I have noticed this phrase several times recently, in the context of systems implementations in large organizations: ‘a single version of the truth’. It refers to an affordance and anticipated benefit of the proposed new systems. The new systems (it is claimed) will improve the quality and reliability of management information, and (it is implied) resolve current problems associated with multiple sources of performance data which are inaccurate or distrusted or both.
Photo credit: Vitorio Benedetti (Flickr CC)
This is surely a laudable aspiration. But is it a realistic one? And why is it so appealing in these large, legacy organizations?
The claim, that a given system will produce ‘a single version of the truth’, seems to overlook the well-attested phenomenon of GIGO: garbage in, garbage out. In other words, the ability of any system to produce a single version of the truth is limited by the tendency of users to input inaccurate or false data. This phenomenon was first observed in the 1950s, and its relevance has grown in step with the increase in processing power. Garbage in, even more garbage out even faster.
But this is not the only reason to distrust the ‘single version of the truth’ claim. The organizations where I have encountered the claim are doing knowledge-based work. This kind of work is invisible and intangible. The work process is collaborative, involving numerous people and functions from across the organization and outside suppliers and partners. Frequently, the work is hypothesis-driven – teams are testing ideas and learning from feedback, to inform their next steps.
The ‘truth’ in these organizations is messy. It is often more qualitative than quantitative in nature. The ‘truth’, if it exists at all, exists only momentarily and fleetingly. If beauty is in the eye of the beholder, truth is in the eye of the stakeholder, and in these organizations, where multiple teams and functions work together to produce complex products, each stakeholder sees a different version of the truth.
The ‘truth’ in these organizations is not amenable to capture in the proposed systems, or reducible to a single version that is recognizable and meaningful to everyone.
The claim of ‘a single version of the truth’ is unrealistic, even dishonest.
That this claim continues to be made reveals something of what is going on in these large, legacy organizations. These organizations were once able to measure and control their activity and output, by worker, team, division, unit. By day, week, month, year. Physical products could be counted, measured, checked and inventoried.
In that pre-digital world, a ‘single version of the truth’ may have been attainable and useful. But now these organizations are producing digital products and services in distributed teams. The products are intangible, the work is hard to see, and the ‘truth’ is far from straightforward.
The promise of ‘a single version of the truth’ appeals to us as it promises to cut through the ambiguity, simplify the complexity, and remove the anxiety which legacy organizations experience when transitioning to digital. But it is a fiction. These organizations must accept the impossibility of a single authoritative source of objective truth, and seek to develop new skills and habits for sharing and interpreting multiple forms of information and insight from multiple sources in multiple locations.
The truth is out there … but probably not in a management system.
Matthew Moran 