Constructing the accurate
forecast: an actor-network
theory approach
Tiina Henttu-Aho and Janne T. Järvinen
Department of Economics, Accounting and Finance, Oulu Business School,
University of Oulu, Oulu, Finland, and
Erkki M. Lassila
Department of Accounting and Finance, Turku School of Economics,
University of Turku, Turku, Finland
Abstract
Purpose – This paper empirically demonstrates the major organizational events of a rolling
forecasting process and the roles of controllers therein. In particular, this study aims to investigate
how the understanding of a “realistic forecast” is translated and questioned by various mediators in
the rolling forecasting process and how it affects the quality of planning as the ultimate accuracy of
forecasts is seen as important.
Design/methodology/approach – This study follows an actor-network theory (ANT) approach and
maps the key points of translation in the rolling forecasting process by inspecting the roles of mediators. This
qualitative case study is based on interviews with controllers and managers involved in the forecasting
process in a single manufacturing company.
Findings – The paper identified two episodes of translation in the forecasting process, in which the forecast
partially stabilized to create room for managerial discussion and debate. The abilities of controllers to
infiltrate various functional groups and calculative practices appeared to be one way to control the accuracy
of forecasting, although this was built on a façade of neutrality.
Originality/value – Prior literature identifies the aims of interactive planning processes as being to
improve the quality of planning. The authors apply ANT to better understand the nature of mediators in
constructing an entity called a “realistic rolling forecast”.
Keywords Translation, Planning, Mediator, Actor-network-theory, Controller, Rolling forecasting
Paper type Research paper
© Tiina Henttu-Aho, Janne T. Järvinen and Erkki M. Lassila. Published by Emerald Publishing
Limited. This article is published under the Creative Commons Attribution (CC BY 4.0) licence.
Anyone may reproduce, distribute, translate and create derivative works of this article (for both
commercial and non-commercial purposes), subject to full attribution to the original publication and
authors. The full terms of this licence may be seen at http://creativecommons.org/licences/by/4.0/
legalcode
The authors gratefully acknowledge the useful suggestions and comments provided on earlier
versions of this paper by Nicolas Berland, Martin Messner and Leona Wiegmann. The authors are
thankful for the feedback from attendees at the joint conference of Enroac and MCA in London 2019,
the MCA Symposium in Nice 2018 and the 11th Enroac in Naples 2017. The authors also gratefully
acknowledge the financial support from the Marcus Wallenberg Foundation. Finally, the authors are
thankful to all anonymous reviewers as well as journal editors for their valuable comments, which
improved the quality of the paper.
MEDAR
31,7
116
Received 6March 2022
Revised 8 September 2022
1 February 2023
Accepted 19 February 2023
Meditari Accountancy Research
Vol. 31 No. 7, 2023
pp. 116-132
EmeraldPublishingLimited
2049-372X
DOI 10.1108/MEDAR-03-2022-1613
The current issue and full text archive of this journal is available on Emerald Insight at:
https://www.emerald.com/insight/2049-372X.htm
1. Introduction
Arguably, rolling forecasts are one of the key tools in organizational planning, in which the
accounting function is often seen as a neutral information intermediary (Bourmistrov and
Kaarbøe, 2013: Goretzki and Messner, 2016). According to actor-network theory (ANT), an
intermediary “transports meaning or force without transformation” (Latour, 2005, p. 39).
However, ANT suggests that scholars should view most associations as mediators.
Mediators, unlike intermediaries, “transform, translate, distort, and modify the meaning or
the elements they are supposed to carry” (Latour, 2005, p. 39). This means that even though
accountants may have an interest in representing both themselves and accounting
technologies as neutral, we view them as active participants, that is, mediators that
influence the outcome of the forecasting process, having a performative effect on the world
the forecast intends to describe (Callon, 1998; MacKenzie, 2006; Revellino and Mouritsen,
2015).
Through continuous planning, organizations strive to improve their decision-making by
keeping up with how current trends in markets and the internal production capabilities of
firms influence future profit-generating potential (Lorain, 2010). This means a combined
understanding regarding the behaviour of financial and operational sales forecast figures
(Huikku et al., 2017), as well as the costs of resources (Bourmistrov and Kaarbøe, 2013).
Rolling forecasts strive to achieve many qualities at once – accuracy (Jordan and Messner,
2020), proactiveness (Henttu-Aho, 2018) and adjusting/reformulating targets (Bourmistrov
and Kaarbøe, 2013; Goretzki and Messner, 2016; Palermo, 2018). Accounting research has
recently been focused on organizational aims of delivering realistic numbers around the
planning process (Østergren and Stensaker, 2011; Henttu-Aho and Järvinen, 2013; Jordan
and Messner, 2020) and hence improving the quality of planning. However, the way
forecasts are constructed, and the influence of management accounting experts and their
calculations on the outcome of this process, are not presented in detail in prior literature
(Sandalgaard and Bukh, 2014). Some beyond budgeting (BB) studies suggest that the role of
controllers in regard to the quality of planning may be ambivalent (Østergren and Stensaker,
2011). On the one hand, controllers use their managerial relations, statistics and trends to
compose more realistic forecasts and targets. On the other hand, they are found to be strongly
affiliated with sub-divisional management in helping them to “sell” their viewpoints to upper
management.
Forecasting typically emphasizes impartiality and forecast accuracy (Jordan and
Messner, 2020). This implies that controllers may depict themselves as neutral experts
(Miller, 1992), claiming the role of intermediaries who would ideally carry the meaning
invested in numbers without transformation. However, this study draws on ANT and views
controllers as a “web of mediators” (Latour, 2005), that is, they are included in a network of
actors that includes both human actors and non-human actants, who in their own ways
influence the outcome of the forecast. In addition to human actors, accounting inscriptions,
technology and money are also seen as mediators that translate the meaning of information
(Lowe, 2001; Hyvönen et al., 2008; Robson and Bottausci, 2018).
To investigate the influence of these mediators in the forecasting process, we draw on a
qualitative case study in a single manufacturing organization. Like most ANT studies, we
use Bruno Latour’s works (Latour, 1987, 2005), where various technologies and inscriptions
are used to establish facts and to organize social knowledge. This means the actor-network
is not defined a priori; rather, it emerges from the complex, open set of relations between
other actants, both human and non-human, through interaction and performativity. Thus,
the relational nature of the network and the heterogeneity of its actors are at the heart of
ANT (Law, 1992; Callon and Law, 1997). Prior research discusses how accounting systems
Constructing
the accurate
forecast
117
never seem to stabilize completely in order to fully achieve “black box” status; instead, they
continue to be contested and develop over time (Quattrone and Hopper, 2006). In fact, rolling
forecasts can be seen as fluid objects that become “immutable mobiles” (Latour, 1987) that
maintain their form and are able to foster a centre of calculation that allows control at a
distance (Robson, 1992). It is precisely the nature of the accounting inscriptions as
immutable mobiles that allows centres of calculation to know what takes place in local
peripheries.
