How Airtable Makes AI Practical

Why flexible applications and well-structured data make it easier to introduce AI into everyday business operations.

Most business software now claims to have AI capabilities. For established SaaS providers, this often means adding AI features to an existing product, with capabilities largely determined by the vendor's roadmap.

The problem is that they are not necessarily designed around the specific processes, information or decisions that matter to your organisation.

At the other end of the spectrum, teams still manage important operational work through spreadsheets, email and manually maintained reports. Introducing AI into these processes can be difficult when information is fragmented, inconsistently structured or dependent on individual knowledge.

We think there's a significant opportunity between these two extremes: flexible business applications that give organisations control over their workflows, data and how AI is introduced.

1. Introducing AI where the work happens

One advantage of flexible systems such as Airtable is the ability to introduce AI directly into existing workflows, rather than waiting for a software provider to develop a particular feature.

Airtable's native AI capabilities make it relatively straightforward to add intelligence to applications that teams already use.

At Rapid Indigo, we’ve added :

• Document processing: Extracting key information from PDFs and other documents into structured records.
• Customer feedback: Identifying themes and sentiment from customer responses.
• Management reporting: Producing summaries of project activity, progress and outstanding issues.
• Account scoring: Assessing and prioritising accounts using available customer information.
• Natural language interaction: Using Airtable's Omni to help users explore data and work with applications - think replacing static dashboards by being able ask questions of your data.

None of these necessarily requires a major AI programme. They can be introduced as relatively small improvements to existing processes.

Crucially, the business isn't limited to whichever AI features a particular software vendor decides to offer. Teams have more freedom to identify opportunities and introduce capabilities that address their own requirements.

2. Making your operational data accessible to AI

One of the more exciting developments is the ability to connect AI tools such as ChatGPT and Claude directly to business systems using the Model Context Protocol (MCP).

Instead of exporting data to spreadsheets or building reports, you can ask questions or even build agents using the information in Airtable using natural language.

For example:

• Ask questions across your data: Which projects are behind schedule? What activities need attention this week?
• Explore trends: Compare complex, segmented data across multiple years, answering questions that would otherwise require time-consuming queries and analysis.
• Combine information: Cross-check Airtable records against emails, invoices and accounting systems to spot missing payments, identify discrepancies and update outdated information.
• Investigate problems: Explore relationships between projects, resources and activities to understand dependencies or identify potential issues.
• Create reports: Generate summaries, comparisons and analysis without having to build a new dashboard for every question.

Why the underlying system matters

The quality and accessibility of your data make a significant difference to what AI can do with it. We have used Airtable to help dozens of organisations organise information into structured, related tables, with straightforward interfaces that help business users keep records accurate and up to date.

Compare this with an ageing specialist SaaS application, where data can be difficult to access, or a collection of spreadsheets where relationships and business rules aren't clearly defined, you already on the backfoot and throwing AI at data isn’t going to yield the results you want.