Read out

Mak­ing busi­ness pro­cesses easier to un­der­stand

Large en­ter­­­prises, com­plex struc­tures, and a mul­ti­tude of di­f­fer­­ent work­­flows: Busi­­ness pro­­cesses are usu­ally made up of many small steps that are im­ple­­men­ted in a defined sequence to achieve spe­­ci­fic busi­­ness goals. Espe­­cially in cases where new em­ploy­ees or di­f­fer­­ent groups of people are work­ing on these pro­­cesses, forms of visual rep­res­ent­a­­tion, for in­stance pro­­cess mod­­els, are needed to make these pro­­cesses as com­­pre­hens­ible as possible for all people in­­volved. Kath­rin Figl from WU’s In­sti­­tute for In­­form­a­­tion Sys­tems and New Me­­dia has in­vestig­ated the po­ten­­tial of visual rep­res­ent­a­­tions of busi­­ness pro­­cesses and looked at how pro­­cess mod­­els need to be designed to be eas­ily un­­der­­stood.

Ac­cord­ing to a study by WU re­searcher Kath­rin Figl, 80% of em­ploy­ees prefer visual rep­res­ent­a­tions of busi­ness pro­cesses over verbal de­scrip­tions. When it comes to im­prov­ing pro­cesses, the pref­er­ence is even stronger, with 90% find­ing visual pro­cess mod­els more use­ful than verbal de­scrip­tions. But even though com­pan­ies are spend­ing large amounts of money on busi­ness pro­cess mod­el­ing, syn­tax er­rors can be found in up to 80% of all mod­els, which is not only in­dic­at­ive of prob­lems in the mod­el­ing pro­cess but also of com­pre­hens­ib­il­ity is­sues. Com­pared to com­puters, hu­mans only have a very lim­ited work­ing memory and can eas­ily be pushed to the lim­its of their cog­nit­ive pro­cessing ca­pa­city by com­plex pro­cess mod­els. In her stud­ies, Kath­rin Figl from the In­sti­tute for In­form­a­tion Sys­tems and New Me­dia in­vestig­ates the po­ten­tial of pro­cess mod­els as cog­nit­ive tools and how pro­cess mod­els need to be designed to be un­der­stood quickly and cor­rectly.

Find­ing the right re­cipe

To find out what pro­cess mod­els should look like to be easy to read and un­der­stand, Kath­rin Figl has car­ried out ex­per­i­ments with more than 700 par­ti­cipants over the past few years. Com­pre­hens­ib­il­ity is­sues could be ob­served espe­cially with regard to com­plex pro­cess struc­tures that in­clude loops and deeply nes­ted con­trol-­flow blocks. The choice of mod­el­ing lan­guage is a par­tic­u­larly im­port­ant factor for com­pre­hens­ib­il­ity, be­cause the lan­guage chosen defines the rep­er­toire of sym­bols that can be used and the rules for com­bin­ing these sym­bols in the mod­el­ing pro­cess. “One example of a pop­ular visual nota­tion for pro­cess mod­el­ing is the Busi­ness Pro­cess Model and Nota­tion sys­tem, BPMN for short. BPMN is an in­dustry stand­ard developed by the Ob­ject Man­age­ment Group, and its basic sym­bol rep­er­toire has been found to be easy to un­der­stand for our test sub­jects,” ex­plains Kath­rin Figl. “The mod­el­ing lan­guage should avoid dif­fer­ent sym­bols rep­res­ent­ing the same mean­ing, and it should not use any sym­bols that have more than one mean­ing, be­cause this can lead to misun­der­stand­ings in the pro­cess of read­ing a model.”