Drawing on the concepts outlined previously, we present an interpretive qualitative case
study and demonstrate through empirical evidence how the rolling forecast emerges in an
actor-network comprised of various mediators. The broad purpose of this paper is to
mobilize ANT concepts of mediators and intermediaries to understand the varying degrees
of translation in the forecasting process. Our research question is why actors in a global
paper company challenge and question the assumptions of a “realistic forecast”, while at the
same time, they strive for forecast accuracy. Our results indicate how not only managers
(Goretzki and Messner, 2016) but also company controllers take part in transforming
assumptions about the future into meaningful forecasts, while still maintaining notions of
objectivity, claiming the status of neutral intermediaries. By this, our study intends to
contribute to studies that emphasize the role of controllers in establishing reliable accounting
inscriptions by discussing how the reliability of an accounting inscription is achieved (Lowe
and Koh, 2007) and how accounting information (partially) stabilizes as a result of their actions
(Andon et al., 2007; Hyvönen et al., 2008). Second, we expand the literature on the quality of
planning (Goretzki and Messner, 2016; Jordan and Messner, 2020) and interactive planning
processes (Østergren and Stensaker, 2011; Bourmistrov and Kaarbøe, 2013) by highlighting
management accounting’s mediating role inherent in forecasting.
The remainder of the paper is organized as follows. First, in Section 2, we provide a
literature review on planning and various aspects of rolling forecasting. Section 3 explains
the principles of ANT (Latour, 1987, 2005), which we aim to follow in understanding how the
“rolling forecast” entity is constructed. Section 4 describes the research design. Section 5
demonstrates how rolling forecasts are constructed within the cross-functional forecasting
process and contemplates the professional aims of controllers. In Section 6, we present a
thorough discussion of the empirical findings and previous literature and conclude the
paper.
2. Rolling forecasting and accuracy
Rolling forecasting is a planning process, where a forecast is updated monthly or quarterly by
adding an extra month or quarter to the period of forecasting (Henttu-Aho and Järvinen, 2013).
Thus, there can be several forecast versions prepared during the year, all of which consist of
different calendar periods and figures. The advantages of rolling forecasting are related in
particular to timeliness and the ability to update plans (Lorain, 2010; Henttu-Aho, 2018). Bøgsnes
(2009) and Sandalgaard and Bukh (2014) note that rolling forecasts seem to differ from budgets in
the sense that they do not have to be officially approved by the board of directors. There seems to
exist a consensus in BB literature that there is usually no penalty for poor accuracy of estimated
figures in a forecast. Such a view seems to differ from Jordan and Messner (2020), who
demonstrate how forecast accuracy can be used as a results control mechanism for managers to
improve the quality of planning, thus being part of the assemblage of performance metrics.
Therefore, the main idea in forecasting is to observe the key changes in the business environment
and estimate their monetary effects on profit, as well as to assess how the figures indicate the
feasibility of achieving organizational targets (Bourmistrov andKaarbøe, 2013).
MEDAR
31,7
118
In empirical studies, sales forecasts have been found to be a crucial starting point for
company revenue and cost planning and for the alignment of demand and supply (Goretzki
and Messner, 2016; Huikku et al., 2017; Jordan and Messner, 2020). Therefore, the accuracy
of sales forecasts, seen as an important element of planning quality, has been of special
interest to organizations. From the perspective of costs, forecasting has been improved by
the organized delivery of various cost factors and cost information collected by information
systems (Bourmistrov and Kaarbøe, 2013; Henttu-Aho and Järvinen, 2013).
In fact, prior empirical studies demonstrate various ways to improve the quality of
planning numbers. First, the growing importance of sales forecasting in uncertain business
environments has led to the development of quantitative modelling tools (Huikku et al.,
2017). Second, there have been aims to incentivize more accurate sales forecasting by using
accuracy measures (Jordan and Messner, 2020). In addition, empirical studies on BB indicate
that interactive (“dynamic”) forms of planning process may provide ways to improve the
quality of planning (Bourmistrov and Kaarbøe, 2013).
Importantly, the aim of accurate forecast figures may conflict with the need to set ambitious
targets. For instance, literature on BB argues that while organizations typically set their target
figures at an ambitious level, the same figures cannot be used for unbiased forecasting (Bøgsnes,
2009). Ambition in constructing forecast figures may create a “polluted forecast” (Østergren and
Stensaker, 2011). Thus, one important aspect of rolling forecasting is the reconciliation of different
and potentially conflicting goals, which calls for increased organizational interaction (Østergren
and Stensaker, 2011).
3. Actor-network theory and accounting
ANT has been part of scholarly discussions in accounting for decades (Robson and
Bottausci, 2018). In Reassembling the Social, Latour (2005) discusses mediators and
intermediaries in the process of translation. Of these, intermediaries carry forward meanings
without transforming them and as such cannot influence other actors. However, empirical
interest in ANT is concerned with mediators, which can influence other actors as well as
having the power to reconfigure actor-networks; thus, mediators translate meanings and
form actor-networks between human and non-human actors.
The process of translation consists of four stages: problematization, interessement,
enrolment and mobilization of allies (Callon, 1986). Problematization refers to focal actors
that actively appeal to other actors with their conceptions of issues and indicate having
solutions to these problems though explicit practices. Interessement is defining and locking
the roles of other actors by the focal actor in accordance with their plans to solve the
problem. Enrolment refers to successful interessement and the creation of alliance networks
in order to build agreement regarding their interests. The enrolling agency thus stabilizes and
cements links to other agencies. Finally, mobilization refers to a mechanism that monitors the
stability of various interests. The process of translation ultimately has the capability to produce
“facts” and even stabilize fully to become a “black box”. According to (Latour, 1987, p. 131), a
“black box” can be considered as a taken-for-granted system where many elements act as one.
Importantly, prior research discusses how accounting systems never seem to stabilize completely
to fully achieve black box status; instead, they continue to develop over time (Quattrone and
Hopper, 2001, 2006). Likewise, managers will most likely understand that financial forecasts are
not unquestionable facts where controllers act as neutral and impartial intermediaries; rather,
forecasted numbers are constantly inmotion, reconfigured by variousmediators.
