Cure for Service Failure?

15 February 2007

Software developed to help Addenbrooke’s Hospital manage performance of soft services contractors is made available at pre-bid stage for potential contractors to calculate their risks, costs and penalties in order to reduce the potential for service failures

THE UNIVERSITY CITY OF CAMBRIDGE sits at the southern edge of the Fens. Visitors approaching the city might expect the University’s colleges to dominate this flat landscape but from the South or the West it’s Addenbrooke’s Hospital which first meets the eye.

Located on a 40 acre site, Addenbrooke’s, part of Cambridge University Hospitals NHS Foundation Trust, serves half a million people locally. With 1,100 beds and over 6,500 staff, it’s the teaching hospital for the University of Cambridge, a centre for cancer services, organ transplants, neurosciences and genetics, and a leader in research and development.

Understandably, FM here is a challenge. The laundry service washes 5,300 tons of linen a year, while the switchboard connects more than 192,000 calls a month. Each day the post-room handles over 35,000 items of mail and the portering staff respond to around 280 requests.

Simon Lewis, estates operational manager, is responsible for the letting of the soft and hard service contracts. Once the contracts have been awarded, he is also responsible for the performance management of the hospital’s facilities contractors. In this demanding patient environment, performance delivery is crucial. “In reality however,” says Lewis, “contractors cannot always deliver what they say they can. In the past we had no real mechanism to monitor contractor performance, our contracts had a limited amount of teeth.”

Last year Lewis and Richard Howe, director of estates infrastructure and facilities, decided they needed to change the way contractor performance was measured. The route they decided on was the PFI model. Lewis explains; “It was Richard who proposed the PFI model. He said that because of the legality that surrounded this type of process it would be better for us to adopt something that was a recognised national approach – a system where contractors are measured on a performance basis and their monthly payment reflects their performance just as it would in a PFI project. We approached a company who had produced such a piece of software and asked them to adapt it for us.”

That company was Service Works Global, facilities and property management software specialists who have extensive experience within the PFI and performance management software marketplace. Lewis started working with one of Service Works Global’s senior PFI consultants in June 2006, and the resulting performance monitoring tool and payment mechanism went online on 1 February this year. What Lewis and Howe asked Service Works Global to design was a measurement tool which would calculate contractors’ payments for the soft services – cleaning and catering worth £6.8m a year – based on service delivery and quality of service.

“During the life of the contract,” explains Lewis, “we’ll be entering performance data weekly. There are two types of measurement – service failures and quality measurements. Our quality measuring team goes round to measure quality all the time, and the service provider is required under the contract to give us certain details relating to service failures. This might be how many times patients receive meals that are not satisfactory, or whether a patient with a special nutritional need receives the wrong meal or many times the service provider fails to give patients the required seven drinks a day.”

Quality levels can be subjective. Taking cleaning as an example, “Here we have minimum expectations as well as cleaning standards,” explains Lewis. “So for instance we would expect the contractor to clean the toilet say every two or three hours, regardless of the quality spec - so that's an input measurement. The output measurement would be an assessment of quality against national NHS standards. This might say that all windows, mirrors and glass surfaces should be free of fingerprints or carpet free from debris or stains. As soon as we see to the contrary, then it's a black mark. When you add up the black marks, there is a calculation that gives you a percentage score.”

Within the database Addenbrooke’s asked Service Works Global to allow an initial 5 per cent margin for error. This means that the contractor has to achieve a minimum service level of 95 per cent each month in order to receive 100 per cent payment. "If they fall below 95 per cent," explains Lewis, “you get a true reflection of service levels. If they score an average of 94 per cent over the year, then they have lost 6 per cent of £6.8m, and that's a lot of money.”

What benefits does Lewis expect to gain from this system? “I think that's the key to the whole thing," he says. “The contractor has got to deliver or they will lose money. The penalties are quite big and they apply to everything.”

Interestingly, Lewis does not necessarily expect to see his management costs reducing. He says, “To manage and monitor a major contract adequately you've got to put in sufficient personnel. But it will be much easier for us to correctly identify the contractor’s performance. The system will give us an array of solid information which we’ll be able to analyse and most importantly we’ll be able to give a better service to our users.”

In developing the database Service Works Global faced two important challenges. "Flexibility firstly", says the organisation’s senior consultant. “We were trying to avoid hard coding of the formulae because things may need to be changed in the future. So we have designed the system in such a way that different parameters can be built into each measurement,plus some weighting values.

“We also wanted to ensure that the client can rerun reports as many times as they want to. So if for some reason there is a dispute with a provider over whether the initial values arecorrect for say cleaning, which is fairly subjective, they can go back and update figures and change values. They can then re-run the reports to see what deductions or otherwise are calculated.”

Ease-of-use is also important, he adds. “Now all the facilities staff need to do on a monthly basis is click on a button and the system will populate the list of criteria and calculate the results. Then if the contractor doesn’t achieve any of the threshold performance scores that they are targeted to achieve, deductions will be calculated from their basic monthly payment.”

Manage risk
Providing the performance measuring tool turned out to be only half the story. While Service Works Global’s MD Gary Watkins was working with Lewis and Howe they started discussing how each of the potential service providers might be modelling their bids, (eg in an Excel spreadsheet), in order to manage the risk. Watkins suggested that having access to this performance measuring mechanism up front would enable bidders to submit more accurate and meaningful bids. Service Works Global gave the contractors access to the system for a specified period of time, which allows them to model their operational risk against actual formulae. Because all contractors had access to the same model to work on, Lewis and Howe hope there will be a closer relationship with the winning bidder prior to the contract commencing.

As Watkins explains, “Where contractors are multi-disciplined, they may be stronger in some areas than others. Accessing the database at the pre-bid stage gives them the opportunity to better estimate that risk and identify what controls or additional spending they need to apply. If they were to face a penalty, it enables them to identify how much the penalty would be.”

By using a replica of the actual tool to be used in the live environment, bidders can produce monthly reports by keying in dummy data or potential scores in the different categories. Plus they can run through scenarios such as staff absences or equipment downtime to see what the potential penalties might be.

“I’m not aware of anyone who has carried this out before, particularly with PFIs and similar types of contract,” said Watkins.” Normally each of the service providers reads through the documentation and inputs the details into an Excel spreadsheet and produces their modelling from that. And at which point they say, ‘OK we’re going to go ahead (or not), and we think there may be some form of risk.’ But it's difficult to quantify. And of course that risk falls back on Addenbrookes. If someone puts in a relatively low bid and wins the contract without having assessed the risk properly, they may go bust or pull out of the contract, then ultimately Addenbrooke’s owns the risk.”

Other organisations could adopt a similar approach claims Gary Watkins. “It could work for any industry sector, because although it is based on the PFI model it’s a comprehensive, flexible performance mechanism solution.” Clearly a welcome move forward which promises to bring benefits for both clients and FM providers.