The Logic of Agility

Agili­ty means effec­tive­ness. It is less about effi­cien­cy, but pri­mar­i­ly about doing the right thing. But how can you know what is right in a world in which it is “nor­mal that many things are dif­fer­ent and are becom­ing dif­fer­ent faster and faster” (Karl-Heinz Geißler)? The answer is sim­ple: You can’t know, you have to exper­i­ment. That’s why agili­ty always means cus­tomer ori­en­ta­tion, because insights can only be gained through feed­back from the field. At its core, agili­ty is there­fore empir­i­cal research into pos­si­ble solu­tions for unknown needs in fast-mov­ing markets.

If the lad­der is not lean­ing against the right wall, every step we take just gets us to the wrong place faster.

Steven R. Cov­ey, 2004. The 7 Habits of High­ly Effec­tive People.

The term empir­i­cal is derived from the Greek word Greek εμπειρία (empeiría) and trans­lates rough­ly as expe­ri­ence or expe­ri­en­tial knowl­edge. It refers to the method­i­cal-sys­tem­at­ic col­lec­tion of data, expe­ri­ences and knowl­edge with the pur­pose of test­ing or refut­ing the­o­ret­i­cal assump­tions about the world’s inter­re­la­tion­ships. Agili­ty thus always begins with the well-known insight attrib­uted to Socrates: “I know that I do not know!” The log­i­cal con­se­quence of this sage real­iza­tion of not know­ing and the hon­est admis­sion of uncer­tain­ty is to work with hypothe­ses. Every pri­or­i­ti­za­tion, every sprint plan­ning is there­fore a hypoth­e­sis for a promised cus­tomer ben­e­fit. Good hypothe­ses will prove them­selves and bad ones will be refut­ed. Fail­ure as a means of gain­ing knowl­edge is thus in prin­ci­ple part of agili­ty. With­out this fail­ure cul­ture there can be no agility.

An empir­i­cal-sci­en­tif­ic sys­tem must be able to fail based on experience.

Karl Pop­per, Logik der Forschung 17

But if agili­ty now essen­tial­ly means gain­ing empir­i­cal insights through cus­tomer prox­im­i­ty and cus­tomer feed­back, then this nec­es­sar­i­ly requires decen­tral­ized struc­tures and a high degree of self-orga­ni­za­tion. The usu­al com­mu­ni­ca­tion and com­mand chan­nels up and down the hier­ar­chy, and then through var­i­ous com­mit­tees, impedes fast learn­ing far too much. Deci­sions must be made where the imple­men­ta­tion takes place and where the infor­ma­tion about the impact of the deci­sions is obtained: in the team and close to the customer.

Most peo­ple do not real­ly want free­dom because free­dom means accep­tance of respon­si­bil­i­ty, most peo­ple trem­ble at such acceptance.

Sig­mund Freud

So agili­ty always means sub­sidiar­i­ty and this requires inter­dis­ci­pli­nar­i­ty. In order for agile teams to be able to work and decide autonomous­ly close to the cus­tomer, they must be inter­dis­ci­pli­nary, i.e. their mem­bers must have all the skills and spe­cial­iza­tions required for this work.

Con­verse­ly, this auton­o­my also means greater respon­si­bil­i­ty than in hier­ar­chi­cal orga­ni­za­tions. The team is always respon­si­ble for its deci­sions. The con­ve­nient excuse of hav­ing only exe­cut­ed com­mands and imple­ment­ed com­mit­tee deci­sions does not exist in agile organizations.