Law (1991) and Latour (1999) provide examples of different kinds of mediators.
Inscriptions materialized into forecasted accounting numbers (Busco and Quattrone, 2018;
Robson and Bottausci, 2018) are one such important category but are often in themselves
Constructing
the accurate
forecast
119
incomplete and never fully stabilize (Quattrone and Hopper, 2006). Many accounting studies
focus on the capability of various accounting-related technologies, particularly information
systems, to reconfigure actor-networks (Lowe, 2001; Quattrone and Hopper, 2006; Hyvönen
et al., 2008). Finally, and not surprisingly, people, particularly controllers, carry forward
meanings and are an integral part of actor-network configuration.
Other relevant ANT concepts include, for example, “centre of calculation”, “action at a
distance”, “trials of strength” and “obligatory passage point”, which illustrate how
representations emerge in actor-networks. A centre of calculation refers to a point in an actor-
network in which the data from distant places and spaces are brought in, which are then
combined and calculated into information bringing a more comprehensive picture of things
happening faraway in time and space and therefore outside the centre (Latour, 1987, p. 232). From
the centre, action at a distance becomes possible due to the translation process of information,
which generates knowledge accumulation of these distant spaces and events. Such knowledge
production, which generates new types of “truths”, is not fully stabilized, that is black-boxed, but
rather creates contestable “facts” which can be challenged and, if necessary, modified according
to the “better” knowledge or “truth” of the given situation being contested. Therefore, trials of
strength are events where actors contest the existing knowledge or so-called “truth” by bringing
in possibly better knowledge or “truth” of a given situation. An obligatory passage point relates
to the previously mentioned problematization stage of translation, where the negotiation space is
under the control of a focal actor (Callon, 1986).
This paper presents a case study of a chain of events related to the forecasting process
and controller involvement in the creation of accurate forecast numbers. The case study
approach will be elaborated further in the section that follows.
4. Research design
This empirical study portrays a global paper company (hereafter Paperco) setting, where
budgeting and planning were undergoing significant changes. This presented a good
opportunity for conducting a qualitative field study to understand the abandonment of the
annual budgeting process. Even though our preliminary theoretical approach was the
budgeting change, from the very first interviews, we recognized that when our informants
talked about recent developments in the budgetary process, they were often referring to
rolling forecasting. Moreover, while the previous annual budgeting process was mainly
focused on manager–subordinate relations, the interviewees seemed to be much more
concerned with certain cross-functional events being central to the novel rolling forecasting
process.
At the time of the study, Paperco represents the traditional core of heavy industry, with
70 production sites in 20 different countries. According to Paperco’s financial report, the
company places great emphasis on protecting its production assets and avoiding any
unplanned production stoppages. A major part of its sales takes place in Europe. Sales are
operated through Paperco’s own distribution centres and global sales offices. Products such
as pulp are sold through the company’s own distribution centres. At Paperco, each business
area has its own sales organization and personnel. Business areas are responsible for their
profit and for achieving their strategic targets.
The dataset for the study consists mainly of interviews, but some secondary data sources
were also used, such as organization charts and annual reports. As the empirical phenomenon of
a study, rolling forecasting, is typically found to become a central activity in organizations that
have abandoned traditional budget control (Østergren and Stensaker, 2011), we felt it important
to conduct interviews with various organizational functions. Our interview data consists of 26
interviews with 19 different people in sales, production plants (mills), business areas, support
MEDAR
31,7
120
functions (energy and wood supply) and group (see Appendix). We interviewed various people
who were involved in the rolling forecasting process. Each interviewee was also asked to
nominate other persons whose working tasks were closely related to forecasting. Finally, we
interviewed accounting personnel, managers and support function personnel to establish an
adequate illustration of the network constructing the forecast (Lowe, 2001). We interviewed the
chief financial officer (CFO), group controllers, business controllers, sales controllers, mill
controllers and support-function controllers, as well as various function managers, such as
operational, business and support-functionmanagers.
Interviews were carried out in several waves, in 2009, 2011 and 2014. The field work
started in 2009, soon after the company had started to prepare monthly forecasts on a rolling
basis. First, we conducted semi-structured interviews with controllers to get an overall
understanding of the rolling forecasting as part of the management control system in the
company. These interviews revealed that the new practice of rolling forecasting required
frequent interaction with various organizational actors. The second round of interviews, in
2011, enabled us to get a deeper understanding regarding organizational interaction around
the forecasting process by extending the interviews to managers and sales controllers, the
latter of which appeared to be a recently established position in the company. The last round
of interviews, in 2014, focused more on the forecasting process of production sites (paper
mills) and the roles of support functions.
Several interview questions were prepared for each theme, but not all of them were
applicable to all interviewees due to their different positions in the organization. The first
round of interviews, in 2009 with controllers at various organizational levels, focused on
four themes: new developments in budgeting, the role of technology in budgeting, new
development projects, and the changing role of the controller. Then the second and third
rounds of interviews, carried out in 2011 and 2014, were more like open-ended interviews
with a few clarifying questions. In 2011 the focus was mainly on new rolling forecasting
practices and the ways planning and control were carried out in the organization by
controllers and managers. Finally, the interviews carried out in 2014 sought to bring deeper
understanding about the main aims of the forecasting process and controlling. For example,
we were interested in issues such as, what is the main purpose of the forecasting process and
whether the aims of accuracy or realism of forecasting figures, or organizational motivation,
can be supported by controlling.
As organizational theorists often emphasize difficulties in crossing inter-occupational
boundaries of humanswith respect to building shared meaning, we found it important not to
focus on ties between actors but rather to investigate the actions and events where the
meaning of accounting information was contested and translated by various human and
non-human actors (Latour, 2005). For this purpose, we used ANT as our method theory to
study the functioning of the rolling forecasting process and the role of mediators for the
quality of planning, both of which were in our focus from the domain-theoretical perspective
(Lukka and Vinnari, 2014).
5. Two episodes of translation in the rolling forecasting process
This section describes the construction of a rolling forecast in Paperco as a web of mediators
by employing the elements of “sociology of translation” (Latour, 1987, 2005). We map the
key points of translation in the rolling forecasting process (problematization, interessement,
enrolment and mobilization) (Callon, 1986) and inspect the roles of various mediators, such
as inscriptions, technical artefacts, human actors (especially controllers) and the money
therein. In assessing the key points of translation, we identified two episodes. The first
episode demonstrates how the initial sales prospects provided by the firm’s field offices were
Constructing
the accurate
forecast
121
questioned, analysed and finally altered by sales controllers and other actors in the company
business area. The second episode shows how the sales forecast was not taken for granted at
the production site but carefully inspected and even resisted by mill controllers and other
actors. Finally, our findings also reveal how monetary incentives acted as a mediator in
constructing accurate forecasts.
