The Impact of Audit Technology on Audit Task Outcomes: Evidence for Technology‐Based Audit Techniques*

As audit technology becomes more widespread, practice and academia are raising concerns about the costs benefits of these technologies. We examine how internal auditors use technology-based techniques (TBATs) TBATs impact efficiency effectiveness their audits. two surveys interviews individual chief executives (CAE) to perceptions TBATs. Auditors perceive as beneficial. Specifically, an increase in is associated with completing audits, finding risk factors, providing recommendations, decreasing days. However, CAEs also be costly. An size function. Finally, suggest that not used often because difficulties quantifying benefits, observing a timely manner, hiring appropriate skills. Overall, stand tasks, but struggle quantify net cost-benefit tradeoff. Our findings validate issues raised by both proponents opponents technologies help provide empirical data inform decision-making process regarding future tools. Additionally, our study prompts several avenues for research can regulators, practitioners, researchers on impacting auditing profession. L'impact de la technologie d'audit sur les résultats des tâches : preuve l'utilité vérification basées Alors que se développe, milieux professionnel et académique soulèvent inquiétudes quant aux couts avantages ces Les auteurs examinent comment auditeurs internes utilisent (TVBT) TVBT ont un l'efficience l'efficacité leurs À l'aide deux enquêtes d'entretiens avec individuels dirigeants l'audit interne (DAI), ils TVBT. perçoivent comme étant avantageuses. Plus précisément, une utilisation accrue est associée à réalisation d'un plus grand nombre d'audits, découverte facteurs risque, formulation recommandations diminution du jours d'audit. Cependant, DAI considèrent également sont couteuses. Une augmentation l'utilisation taille fonction interne. Enfin, entretiens suggèrent ne pas utilisées souvent en raison difficultés quantifier avantages, observer temps opportun engager possédant compétences appropriées. Dans l'ensemble, susceptibles d'accroitre d'audit, mais mal leur rapport cout-avantage net. Ces permettent valider questions soulevées par partisans opposants contribuent fournir données empiriques pour éclairer processus décisionnel au regard l'avenir outils. De plus, cette étude propose plusieurs pistes recherches futures qui peuvent contribuer informer autorités règlementation, praticiens chercheurs l'impact profession This examines influences efficiency, effectiveness, costs.1 Technology becoming increasingly important aspect external audits clients become sophisticated users technology; face significant pressure reduce fees, better able perform tasks.2 Although adopted goal improving costs, studying this phenomenon settings difficult relative lack archival sets. Prior attempts address issue finds no results or even mixed (we review literature section 2). In line findings, find there less adoption at firms than anticipated (Lovata 1990; Janvrin al. 2008; Vasarhelyi Romero 2014; Eilifsen 2020). gap—technology holding promise evidence realization positive outcomes—has led numerous calls additional area (Earley 2015; Wang Cuthbertson Alles Gray 2016; Wood Moffitt Austin 2021). To knowledge, among first examining proprietary set auditors' influence actual outcomes field. Although, glance, it would appear relatively obvious should improve outcomes, reasons why may case. First, enhance audit, rather redeploying captured gains into coverage, organization could budgets (for audit) cut fees auditors) so overall quality does improve. Second, effective, must only purchase advanced have personnel capable using successfully. The accounting currently faces challenges technological expertise, given high demand pay individuals skills other disciplines (Christ 2021; Maksymov Third, cost companies willing pay. For example, Christ (2021) demonstrate automating inventory counts drones dramatically improves effectiveness; however, still little field “drones expensive” (Vien 2018, para. 32). three complementary approaches gather we conducted survey 268 auditors, asking beliefs (perception data), company's function (IAF) (overall respondents' most recent engagements (engagement various measures tasks. another 505 CAEs, (in terms budget) companies' IAFs, whether increases decreases costs. interviewed 11 gain further insight decision adopt evaluate TBATs' benefits. from show make efficient effective—and result robust across amount TBAT usage organization, rank respondent (chief executive not), IAF respondent. non-perception collected supports perception findings. one standard deviation 18.5% 10.8% 12.3% days necessary complete 13.5%. words, higher effective audit. When looking specific phases reduction number prepare conducting fieldwork, reports. they do consider achieve Therefore, analyze second wherein ask 15.5% (i.e., auditors). open-ended responses factors barriers implementation, top five implementation competence issues, operating access issues. split respondents usage, IAFs low much concerned financial already level. negative usage. greater (our auditors), come CAEs). Given different respondents, unable benefit shed light decide invest assess TBATs, CAEs. interviewees noted largely responsible deciding management board dictate decision) generally informal analysis. analysis quite complicated time all difficulty assessing Interviewees sometimes value non-monetary which always recorded engagement teams; ignored when indirectly related acquisition such retraining employees new technology, operational downtime installation maintenance, on. Furthermore, seeing implementing take long time, partially skill sets properly implement Thus, some likely measuring challenge qualified realize potentially problem identifying measurable over technology. helps engagements. contributions literature. contribute expect see improvements adopting technologies; exceed large public investing billions (EY 2018; KPMG 2019; PwC 2019a), quality, will necessarily lower especially immediately after introducing Since many people thus up