Phil Rosenzweig

Who they are

A note on the name before anything else, because the two are constantly merged and this site keeps them apart. The perception-psychology halo effect is about how one impression of a person colours judgments of their unrelated attributes. Rosenzweig's halo effect is about how a company's known results colour every subsequent description of how that company operates. They share a word and very little else, and this page is about the second only.

Rosenzweig's target is the genre of business books that studies successful companies, identifies what they have in common and presents those commonalities as the causes of the success. His named targets were In Search of Excellence and Good to Great, and the argument transfers without modification to every book of that shape since.

Four of his nine delusions do almost all the work. The halo effect, where attributes are attributed to a company because its results are already known, and reversed when the results turn. The delusion of correlation and causality. The delusion of connecting the winning dots, which is selection on the outcome: study only winners, and read their common features as causes. And the delusion of rigorous research, where volume of data is mistaken for quality of design.

The sharpest part of the argument is the part most often missed. His halo objection is not that observers are sloppy and should try harder. It is that the data the whole genre is built from are contaminated at the point of collection, because the interviews, surveys and press coverage describing a company's internal qualities are almost always gathered after its results are known. The independent variable is therefore partly a function of the dependent variable. That is a measurement problem rather than a bias problem, and no amount of care fixes it.

The other side has a case, and it belongs here. Defenders of the genre argue that formulas derived from successful firms are heuristics for practitioners rather than causal claims, and that judging them as controlled research asks the wrong question of them. That defence works where the book itself makes the distinction. It is also precisely the distinction most books of this kind decline to make.

Key papers

Rosenzweig, P. (2007). The Halo Effect... and the Eight Other Business Delusions That Deceive Managers. Free Press. Business Book of the Year at the Frankfurt Book Fair, 2007.

Rosenzweig, P. (2007). "Misunderstanding the Nature of Company Performance: The Halo Effect and Other Business Delusions." California Management Review, 49(4). The peer-reviewed statement of the same argument, and the reason he is citable rather than only quotable.

Neither is read in this vault. Nothing is attributed to either beyond authorship and the standard account of the nine delusions, which was established in a separate verification pass.

Key ideas

This is the business halo effect, and it is not the perception one. A company's known results colour every later description of how it operates. The site keeps the two senses separate and so must this page.

The contamination is at collection, not at interpretation. Interviews and press coverage about a firm's internal qualities are gathered after the results are known, so the independent variable is partly a function of the dependent variable. Care does not repair that.

Connecting the winning dots is selection on the outcome. Study only the winners and the common features look like causes.

Volume is not design. More data collected the same contaminated way is more of the same problem.

The defence, given properly. These books offer heuristics rather than causal claims, and should be judged as such where they say so. Most of them do not say so.

He is the citable source for an objection this site had been making in its own voice, which is why the page exists.

---