5.1 Episode 1: problematizing the sales forecast
Paperco was operating in a volatile business environment, and predicting customer demand
was seen as a challenging task. The company had field sales offices around the world, where
sales office personnel were continuously assessing the sales demand prospects among
current and potential customers. These first stage sales prospects were related to demand
and sales volume estimates for the next 12–15months. However, sales controllers and
product segment representatives were setting initial sales targets twice a year, which guided
the level of sales for the next six months:
In our organization, the sales (function) doesn’t set their own targets, but it is especially me and
my team’s task to be involved in the target setting. [. . .] We aim to tell a story of how we see that
the demand and sales price will develop in the future, what type of outside indicators and
parameters we have taken into account (in calculations), and how we see the competition will be
in different areas. [. . .] This can be seen as a sort of sparring and helping – providing a direction
to them. (Sales controller, 2 July 2014)
Field sales office people updated their first-hand estimates of demand straight into the sales
information system. As the sales offices were usually located far from actual business area
management and paper factories, these inscriptions uploaded to the system were the only
available outlook of future demand. At Paperco, the importance of this sales prospect
information as a mediator became evident as it immediately started to mobilize various
actors in business area sales.
First, the sales volume figures of field sales offices were uploaded into the system,
stimulating lively discussion in the business organization. In particular, controllers were
problematizing the accuracy of these sales figures. One interviewee, an experienced sales
controller, stated that: “the information considering the first three months is fairly accurate,
but the accuracy of later months can be seen declining strongly”. The sales controller
challenged the assumed truthfulness of the given sales prospects by arguing that
salespeople at field offices usually painted too optimistic a picture of sales possibilities.
Next, controllers, in conjunction with organizational salespeople, started to inspect these
given sales prospects within the various product segment teams. At Paperco, business area
sales organization was divided into product segments, and each product segment had its
own team. Furthermore, each product segment team consisted of managers and one sales
controller. The interaction within these product segment teams was daily and informal
(Østergren and Stensaker, 2011). Teams started to assess the feasibility of given sales
prospect figures as soon as they were available. They simulated the given figures and
calculated profitability at various price levels with the assistance of an analysis tool. The
forecasting period for the business area was 12months plus three months, for which the top-
line price for each product segment and corresponding sales volumes were forecasted.
Product segment teammembers, including the sales controller, became important mediators
in questioning and altering the prospective demand figures given by remote sales offices.
The important skill of sales controllers was related to product and customer profitability
calculations, as well as the verification of sales prospect data in cooperation with product
segment managers. One sales controller explained how, by conducting historical data
MEDAR
31,7
122
analysis (Bourmistrov and Kaarbøe, 2013), they were able to compare the sales prospect
numbers (provided by field offices) with the company’s historical sales trends:
And, well [. . .] we don’t quite trust those sales prospects as such, but we make some analysis of
them using our historical data. We build confidence (on the sales data) with the help of this data
analysis. There we can see how the price and volume will develop in the future. We provide some
data points to these segment teams to be able to construct a rolling forecast. [. . .] And then we
have quite good forecast patterns, which we believe in, especially from the perspective of price
and volume development. (Sales controller, 20 June 2011)
This sales forecasting tool was an important technical artefact in the forecasting process as it
acted in concert with controllers to adjust the initial sales estimates in a direction that had more
credibility with other actors. The tool proved to be a critical mediator for the acceptance of sales
forecasting information in the organization. There were knowledgeable experts in each product
segment team, who had the opportunity and skills to contest the information given by their
colleagues at the field offices. However, the data analysis tool, combined with the competence of
the sales controllers, formed an interessement device, through which the segment teams were
enrolled in the business area sales alliance (Munro, 1999) to craft “realistic sales information”. This
analysis tool scaled the field’s sales prospect figures by the historical sales trends and thus
represented the initial sales prospects in a different light across thewhole organization.
Finally, it was not unusual for the questioning of field offices’ sales prospects tomobilize
sales controllers and business controllers to visit customers directly to track the origins of
numbers and to get more information on the actual market situation:
It is not simply that some sales guy says that this is what all the others do, and we should follow
them if we want to be in this business. That’s a common slogan for salesmen. But now we try to
find the facts that these slogans are based on. (Sales controller, 20 June 2011)
The validation of sales forecast figures, usually seen as a managerial task, appeared to be an
important part of the work of sales and business controllers (Bourmistrov and Kaarbøe, 2013).
This strengthened “the system of alliances” around “realistic” sales forecasts. However, the final
acceptance of modified sales figures was given by the sales manager, who reviewed the trend of
figures andmade the decision from the business area viewpoint.
These three examples show the scepticism towards the sales volume estimates. At first, sales
controllers and their calculations were problematizing the initial sales prospects of field offices
and building a network of allies to support the conversion of sales figures. Then, sales controllers
acted as important mediators in shaping the figures to be more realistic and trusted within the
business area. The historical analysis tool represented a technical artefact that was used to shape
the “realistic forecast figures”. By using historical sales data, the controllers aimed to show the
acceptable range of trends for the prospective figures, whilst also building allies, which appeared
to have a strong altering and stabilizing effect for the rolling forecast. During the first episode, the
sales forecast became “partially stabilized” sufficient to be communicated, which in our case
means that, on the one hand, it was trusted that it could be mobilized further, but on the other
hand, the actors knew that it was just a forecast that could still be contested. Thus, in the sales
organization, the sales forecast could still not be considered as a black box or an unquestionable
fact at that stage of the forecasting process. There was to be a subsequent test when the sales
forecast was presented to the production site.
