By Paul Walker, Global Solutions Director, iManage
Is the enterprise getting carried away with trying to solve complex problems with AI, when there are numerous simpler, everyday tasks that could immediately be improved with the help of this technology?
Ask any finance professional about the pain points they regularly encounter, and they will gush at length about how frustrating they find mundane and time-consuming tasks such as cleaning up formatting in a Microsoft Word document, or fixing merged cells in a spreadsheet.
These tasks often interrupt deeper analytical work, forcing professionals to shift mental gears and lose momentum. Even more irksome, clients are rarely willing to foot the bill for this kind of work.
In parallel with more ambitious applications of AI that aim to tackle complex, higher-level challenges, perhaps it’s time for AI to squarely take aim at these “low value” tasks to generate immediate efficiencies. Shifting AI’s focus towards these “lower-value” tasks isn’t a lowering of the bar – it’s a strategic reframing of AI priorities that will ultimately be to the benefit of finance professionals and the organisation at large.
The real time sink is repetitive admin work
Many financial tasks – like reconciling spreadsheets, formatting decks for client or investor presentations, renaming files, logging notes, and handling compliance forms – remain manual and repetitive, taking up significant time for analysts, advisors, and other busy finance professionals.
AI can quickly improve these standardised, low-risk activities. Errors such as mislabeling a folder or misformatting a slide have minor effects, unlike mistakes in risk calculations or regulatory interpretations, which carry much higher stakes.
AI vendors often target high-profile tasks like strategy creation or market analysis with their tools, but many finance professionals prefer to handle these tasks themselves. It’s the part of the job they enjoy. What they want is relief from time-consuming administrative work, such as manual data reconciliation. AI vendors should prioritise automating repetitive, frustrating chores over flashy features with no real-world impact on how work gets done every day.
Let “the machines” handle the mundane
Consider the daily grind of preparing client-ready materials. A junior associate might spend hours aligning bullet points, adjusting font sizes, and cross-checking figures between Excel and PowerPoint.
These tasks are ripe for automation – not because they’re unimportant, but because they’re predictable and rule-based. AI can handle them with speed and accuracy, freeing up human bandwidth for more nuanced work.
Likewise, developing tailored investment strategies for high-net-worth clients represents meaningful work, while inserting standard boilerplate text into quarterly reports offers comparatively little value.
Even within critical activities such as market analysis or risk assessment, there are repetitive components that lend themselves well to automation. It is essential for finance companies to clearly identify which pieces of the workflow fall into this category.
Meanwhile, for their part, AI vendors should be capable of articulating what particular task their AI solution addresses – e.g., reconciling data, updating assumptions, verifying formula references, etc. – and what efficiencies can be realised. Clear articulation by the vendor – rather than broad but vague claims about “streamlining finance workflows” or “empowering finance professionals” – also allows finance teams to better align AI-enabled automation efforts with measurable ROI.
What’s the tradeoff?
Ultimately, the goal is for AI to empower finance professionals to focus on more strategic responsibilities – but consideration must also be given to potential trade-offs involved with automation.
Some manual processes inadvertently produce value. For instance, assembling documents for audits may seem monotonous, but it fosters a deep familiarity with key information, often benefiting team members during client discussions or strategic planning.
Automating such tasks might diminish this incidental expertise. As a result, when introducing AI, it is important to ensure that any value that is inadvertently discarded when manual processes are automated is recaptured by alternative means. People, process, and technology all need to work in concert to make sure the proverbial baby isn’t thrown out with the bath water.
“Small” can be big
Low-value tasks often get overlooked because they lack excitement, but automating them is a path well worth taking. Targeting these tedious, repetitive tasks helps build trust in AI’s usefulness every time it reliably tackles one of these small but vexing tasks: renaming a file, formatting a table, or flagging inconsistencies in a report.
To be clear, financial enterprises can simultaneously take aim at more complex and strategic problems with AI. But these small AI wins are a worthy end in their own right, and as they accumulate, they create a foundation for broader adoption centered on higher-level, high impact use cases. That’s proof positive that AI doesn’t have to be big to be impactful.
