May 23, 2024


Digitally first class

5 ways to improve on spreadsheets for business workflows


I recently wrote about five ways spreadsheets kill your business. Spreadsheets are a versatile tool for performing quick, ad hoc data analysis, developing presentation materials, creating small knowledge bases, and collaborating with small teams. But when spreadsheets become operational tools in recurring business processes, the manual effort required to maintain data quality and develop workflows can be costly and error-prone.

When spreadsheets become part of an ongoing process, it’s time for everyone using them—business leaders, solution architects, data scientists, and software developers—to review migration options to more robust platforms.

Given that spreadsheets are the Swiss Army knives of office tools, the suitable  alternatives for modernization span several different types of platforms. Some spreadsheets can be replaced with configurable software-as-a-service tools, while others require a more customizable solution. Low-code options enable developers to more easily create and support enhancements, and no-code options empower users to develop their own solutions.

For larger-scale and more complex workflows, development teams might consider custom-developed applications, business process management (BPM) platforms, or other enterprise application development frameworks. Once a workflow fulfills a domain-specific or industry-specific need and becomes used by multiple departments, it’s worth reviewing platforms specific for the job at hand rather than replacing spreadsheets with lighter-weight tools.  

Depending on the data types, workflow, presentation requirements, security considerations, and collaboration required, different platforms are worth considering. In many cases, platforms address more than one type of use case, and many of the ones I list can fall into multiple categories.

Here are five approaches to replacing spreadsheets that draw on different platform types. Most organizations will need several options to cover their broad spreadsheet use cases.

Build a data visualization or dashboard

If a spreadsheet is largely a conduit to display charts and graphs, one option is to rebuild the data visualizations in a self-service business intelligence (BI) platform. The best approach is to rediscover the underlying requirements and rebuild dashboards that serve business needs rather than porting the existing charts into a new platform.

Developers and citizen data scientists should consider identifying user personas, documenting the questions dashboards will answer, addressing data quality issues, and following data visualization standards. Some platforms are stand-alone data visualization tools, but many enable developers to embed visualizations directly in applications and other platforms.

Platforms to consider include Domo, KNIME, Looker, Power BI, Qlik, Sisense, Tableau, and ThoughtSpot.

Migrate to a SaaS or no-code database

If the spreadsheet is storing original data for a knowledge base or workflow, you’re likely to need a database to migrate and manage the data. Assuming this is structured data (sheets of rows and columnar data), you could import it into a managed public cloud database such as AWS Relational Database Service or Azure SQL, but then you’ll be left with additional development work around workflow tools and dataops.

Another option is to review SaaS and low-code databases that allow you to create the database structures, load data, develop role-based workflows, and create dashboards. Platforms to consider are Caspio, Kintone, and Quickbase.

Outside of SaaS, low-code, and cloud databases, architects and IT leaders should also research domain-specific data lakes and data warehouses that support full, end-to-end workflows. Examples include customer data platforms for customer data, AIops solutions for IT operations, marketing automation platforms, and enterprise search platforms. These platforms enable easier connection to the common data sources for their domains and then provide a mix of workflow, machine learning, and reporting capabilities.

Enable departmental collaborations and workflows

Let’s consider a use case where a department or team uses spreadsheets to manage a workflow. The workflow could be a marketing team’s editorial calendar, a production department’s Kanban to manage work intake and fulfillment, or a field service team tracking their jobs. These spreadsheets typically store just enough data to manage the workflow but lack the configuration to track and govern the process.

SaaS and no-code tools focus on enabling a hybrid of workflow and collaboration capabilities. Platforms to consider include Airtable, Asana, Jira Work Management, Monday, Smartsheet, and TrackVia.

For larger scale and more complex workflows, you’ll want to review domain-specific platforms with data models, workflow configurations, and embedded best practices. These include a wide range of platforms, including enterprise resource planning, customer relationship management, content management, IT service management, agile collaboration, talent management, and financial planning. 

Hyperautomate integrations and data flows

Sometimes, spreadsheets are used as intermediary steps in a data flow, integration, or low-level automation. A marketer might use a spreadsheet to merge email lists, or an IT ops engineer might have one for documenting the steps to restart an application server. There are many spreadsheet-in-the-middle use cases where someone connects two ends of a flow and performs some manual work in the middle.  

Several different types of platforms can replace the spreadsheet-in-the-middle use case:

  • Extract, transform, and load (ETL), data prep, and other dataops platforms such as Alteryx, Informatica, Tableau, and Talend support data flows and pipelines.
  • Robotic process automation (RPA) tools such as Automation Anywhere, Blue Prism, and UiPath can automate information extraction from websites, SaaS, and other sources.
  • Citizen integration technologies support if-this-then-that integrations between SaaS, cloud, and other tools, including Zapier, IFTTT, and
  • Integration platforms support data, API, and workflow integrations and include robust platforms such as Boomi, Celigo, Jitterbit, MuleSoft, and SnapLogic.
  • IT automation, quality assurance test automation, CI/CD, and configuring infrastructure as code (IaC) are examples of IT platforms used to consolidate scripts, manual work, and spreadsheets into repeatable and automated processes.

Build a low-code or no-code application

If a spreadsheet takes on multiple functions, including loading data, processing it, enabling collaboration, and presenting results, you may need to develop an application to replace it. Of course, you can code an app, deploy it to a serverless architecture, and maintain it through an agile development process. But one reason these workflows evolved using spreadsheets is that IT can’t easily keep up with the business demands for new apps and improvements.

Today, a significant share of apps is developed on low-code and no-code platforms. Consider reviewing my previous posts on seven keys to selecting a low-code platform and the seven low-code platforms developers should know. Some platforms that help developers and citizen developers build spreadsheets into apps include Appian, Betty Blocks, Bizagi, Caspio, Claris, Creatio, Google AppSheet, Mendix, Microsoft PowerApps, Oracle APEX, OutSystems, Pega, Quickbase, Retool, Unqork, and Vantiq.

Before you abandon that spreadsheet

Although there are many options for replacing spreadsheets, keep in mind that apps can’t easily replace the openness and all the flexibilities of using them. People love the versatility of spreadsheets and the ability to add columns, change data, and create formulas on the fly—something that you’ll likely control when migrating the data and workflow to other platforms. It also means people must get used to new experiences, tools, and user interfaces, along with the idea that aspects of what people did manually will now be automated.

Before getting started, it’s important to collaborate with the primary users. Unwind how they do things today and rebuild the requirements on what they are trying to accomplish with an eye toward the business purposes, quality, and scale. Only then can you explore platforms, prototype approaches, develop solutions, and evolve improvements.

Copyright © 2022 IDG Communications, Inc.


Source link