5.2 Episode 2: problematizing the operative forecast
After the sales forecast was sufficiently stabilized to allow for its use in planning, it was the
sales organization’s (of which the sales controller was a member) responsibility to carry out
further profitability calculations and data analysis. The aim of this was to determine how to
Constructing
the accurate
forecast
123
divide the sales volumes between production sites and factories. Thus, the sales forecast had
the utmost importance for the mills, as the sales allocation plan practically determined the
level of capacity usage at the mill and might, at worst, have severe implications for mills in
the form of employee layoffs and shutdowns of production lines. The following quotation
indicates that the interviewed mill director saw the sales forecast as a credible set of
numbers but was still worried about the origins of the presented figures:
Apparently, there is a lot of information which is known only by the financial function and which
has already been processed. Then, when we [mill management] participate in the RF [rolling
forecasting] meeting, where people from production and sales are also present, the future outlook
is only the tip of the iceberg. We actually don’t know the whole process behind those figures. (Mill
director, 11 April 2014)
This quote signifies that the forecast had at least partially stabilized (Quattrone and Hopper,
2006), stable enough to achieve mobilization and to be acted upon, whilst remaining an
inscription that could be challenged and questioned in the process of being continuously
made and updated. Therefore, actors at the mill wanted to problematize the allocated sales
figures, especially when all the details of the sales plan were not transparent to all actors. In
addition to the mill director, the mill-specific sales forecast involved various actors at the
production sites in the process of interessement, such as the mill controller, production
managers, logistics manager and purchasing staff. Sometimes, the suggested sales forecast
gave rise to debate, and the effects of the forecast were discussed in detail. In particular,
large-scale under-utilization of production capacity was not easily accepted at the mill.
Actors at the mill could resist the plans of the sales organization, and sometimes they led to
significant changes at the mill:
For example, we [in the sales organization] had planned under-utilization, as we saw that there is
not enough demand for that type of business [. . .]. Right away, we got input from them [at the
mill] that in that case they have capacity to sell pulp instead. Then we discovered that there is
demand for pulp in the markets, so we increased the sales of pulp at that mill, even though the
sales of paper were decreasing. So, it is not top-down, but it is rather iteration. (Sales controller, 20
June 2011)
For the paper mill, it was important to see whether the new forecast created changes in
utilization rates and affected the stock and working capital goals. Thus, the sales forecast
was a strong mediator for the various actors at the mill. The mill controller explained how
both sales and production volumes, along with stock levels and prices, were jointly
discussed in a monthly estimate meeting:
At the mill, we have an estimate [RF] meeting at the end of each month. Then I sit down with the
mill director, sales manager, production manager and logistics director. We review how our next
forecast looks and what kind of forecast we are going to send to the business area next month,
and what kind of changes we have. (Mill controller, 9 August 2011)
The sales manager started the meeting by presenting the near-term outlook and the
anticipated sales volume figures for the mill. When the sales figures had been introduced,
the mill management wanted to know what implications this would have for production site,
the production plan and the product mix. In the background of the meeting, the mill
controller was updating the income statement calculation in their spreadsheet computation.
The cost side of calculation was partially based on monthly information, which was
collected from various organizational support functions and contained information such as
raw material prices, energy prices and exchange rates. The interviewed mill controller
MEDAR
31,7
124
explained how they combined these key cost figures with the sales forecast information in
their own calculation during themonthly estimate meeting:
During that estimate meeting, the controller at each mill usually has their own formulated
template in an Excel worksheet, where they update those estimate calculations [. . .], and as the
discussion goes on, we update those estimate figures in the worksheet and are then able to see
how it affects our profit and cash flow. (Mill controller, 9 August 2011)
The delivery of cost information was well organized and collated in electronic form on
SharePoint, a team collaboration software tool connected to the company’s enterprise
resource planning system. The persons responsible for delivering the cost factor information
appeared to be called controllers (e.g. wood supply, logistics and energy supply). Mill
controllers often contacted their colleagues in these support functions, which helped the mill
controllers in knowledge accumulation, as well as in building an alliance network, as these
colleagues understood the functioning of the calculations and the importance of providing
truthful information.
This calculation could be seen an “interessement device”, which divided people into those who
understood its functioning and those who needed the output of the calculations to justify their
operational existence. However, and importantly, forecasting is a recurring exercise that is to a
great extent (but not perfectly) stabilized. The interessement was more linked to aspects of being
a professional who possesses knowledge about the accounting inscription and who has the
capacity to challenge numbers (Ezzamel, 1994; Alcouffe et al., 2008).
The monthly RF meeting at the mill was a culmination point for the entire rolling
forecasting process. The meeting brought together various actors, each of them having the
possibility to affect the direction of the final forecast. In particular, the income statement and
related accounting calculations were important technical artefacts in translating the
information into a form that visualized the implications of the sales allocations for the mill
operations and created a new “truth”. The spreadsheet tool controlled the discussion and
enrolled the interests of various stakeholders into the larger common interests that were in
play (Munro, 1999). After the representatives of sales, production and accounting had
achieved a shared understanding and agreement on the mill forecast, and once the other
mills had prepared their own, all the mill forecasts were uploaded into the company’s
Hyperion consolidation tool. Finally, the business controllers gathered the forecasting
information and prepared an overall outlook for the whole business to be discussed by the
business management and the CFO.
5.3 Promoting quality and accuracy of planning
In Paperco, the accuracy of the forecasting was monitored on a regular basis. At the beginning of
each month, the variances between actual and forecasted figures were analysed. From the
business controller’s point of view, if the forecast was compiled well, it should be fairly accurate.
Accuracy and dedication were seen as the main characteristics of good quality in rolling
forecasting. The group controller of the company used the word “mechanism”, which indicated
that the issue of qualitywas related to thewhole process of forecasting:
I trust that our production sites work so that they try to generate a realistic forecast. So, I sort of
trust this (process), if I dare to use the term ‘mechanism’. In our firm, people want to do things
right, and they also understand what is right. And I know that individual controllers’
performance bonuses can be tied to the accuracy of forecasting: the issue that one has considered
the right things [. . .] [. . .] what may happen in the future and what things have an effect. [. . .]