interesting investigate key stakeholders (e.g., managers, members, regulators) respond increasing if savings expect. fee setting adversarial, understanding dynamic. identify reason highly auditing—identifying quantifiable difficult. contribution paper identifies objective, sample firms. science measurement, encourage develop ways measure settings—especially companies. Audit managerial enable governance contexts. around used. gathering propriety those might clear-cut choice exposure hype emerging solving problems (Austin 2021), suggests explore person, task, environmental impactful outcomes. provides (Prawitt 2009, 2012; Ege Abbott Bills 2022; Jaggi 2022). continues mature organizations, work launch organizations' management, Increasing tasks receives attention professionals (PwC Hood 2018). know tangible tasks' efficiency. Professionals' mixed. Some express optimism technology's impact, believing task Braun (2017, 41) report “that successfully implemented sustainable analytics activities been clearly visualize articulate deliver functions broader business, started enhanced awareness.” contrast, others optimistic auditors. A Gartner (2018) notes risks surrounding tools Protiviti (2019) few organizations indicate teams understand importance innovation core IAF. (2019b, 15) reports “struggle right fit” work. result, repeated practitioners academics (Bierstaker 2001; Ramamoorti 2003; Brown-Liburd Cao Earley Griffin Wright Yoon AICPA 2017; Richins Gepp Almost universally, commentaries call pessimism impacts warranted.3 prior client's investments mixed, IT investment client lags (Johnston Zhang 2018) increased probability issuance going-concern opinions, likelihood Type II errors (Han 2016). Additional Chen (2014) similarly shows capability mitigates has effect delay. Lim clients' decreased delays. There studies firm- engagement-specific handful extent Dowling 2009; 2014). (2020), points used—but enhances Li within IAF, performance directly specific, engagement-level complements being stages Extant effectiveness. allow direct testing transactions, (Ballou 2020; Barr-Pulliam Cardinaels Emett addition, allows displaying unique interpret evidence, procedures performed without (Dilla 2010; Jans Rose Anderson Pickard Loraas Holt Eulerich (see hypothesis), any realized redeployed testing, (KPMG CA Worldwide 2017). (Brazel 2004; Commerford Holmstrom believe literature, whole, leads hypothesis. Hypothesis 1 (H1).The positively planning sources assessment (Lynch Koreff traditional efficiently (Dowling Leech Cooper 2019, M. examples, (2022) robotic automation reduces takes calculate distances evaluating compliance tests mileage reimbursement, documents downloaded enterprise resource (ERP) systems, taken extract information ERP systems. Practitioners suggesting digital confirmations, natural language processing systems legal contract analysis, application programming interfaces speed extract, transform, load processes (Davenport Raphael IDC Research suggested (Maksymov Negangard us 2 (H2).The It effects costly, need upgrade latest tools, training needed educate apply specialized hired numbers salaries what required Alternatively, hours procedure; insights assessments targeted, smaller program; entirely remove certain performed; level assurance similar procedures, reducing gathered. dynamic, following non-directional 3 (H3).The create set, surveyed auditors.4 developed consulting academic practitioner receiving feedback working German chapter Institute Internal (IIA). final version consisted open- closed-response respondent's engagements.5 Questions cover details type nature engagement, engagement. provided definition participants conceptualized consistently uniformly. Survey were annual conferences national IIA chapter. One authors distributed copies each events. Approximately 1,100 conferences, achieved approximately 24% participation rate. From obtained 205 observations run (column (3) Table 3). mentioned, asked engagements, yielding possible 536 (if data) described detail later, effectiveness/efficiency level: (i) found, (ii) recommendations provided, (iii) total days—we break out parts model analyzing (recommendations) (total days), 300 (340) (361) included regression model. CAEs.6 was collaboration between extensively pilot tested before collecting headquartered Germany, Austria, Switzerland. contacted received electronic invitation participate hyperlink survey. All who took part completed anonymously. 4,009 invitations sent participants, response rate 12.6%. Respondents include broad range industries different-sized answered full. CAE responses, 349 organization. outcomes: budget proxied employees. (IAF size), 309 (320) reached personal contacts specifically selected ensure wide representation based On average, had 20 years experience, lasted, 27 minutes. semi-structured interview approach Appendix script). summarize things learned later paper. accepted (Arena Azzone A. Conceptually, deals degree “performs way accomplish objective” (Dittenhofer 2001, 447). ways. department complete. audit's coverage scope. conduct areas improvement, matters concern main independent variables interest, teams' during six process, including schedule, performing analyses, reporting More specifically, question engagement: “How you your stage audit?”7 then answer seven-point Likert scale (1 = very 7 high). averaged together Engagement Level, engagement.8 equations (2) (3), Controls vector control full description variable). team worked (Auditing Days). indicator capture entity audited Subsidiary, Plant, Project, Process). focus, variables: Financial Focus, Operational Compliance Managerial Focus.9 Assurance variable controls relates project. (Team Size) scope (Audit Size). object's (Risk before) (Performance object Also, Employees), head functionally supervisory committee (Report Board/AC), IAF's (Assurance PCT), outsourcing (Outsourcing average (Training same company characteristics equation (1). (3). Before discuss models above, depict analyses 1. objective dependent variables, like primary useful view potential TBATs.11 opinion ranging (very little) much). 1, panel presents