1 Christiaan Katsma & Ton Spil. Agenda Dynamics of the ES Implementation process Suitability of agile or incremental approaches Inspection of 5 cases.

1

Christiaan Katsma & Ton Spil

Agenda

• Dynamics of the ES Implementation process

• Suitability of agile or incremental approaches

• Inspection of 5 cases in healthcare

• Results

• Discussion

Markus & Tanis (2000)ES Experience cycle

Parr & Shanks (2001)

Backgound

• ES Implementations traditionally are executed in a staged approach.

• Current Web applications, SOA, SAAS,.. based projects often use more incremental or agile approaches

• In the last century the incremental and agile implementation of ES was explained and described (Fichman & Moses, 1998; Stender 1999)

• But the monolithic technology behind the ERP systems at first prohibited actual incremental implementation approaches.

RDI (Fichman & Moses, 1998)

• “One question we have been asked is whether this methodology can be applied to enterprise resource planning (ERP) offerings from vendors like SAP, PeopleSoft or BaaN. We believe that except for the instance of a small company implementing a fairly well understood module,

the answer will usually be no. “

Sollution: Subsequent implementation efforts (Fleisch et al. 2004)

Information System oriented approach

R/3 Implementation

R/3 Implementation

Solution: Rapid implementation

Squeeze

• Currently we see an initial trend (Mezaros & Aston, 2007; Karim et all; 2007) for system implementers to gradually come up with cyclical implementation approaches.

• In the perspective of implementation time and cycle usage these approaches more and more come close to the agile philosophy (Alleman, 2002)

Organizations Learn: Cyclical implementation

ESImplementation

cycle

So we questioned….

• One decade after Moses & Fichmann (and others).

What is the current state of implementing ES packages using cyclical or incremental approaches?

Research set up

• 5 cases of ES implementations in the Netherlands in Hospitals.

• Different ambition and maturity levels

• Qualitative & longitudinal approach (started 2004)

• Interviews with at least three different “roles”in the project (Consultant, Healthcare professional, (Top) Manager)

Why healthcare?

• The specific distinction between the flexible care processes on the one hand and the structured and repeating operational hospital management processes, like purchasing and controlling (Merode et al., 2004).

• This distinction was/is seen as impeding and complicating the implementation process. (end of the 90’s).

• We currently see that this typical characteristic is becoming a foundation for cyclical implementation efforts.

What research perspectives on the implementation process?

• Cycles

• Ambition level

• Approaches

14

Cycles

• Based on Markus & Tanis we define a cycle as the entire proces from envisioning towards adoption of the ES.

Phase I

Project Chartering

Phase II

The Project

Phase III

Shakedown

Phase IV

Onward and Upward

Ambition level

• Based on the MIT Alignment model and its developments (Venkatraman, Scott Morgan, Yetton et al.) we distinguish three ambition levels for ES implementations

• IT driven Replacement (IDR)

• Package enabled Reengineering (PER)

• Human driven renewal (HDR)

Approaches

• Practitioners choose to utilize implementation methodologies based upon their own experiences and habits (Aydin and Harmsen, 2002, Hirschheim and Klein, 1989)

• Based upon (Goles and Hirschheim, 2000) we discern three distinctive approaches– Functionalistic– Integral – Interpretative

Combining perspectives

Our expectations

Overview of the results

Case A Case B Case C Case D Case ENumber of cycles

1 1 2 2 (ES discontinued) Multiple

Initial Ambition

o Replacement of legacy systems after merger

o Realization of a new Hospital Information System

o Improve workflow of healthcare processes

o Addition of care specific modules and possibly reengineering of existing ES

o Full implementation of an ES

Initial Approach

o Software oriented and firm approach by the vendor

o Testing and training planned, but not realised

o Functionalistic approach, sometimes indifferent to org. members

o Late training of the users

o Integral and structured approach

o Much communication with end users

o Functional design with many changes to the standard package

o Small amount of user involvement

o Standard waterfall method

o Functionalistic perspective

o Good user involvement

Result o Integration plus management information

o Partly changed processes in administration

o No success in care.

o Process continues ( up for a 2nd cycle)

o Most targets reached but the ambition level is increased continuously. (Project still continuing)

o Straight-forward implementation but no satisfaction with end product.

o Disposal of ES.

o Partly satisfied. (Project still continuing)

o Learning approach

Results

• We observe an increase in the use of cyclical approaches resulting in what we call macro dynamics at work in four out of the five hospitals studied.

• Macro dynamics means that the hospitals start drifting in either their ambition level, implementation approach or even both.

• Within the limitations of 5 cases we conclude that such a drift in most cases leads to implementation problems if not aligned within one implementation cycle.

• On the other hand persistence in ambition level and implementation approach results in a positive perception of the implementation project and its results.

Conclusions

• Functionalistic approaches still dominate in the ES domain

• Only one hospital adopts a mere IT- Driven Replacement ambition. (compared to 5 years ago the healthcare domain is advancing to proces reengineering in the care processes.

• Humand driven renewal still is lacking as phenomena in the ES domain.

• There is great dificulty in the cases to be consistent in the adopted approach and ambition level.

• Changes within one cycle do not show good implementation results. Changes between cycles show moderate to good results

Discussion

• Incremental implementation still means : following the “entire” cycle, but shortening the stages.

This makes us question:

• What are aspects in the project stage that are suitable for more agility?

• What technological barriers do we currently see to adopt agile approaches in the ES domain?