Therefore, I’m satisfied with the way this process works, even though I know that the
psychological effect might drive towards goal-orientation. (Group controller, 7 June 2011)
Constructing
the accurate
forecast
125
As the quotation indicates, one way of confirming the controllers’ dedication to the
forecasting accuracy issue was that their monetary rewards were tied to the ex post accuracy
measures of the forecasts. Thus,moneywas also an important mediator guiding controllers’
focus on the accuracy of forecasts. However, controllers seemed to keep a low profile
regarding their aspirations to control the process:
This is not a process led by controllers anymore [as in annual budgeting]; also, the whole business
organization works in it, having ownership of the process. We just support and participate in that
process. We sort of coordinate the process. (Business controller, 6 June 2011)
Even though controllers argued that they only coordinated the process, it became evident
that they were also active mediators. It was important for them that the forecast would
mobilize the planned actions so that they would finally achieve their accuracy-related
incentive targets and meet their professional expectations. A mill controller illustrates the
situation:
However, the forecast may bring about different sorts of issues that might need further
investigating [. . .] [. . .] [. . .] we need to start thinking of actions by which we perhaps could do
something to remedy such issues. [. . .] The forecast may also bring about individual more
nuanced subjects of interest that might not be added to the official list of action plans. [. . .] In our
case, the business area monitors each month how well we were able to estimate. Did we hit it
right? It has a direct influence on our bonuses. (Mill controller, 9 August 2011)
In conclusion, we have described two episodes in the rolling forecasting process which show
how a “realistic forecast” is continuously questioned during the forecasting process and how
the controllers at various points of the process perform conscious manoeuvres as a part of
the “forecasting mechanism”. Therefore, controllers’ ability to infiltrate various parts of the
organization and calculative practices seemed to be one way of ensuring that the realized
figures would be fairly close to those that were jointly forecasted. Also, the point that
controllers’ bonuses were tied to the accuracy of forecasting (measured ex post) may lead to
some element of risk aversion in the controllers’ work and in the level of the forecasted
figures.
6. Concluding discussion
In this study, we draw on ANT to demonstrate how rolling forecasts emerge in an actor-
network. We show how the “partial stabilization” of the elements of forecasting provided
simultaneous means to challenge the “conceptions of realistic forecast” while still aiming to
control the hypothetical accuracy of the forecast (Latour, 2005; Quattrone and Hopper, 2006). Our
paper empirically documents two episodes of rolling forecasting, where “the understanding about
what is a realistic figure” was repeatedly translated and challenged to build the outcome of an
accurate forecast. Both episodes indicate that various mediators, such as inscriptions, technical
artefacts, human actors andmoney, played important roles in the process of translation.
Our case study indicates how the expected accuracy of the forecast was an important
issue in a large capital-intensive manufacturing firm operating in fluctuating markets. The
accuracy of the monthly version of the forecast is usually measured by comparing the
previously compiled forecast to actual figures ex post (Jordan and Messner, 2020), and
therefore the perceived accuracy of the forecast can be seen as hypothetical at the time of
preparing a forecast. However, this more anticipatory nature of the concept of “forecast
accuracy” became evident in our case study through numerous courses of action that
various actors were carrying out (Latour, 2005) to improve the quality of the planning.
In the previous section, we presented two forecasting episodes representing the key
points of translation, where the strength of the situational “conception of a realistic forecast”
MEDAR
31,7
126
was contested and the hypothetical accuracy of the forecast estimated. Our findings suggest
that the “partial stabilization” of some elements of rolling forecasting was important so that
the forecast, which was collated from numerous and heterogeneous remote inscriptions in
association with many organizational actors, could be discussed and challenged at the very
centre of events. Lowe and Koh (2007) point out that seemingly trivial elements in the
network may be critical for the credibility of the inscription. We argue that the partial
stabilization of the sales forecast, where some parts of the process could be challenged and
some could not, enabled building trust in an otherwise heterogeneous system, referring to a
system in which various mediators, such as inscriptions, technical artefacts, human actors
andmoney, associate in the process (Callon and Law, 1997).
Prior research discusses how accounting systems never seem to stabilize completely to
fully achieve black box status, rather continuing to develop over time (Quattrone and
Hopper, 2001, 2006; Busco and Quattrone, 2018). In our case, we were able to identify two
forecasting episodes where the constructed accounting inscriptions partially stabilized as
“facts”, mobilizing various actors to discuss and challenge the set of numbers. The first
episode was in the sales organization and showed how the use of a technical artefact (sales
analysis tool) played an important role in shaping the idea of realistic sales figures. Ideals,
such as being “realistic”, are typically abstract entities unless they are operationalized and
enacted (Quattrone and Hopper, 2006). In particular, the historical analysis tool of the sales
team was a powerful mediator in bringing “the absent very much present”, namely, historic
trends (Justesen and Mouritsen, 2011). By modifying the sales figures to comply with the
criteria of historical trends, actors were able to translate the organizational understanding of
what is meant by a “realistic sales figure”. Thus, actors were able to bring the past and the
future into the present (Palermo, 2018) by viewing the current understanding of trends. This
way, figures presented in the historical scale became more valid “facts” and could be agreed
on in sales and the whole business area organization. This way, the sales forecast appeared
to achieve a partial, limited-time black box status (stabilization for now) in the forecasting
process. Nevertheless, we can speculate that the range of possible outcomes of sales volumes
and price might be wide in uncertain environments. This indicates that a forecast becomes
constructed by various types of constellations and shifting connections, which may cause
relational drift (Andon et al., 2007). Thus, the systems cannot be seen to stabilize completely,
as the ways in which the forecast mobilizes other actors change constantly (Quattrone and
Hopper, 2001, 2006). Therefore, the controller, even while being capable of inventing
practical solutions to confirm the quality of information, cannot completely control the
outcome of the process (Lowe, 2001; Andon et al., 2007).
The second episode of stabilization took place in the production site’s forecast meeting
and focused on the concerns of production-side personnel regarding the under-utilization of
production capacity. While the sales organization and controllers had the authority to
orchestrate the allocation of sales volumes between the production sites, their sales forecast
was a powerful inscription, as it mobilized production personnel and controllers at the mill.
One starting point for the second stage of stabilization was the technological platform that
gathered all available information regarding the operations of the mill for the use of
controllers, enhancing the knowledge accumulation cycle (Latour, 1987). The platform as an
inscription device enabled mill controllers to become a “centre of calculation” in the forecast
meeting and objectively represent the operations of a company in a form of simplified Excel
sheet by bringing the “outside in” (Law, 1992; Robson, 1992). As experts in financial
calculations, mill controllers sketched the paper mill’s future, visible to others in the form of
profit and loss, using these calculations as truthful representations of the future (Lowe,
2001). However, the validity of the truth (forecast accuracy measure) could only be assessed
Constructing
the accurate
forecast
127
afterwards (Lambert and Pezet, 2010). From the quality of planning point of view, our study
indicates that the “partial stabilization” of rolling forecasting was a controlled action, where
company controllers, through their calculative apparatus, defined those elements that were
open to challenge and those that were not. Our findings demonstrate the perceived strength
of accounting’s functional network and their inscriptions over the other functional actors
(Lowe and Koh, 2007).
