems to have been done by much more junior solicitors, trainees or paralegals.  I’d suggest this is not good practice.

No-one likes giving estimates which may at some stage later “be used against them”.  Returning again to software development, the classic scenario is where the development team give an estimate for completion, only to find that Sales and Marketing treat those estimates as fixed, certain dates and are presented to customers as such.  The rest is easy to imagine: the ‘release date’ is missed, customers are disappointed and the business managers complain about ‘lost revenue opportunities’.

The whole point of Project Management of course is to be able to manage projects in such a way that scenarios like the one above are avoided, or that at least any damage caused is minimised.   A starting point is to recognise and acknowledge that estimates of any kind are just that – estimates.  As such they are subject to uncertainty.  Research and experience over different disciplines confirms an obvious, but often overlooked, point: estimates become more accurate, and those providing the estimates become more confident about them, as a project progresses.  This is explained particularly clearly by Steve McConnell in his brilliant book “Software Estimation – Demystifying the Black Art”.  Steve’s book illustrates the notion of estimation uncertainty by reference to the “Cone of Uncertainty”.  This is also explained at the web site of Construx a software development consulting organisation set-up by Steve McConnell.  The first and third diagrams below are reprinted by kind permission of Construx.

The Cone of Uncertainty is a graphical representation showing that estimates are subject to predictable amounts of uncertainty at various stages during a project.  The horizontal axis is made up of project milestones and the vertical axis shows the typical degree of error found in estimations by skilled estimators at various points during a project lifecycle.  In the context of software development the Cone of Uncertainty looks something like this:

Substituting litigation stages for software development stages may result in a cone of uncertainty looking something like the one below.  I am not suggesting that these are the litigation milestones (Pre-Action, Case Management Conference, Disclosure of Evidence, Exchange of Witness Statements, Listing and Trial) which have to be used, they are suggestions and I am sure that more experienced litigators than me will have their own suggestions.  Nevertheless this should suffice for illustrative purposes.

My plotting and drawing skills are not very good, but I hope you can see the point I am making.  Another thing to note, and which my extremely basic litigation plot does not convey particularly well, is that estimation accuracy typically improves rapidly for the first 30% of a project.

It seems obvious that the levels of uncertainty narrow as the project progresses.  In fact, it is not obvious at all.  The cone does not narrow of its own accord.  A project has to be managed in such a way that variability’s and uncertainties are driven out as much as possible.  A poorly managed project means that any estimates do not become more certain as the project progresses.  Rather than a Cone of Uncertainty, one is simply left with a project “Cloud of Uncertainty” which means that estimates can be subject to even greater error somewhere within the cloud as illustrated in blue below.

Steve McConnell points out that estimators cannot produce estimates more accurate than is possible at the projects current (at time of estimation) position within the cone.  As Steve puts it “it isn’t possible to be more accurate; it’s only possible to be more lucky”.

Returning to the cost pilot scheme, when drawing up CMO’s in reaction to cost estimates given by solicitors, did the Judges take into account at which point in the Cone of Uncertainty the estimates at issue were provided?  I suspect that at one level, unconsciously or intuitively, they did, but if I were a litigant I would feel more reassured knowing that Judges and Solicitors have received adequate training to familiarise themselves with concepts such as these.

It is also interesting to note that some solicitors suggested including “legal issues”  – effectively “complexity factors” – on the Form HB for reference.  In order to help manage the variability associated with the Cone of Uncertainty Steve McConnell suggests having one person create the “how much” part of the estimate and another person create the “how uncertain” part of the estimate.  In litigation this may translate as having a Cost Draftsman do the “how much” with a Solicitor doing the “how uncertain”’ (ie legal issue variability) part of the estimate.  Some may object that this would increase the cost of preparing the case budgets but the pilot report authors note that, in London at least, costs draftsmen were being used in addition to solicitors to produce the Costs Budget anyway.

The task of estimating project duration and cost is fundamental to good project management.  It is a hard task to perform well, despite the wealth of literature, estimating tools and guiding methodologies offered by project management.  Project Estimation remains a fascinating subject nevertheless and I have barely been able to scratch its surface during this post.  It is a subject of rising importance to lawyers as their clients and the courts come to require estimations of increasing reliability.