Reporting bias and failure: lessons for sponsors

How do you know what’s going on?

On target -- how do you know?Some recent research into bias in reporting about project status has implications for change governance. Two recent examples of poor reporting have surfaced recently: the fiasco on the launch of the IT to support the US Government Health Care programme (Obamacare) and the under-performance of the Universal Credit IT system in the UK. In both cases the senior civil servants and politicians claimed there were no problems only to find major problems when the systems went live (in the US case) or were scrutinised (in the UK). Why do the sponsors not find out (or not tell us) about problems in change projects until Its too late? What can a sponsor do about it?

The research identifies thinking biases that help explain the poor exchange of information between the managers of change and controllers (sponsors) of change.

Sloan Management ReviewProject staff can’t be relied upon for accurate reporting

It seems obvious, but project and change staff tend to hide information when problems arise. There is a fear of being blamed for problems and their consequences in many workplace cultures. The pressure for successful change does not help. Its not that the problems are deliberately hidden, they are just not mentioned!

Whilst senior managers are expecting their staff to speak up when problems arise, they don’t realise the pressures they place on those managers. Consequently many managers produce biased status reports with the most common bias towards optimism. Managers want to look competent, who wouldn’t, managers are also on the wrong side of a power relationship.

Of course, once an optimistic report has been given it establishes ‘facts on the ground’ which can not be later denied. Thus a spiral of poor and eventually false reporting is started.

The best fix for this bias is for the sponsor to find independent sources of information about progress and problems. A programme or project office can be very useful for this. If a programme or change manager knows their manager has access to independent information then it helps to keep their reporting honest and open.

I remember one organisation I worked in where the Programme Office (PO) issued status reports on all the change activities every month prior to an executive board meeting. All the project and programme managers would gather in the PO when the reports were released and always argued about the information and summaries; especially the RAG status of their own activities. The reports were all derived mechanically from the project and programme database; and most of the information in there was entered by the project and programme managers! It didn’t stop the managers complaining that their activity was not put in a good light by the reports and made them look incompetent! However, the executives set a lot of store by the information in the reports.

Bias is complicated

There are lots of reasons why people misreport their change activities. These include individual personality traits, the climate of the workplace and cultural norms. For instance, someone who has an above average appetite to take risks is more likely to take risks and then misreport their status (about the risk exposure as well as failures). Staff who see the world as ‘glass half full’ will tend to report more optimistically, rather than realistically. Whereas ‘glass half empty’ people will produce more accurate reports as their pessimism will tend to offset any errors in reporting.

The climate in the workplace can differ significantly across organisations. In my experience the reporting climate in change activities is very different from the operational part of the organisation because there is an expectancy of reporting and problems. The research suggests that people in a ‘self-interested’ climate are much more likely to misreport. Having consultants in a change activity can promote a ‘self-interested’ climate. Whereas  people in a ‘rule-based’ environment tend to provide more accurate reports.

The research compared ‘individualistic’ cultures, such as in the US, with ‘collectivist’ cultures, such as those in the far east. Staff in an individualistic culture were more likely to shift blame where possible in reporting, or misreport if they couldn’t; however they are more likely to respond to rewards for accurate reporting. Staff in a collectivist culture would be more willing to take responsibility for problems but may hide them for longer to give the team more time to address them.

The response to bias is not to fight it. The senior manager should take individual and team bias into account when building teams and the working climate. Recognise where bias might arise and compensate for it by balancing a team and following through with questions and demanding information.

Audits don’t find bias in reporting

If your change staff are hiding information from the senior managers and misreporting then they will do the same to an audit team. Sending in the auditors to seek out information and correct reporting bias causes the change team to adopt a more defensive attitude resulting in less accurate reporting and even harder ‘facts on the ground’ than before.

The hunkering down by the change team leads to a cycle of misreporting and mistrust that builds into high walls and dysfunctional behaviour. It’s best not to go there if you can avoid it. The right approach is to generate trust and maintain trust between the change team and the sponsor.

Senior managers exacerbate the problem

One of the key factors for success in change activities is sponsorship by senior managers and especially having the accountable sponsor as senior as possible in the organisation. Whilst this is very important in obtaining resources and political clout to deliver successful change it is the opposite for reporting bias. The power distance between the managers in the change team and sponsor can lead to tensions and misreporting behaviours.

Change managers in the research talked of ‘career’ impact on the information being reported. How are you going to tell a senior executive that his pet idea is not going to work and the plug should be pulled. There are a lot  problems in talking truth to power.

There are two things that can be done to help. The first is the relationship the sponsor establishes with the change managers. The power distance needs to be carefully managed and trust established so that the change manager is not in a career threatening position.The second thing is to establish a safety valve in the governance for the change activity. This usually means the change manager has access to another senior manager (at a similar level to the sponsor) who is independent of the change. The safety valve manager can then address problems in the change directly with the sponsor without the threat of the power distance. An independent programme office reporting to an executive board member who is not allowed to be a change sponsor is one way I have seen.

Senior managers ignore bad news

Not surprisingly, if change managers are reluctant to pass on bad news so are the sponsors; even more so. This is exactly why politicians leading failed changes bluff the reports all the way to the line. It turns out that senior managers are as prone to self-interest and bias as any other manager. The bad news message has more chance of being ignored if the messenger was internal. For instance, reports from auditors or risk managers were discounted less often than the same information from project and change staff.

The only way round a senior manager is another senior manager (the safety valve we discussed above); or a whistle blower. So a good adjunct to the safety valve is an established whistle blower scheme with rewards. If you can get 10% of the fines paid for blowing the whistle on law breaking activity within a company; can you get 10% of the savings from identifying a doomed change?

What to do?

The above research applied to change activities in an organisation suggests two things that sponsors of change need to think about:

  1. How are you recognising potential sources of bias in the reports you receive and have a plan to mitigate it?
  2. Do you have governance features which will help when item 1 fails, or you become the problem?

Leave a Reply

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.