Our findings extend previous studies on interactive forms to improve planning quality
(Østergren and Stensaker, 2011; Bourmistrov and Kaarbøe, 2013), and our paper is, as far as we
know, the first to give amore detailed demonstration of the rolling forecasting process and the role
of a varied group of controllers therein (cf. Goretzki and Messner, 2016). Controllers were in the
centre of the calculation, and there was a strikingly wide group of controllers working in various
support functions (paper mill, wood supply, energy supply, logistics, etc.). The strength of the
accounting network in the forecasting process was its ability to use “obligatory passage points”
(Callon, 1986; Latour, 1987), such as the historical tool of sales trends and the profit calculation,
which combined the concerns of all (the networks of sales, production and accounting) (Lowe and
Koh, 2007). While Bourmistrov and Kaarbøe (2013) point out that rolling forecasting provides
many possibilities for discussion and debate, we specify that the obligatory passage points
in rolling forecasting (in relation to the key problematization stages of forecasting) were
in fact important fabrications in funnelling the flexibility in forecasting and creating a
controlled space for interaction. In line with Østergren and Stensaker (2011), controllers’
calculative practices created room for managers to assess the feasibility of figures as the
decision-making situations were continuously changing. Therefore, the forecasting can
rather be seen as the “interaction of heterogeneous elements that are shaped and
assimilated into an open-ended network” (Law, 1991). This means that the forecast was
mediated through relational attempts to define what it “really” represented (Andon et al.,
2007).
Our findings show that the controllers were rhetorically willing to claim their neutrality
in the rolling forecasting, even though they seemed to be strong mediators in the process.
For example, controllers could initiate investigations for remedial actions in a proactive
manner, if forecasted figures started to indicate deviation from targeted levels. Moreover,
controllers were incentivized by money to promote the accuracy of the forecasts. Monetary
incentives may inhibit controllers from advancing managerial interests in the organization,
which can be a risk with more interactive planning processes in general (Østergren and
Stensaker, 2011). In fact, strong authority of controller function is found to foster risk
aversion (Lambert and Sponem, 2012), which was also evident in our case, as controllers
were reluctant to take risks with excessively anomalous forecast numbers. It rather seems
that attempts were made to construct a “façade of neutrality”, for acting like an intermediary
that delivers impartial information to managers. This finding contributes to Hyvönen et al.
(2008) by indicating how controllers acted in concert with other mediators (accounting
inscriptions, technology, money/resources) to mobilize action.
One explanation for controllers’ claims of neutrality might relate to their willingness to
promote the trust in figures. Accounting inscriptions are typically seen to represent
neutrality but also to build trust in figures; therefore, the controllers may aim to keep the
neutral illusion of their intervention. Thus, the powerful role of controllers is not given or
“invisible” background information but rather becomes evident as an outcome of the actor-
network. Our study points out that when the focus is on measuring the forecast accuracy,
this guides the attention to the issues that are typically important for the accounting
function, that is, how well all the available information has been integrated to compile the
MEDAR
31,7
128
forecast and how well the forecast is finally enacted in the organization (Jordan and
Messner, 2020).
Finally, this was a single case study illustrating a construction of numbers in a firm’s
forecasting process and the role of controllers therein. This is one interpretation of
controllers’ aims of keeping a neutral illusion of their intervention, while being a powerful
mediator in the process. Therefore, we propose more research in other management
accounting settings to shed further light on the role of controllers in cross-organizational
planning processes.
References
Alcouffe, S., Berland, N. and Levant, Y. (2008), “Actor-networks and the diffusion of management
accounting innovations: a comparative study”,Management Accounting Research, Vol. 19 No. 1,
pp. 1-17.
Andon, P., Baxter, J. and Chua, W.F. (2007), “Accounting change as relational drifting: a field study of
experiments with performance measurement”,Management Accounting Research, Vol. 18 No. 2,
pp. 273-308.
Bourmistrov, A. and Kaarbøe, K. (2013), “From comfort to stretch zones: a field study of two multinational
companies applying ’beyond budgeting’ ideas”, Management Accounting Research, Vol. 24 No. 3,
pp. 196-211.
Busco, C. and Quattrone, P. (2018), “Performing business and social innovation through accounting
inscriptions: an introduction”,Accounting, Organizations and Society, Vol. 67, pp. 15-19.
Bøgsnes, B. (2009), Implementing Beyond Budgeting –Unlocking the Performance Potential, Wiley, New
York, NY.
Callon, M. (1986), “Some elements of a sociology of translation: domestication of the scallops and the
fisherman of St Brieuc Bay”, in Law, J. (Ed.), Power, Action and Belief: A New Sociology of
Knowledge?, Routledge, London.
Callon, M. (1998), “Introduction: the embeddedness of economic markets in economics”, in Callon, M.
(Ed.),The Laws of the Markets, Blackwell Publishers, Oxford, pp. 1-57.
Callon, M. and Law, J. (1997), “After the individual in society: lessons on collectivity from science,
technology and society”, Canadian Journal of Sociology/Cahiers Canadiens De Sociologie, Vol. 22
No. 2, pp. 165-182.
Ezzamel, M. (1994), “Organizational change and accounting: understanding the budgeting system in its
organizational context”,Organization Studies, Vol. 15 No. 2, pp. 213-240.
Goretzki, L. and Messner, M. (2016), “Coordination under uncertainty: a sensemaking perspective on
cross-functional planning meetings”, Qualitative Research in Accounting and Management,
Vol. 13 No. 1, pp. 92-126.
Henttu-Aho, T. (2018), “The role of rolling forecasting in budgetary control systems: reactive and
proactive types of planning”, Journal of Management Control, Vol. 29 Nos 3/4, pp. 327-360.
Henttu-Aho, T. and Järvinen, J. (2013), “A field study of the emerging practice of beyond budgeting in
industrial companies: an institutional perspective”, European Accounting Review, Vol. 22 No. 4,
pp. 765-785.
Huikku, J., Hyvönen, T. and Järvinen, J. (2017), “The role of a predictive analytics project initiator in the
integration of financial and operational forecasts”, Baltic Journal of Management, Vol. 12 No. 4,
pp. 427-446.
Hyvönen, T., Järvinen, J. and Pellinen, J. (2008), “A virtual integration – the management control system
in a multinational enterprise”,Management Accounting Research, Vol. 19 No. 1, pp. 45-61.
Jordan, S. and Messner, M. (2020), “The use of forecast accuracy indicators to improve planning quality:
insights from a case study”, European Accounting Review, Vol. 29 No. 2, pp. 337-359.
Constructing
the accurate
forecast
129
Justesen, L. and Mouritsen, J. (2011), “Effects of actor-network theory in accounting research”,
Accounting, Auditing and Accountability Journal, Vol. 24 No. 2, pp. 161-193.