Col­ors, shapes, and lay­out

Sym­bols with sim­ilar shapes and col­ors can also con­fuse read­ers. Sym­bols tend to be easy to un­der­stand and to learn if read­ers can in­tu­it­ively as­so­ci­ate them with their re­spect­ive mean­ings. Once busi­ness pro­cess mod­els reach a cer­tain size, it can be use­ful to de­com­pose them into vari­ous sub­-­mod­els. Even though it is gen­er­ally ad­vant­age­ous to avoid com­mu­nic­at­ing ir­rel­ev­ant in­form­a­tion to read­ers, it can be use­ful to provide them with over­view mod­els to give them bet­ter ori­ent­a­tion and help them find their way around large model hi­er­arch­ies. Kath­rin Figl re­cently pub­lished a lit­er­at­ure re­view which shows that aside from the factors men­tioned so far, model lay­out may also be a “secret re­cipe” for un­der­stand­able pro­cess mod­els. The first res­ults of an eye-track­ing ex­per­i­ment car­ried out in co­oper­a­tion with the Tech­nical Uni­versity of Den­mark sug­gest that cer­tain lay­out choices (e.g. high vis­ib­il­ity of con­trol-­flow blocks, avoid­ing un­ne­ces­sary changes of model flow dir­ec­tion) can make busi­ness pro­cess mod­els much easier to read.

Com­ing up with bet­ter ideas

In a joint pro­ject with Aus­tralia’s Queensland Uni­versity of Tech­no­logy, Kath­rin Figl also in­vestig­ated the ef­fects of visual pro­cess mod­els on em­ploy­ees’ power of ima­gin­a­tion and the qual­ity of their ideas, as com­pared to the ef­fects of tex­tual pro­cess de­scrip­tions. This re­search fo­cused par­tic­u­larly on ideas for im­prov­ing ex­ist­ing pro­cesses. Pre­vi­ously, no re­search had been done that provided an un­equi­vocal answer to the ques­tion of whether visual pro­cess mod­els help ana­lysts to come up with in­nov­at­ive solu­tions or, on the con­trary, if visual mod­els restrict their mental per­spect­ive. An ex­per­i­ment showed that this lat­ter con­cern is un­war­ran­ted: Visual mod­els worked a little bet­ter than tex­tual de­scrip­tions in stim­u­lat­ing the cre­at­ive qual­ity of sug­ges­tions for im­prove­ments. They do not in­crease the over­all num­ber of ideas for im­prove­ments, but they en­hance their use­ful­ness and ap­pro­pri­ate­ness for en­ter­prises. “Our test sub­jects, for in­stance, came up with more ideas on how to use new tech­no­lo­gies in the pro­cess. This means that visual pro­cess mod­els do not lead to more ideas over­all, but they help us to gen­er­ate ideas of higher qual­ity that are more use­ful,” Kath­rin Figl ex­plains. “Di­git­al­iz­a­tion is bring­ing changes to en­ter­prises world­wide and across many in­dus­tries. To re­main com­pet­it­ive, it is es­sen­tial for com­pan­ies to em­ploy tools that stim­u­late cre­at­ive ideas. For this reason, visual busi­ness pro­cess mod­els are an im­port­ant tool for op­tim­iz­ing pro­cesses and tap­ping the full po­ten­tial of di­gital tech­no­lo­gies.”

The stud­ies

The stud­ies Figl, Kath­rin (2017): Com­pre­hen­sion of Pro­ced­ural Visual Busi­ness Pro­cess Mod­els – A Lit­er­at­ure Re­view. Busi­ness & In­form­a­tion Sys­tems En­gin­eer­ing (BISE) (59) 1.

Figl, Kath­rin, Recker, J. (2016): Pro­cess in­nov­a­tion as cre­at­ive prob­lem solv­ing: An ex­per­i­mental study of tex­tual de­scrip­tions and dia­grams. In­form­a­tion & Man­age­ment (53) 6, pp.767-786.

Figl, Kath­rin, Recker, J. (2016): Ex­plor­ing Cog­nit­ive Style and Task-spe­cific Pref­er­ences for Pro­cess Rep­res­ent­a­tions. Re­quire­ments En­gin­eer­ing 21 (1), pp.63-85.

Mag. Anna Maria Schwendinger
Tel: + 43-1-31336-5478
E-­Mail: anna.schwendinger­

Back to overview