Lambert, C. and Pezet, E. (2010), “The making of the management accountant – becoming the
producer of truthful knowledge”, Accounting, Organizations and Society, Vol. 36 No. 1,
pp. 10-30.
Lambert, C. and Sponem, S. (2012), “Roles, authority and involvement of the management accounting
function: a multiple case-study perspective”, European Accounting Review, Vol. 21 No. 3,
pp. 565-589.
Latour, B. (1987), Science in Action, Open University Press, Milton Keynes.
Latour, B. (1999), “On recalling ANT”, in Law, J. and Hassard, J. (Eds),Actor Network Theory and After,
Blackwell, Oxford, pp. 15-25.
Latour, B. (2005), Reassembling the Social: An Introduction to Actor-Network-Theory, Oxford University
Press, New York, NY.
Law, J. (1991), “Introduction: monsters, machines and sociotechnical relations”, in Law, J. (Ed.),
A Sociology of Monsters: Essays on Power, Technology and Domination, Routledge,
London.
Law, J. (1992), “Notes on the theory of the actor-network: ordering, strategy, and heterogeneity”,
Systems Practice, Vol. 5 No. 4, pp. 379-393.
Lorain, M.A. (2010), “Should rolling forecasts replace budgets in uncertain environments?”, in Epstein, M.J.,
Manzoni, J.F. and Davila, A. (Eds), Performance Measurement and Management Control: Innovative
Concepts and Practices (Studies in Managerial and Financial Accounting, Vol. 20, Emerald Group
Publishing, Bingley, pp. 177-208.
Lowe, A. (2001), “After ANT – an illustrative discussion of the implications for qualitative
accounting case research”, Accounting, Auditing and Accountability Journal, Vol. 14
No. 3, pp. 327-351.
Lowe, A. and Koh, B. (2007), “Inscribing the organization: representations in dispute between
accounting and production”, Critical Perspectives on Accounting, Vol. 18 No. 8,
pp. 952-974.
Lukka, K. and Vinnari, E. (2014), “Domain theory and method theory in management
accounting research”, Accounting, Auditing and Accountability Journal, Vol. 27 No. 8,
pp. 1308-1338.
MacKenzie, D. (2006), An Engine, Not a Camera: How Financial Models Shape Markets, MIT Press,
Cambridge, MA.
Miller, P. (1992), “Accounting and objectivity: the invention of calculating selves and calculable spaces”,
Annals of Scholarship, Vol. 9 Nos 1/2, pp. 61-86.
Munro, R. (1999), “Power and discretion: membership work in the time of technology”, Organization,
Vol. 6 No. 3, pp. 429-450.
Palermo, T. (2018), “Accounts of the future: a multiple-case study of scenarios in planning and
management control processes”, Qualitative Research in Accounting and Management, Vol. 15
No. 1, pp. 2-23.
Quattrone, P. and Hopper, T. (2001), “What does organisational change mean? Speculations on
taken for granted category”, Management Accounting Research, Vol. 12 No. 4,
pp. 403-435.
Quattrone, P. and Hopper, T. (2006), “What is IT?: SAP, accounting, and visibility in a multinational
organization”, Information and Organization, Vol. 16 No. 3, pp. 212-250.
Revellino, S. and Mouritsen, J. (2015), “Accounting as an engine: the performativity of
calculative practices and the dynamics of innovation”, Management Accounting
Research, Vol. 28, pp. 31-49.
MEDAR
31,7
130
Robson, K. (1992), “Accounting numbers as ’inscription’: action at a distance and the development of
accounting”,Accounting, Organizations and Society, Vol. 17 No. 7, pp. 685-708.
Robson, K. and Bottausci, C. (2018), “The sociology of translation and accounting inscriptions:
reflections on Latour and accounting research”, Critical Perspectives on Accounting, Vol. 54 No. 1,
pp. 60-75.
Sandalgaard, N. and Bukh, P.N. (2014), “Beyond budgeting and change: a case study”, Journal of
Accounting and Organizational Change, Vol. 10 No. 3, pp. 409-423.
Østergren, K. and Stensaker, I. (2011), “Management control without budgets: a field study of ‘Beyond
budgeting’ in practice”, EuropeanAccounting Review, Vol. 20 No. 1, pp. 149-181.
Constructing
the accurate
forecast
131
Appendix
Corresponding author
Tiina Henttu-Aho can be contacted at: tiina.henttu-aho@oulu.fi
For instructions on how to order reprints of this article, please visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm
Or contact us for further details: permissions@emeraldinsight.com
Table A1.
The interviews
Interviewee Location Date Duration
Mill controller No. 1 Production site 29 Sept. 2009 1 h 20min
Group controller Group headquarters 02 Oct. 2009 1 h
Business controller No. 1 Group headquarters 02 Oct. 2009 1 h 30min
Business controller No. 2 Group headquarters 26 Oct. 2009 1 h
Business controller No. 3 Group headquarters 26 Oct. 2009 1 h 15min
Mill controller No. 2 Production site 17 Nov. 2009 1 h 10min
Business controller No. 2 Group headquarters 06 Jun. 2011 1 h 15min
Business controller No. 1 Group headquarters 06 Jun. 2011 1 h 10min
Business controller No. 3 Group headquarters 07 Jun. 2011 1 h 10min
Business area controller No. 1 Group headquarters 21 Jun. 2011 50min
Mill controller No. 2 Production site 09 Aug. 2011 1 h
Group controller (new) Group headquarters 07 Jun. 2011 1 h 20min
Sales controller No. 1 Group headquarters 20 Jun. 2011 45min
Business area controller No. 2 Group headquarters 20 Jun. 2011 45min
Sales controller No. 2 Group headquarters 20 Jun. 2011 1 h
Sales support manager Group headquarters 21 Jun. 2011 1 h 5min
Business controller No. 4 Group headquarters 21 Jun. 2011 1 h
IT manager Group headquarters 22 Jun. 2011 50min
Business manager Group headquarters 08 Nov. 2011 50min
BI manager Group headquarters 08 Nov. 2011 50min
Mill controller No. 2 Production site 11 Apr. 2014 40min
Mill director, production
manager and financial manager
Production site, Group
interview 11 Apr. 2014 45min
Business manager Group headquarters 26 May 2014 1 h
Business Controller, Wood
supply By phone 09 Jul. 2014 40min
Business Controller, Energy
supply By phone 11 Jul. 2014 45min
Sales controller No. 3 By phone 02 Jul. 2014 50min
Interviews 26 21 h 45min
MEDAR
31,7
132