What’s wrong with your Airtable database and how to fix it?

Airtable databases often have poor data design and lack of standardization

I've spent the past few years consulting organizations that use Airtable. I'm going to share what I consider the big issues I've come across, some thoughts on why these problems are nearly always present, and how you might remedy them.


Airtable is a cloud-based, no-code/low-code, relational database system. It has gained in popularity because its spreadsheet-like interface allows anyone to create a database that can be accessed and updated by multiple people.

In truth, you could have a database up and running, with multi-user access, in an hour or less.


Airtable has a growing list of useful features. Here are three I find compelling.

Views: Airtable allows you to create custom views for your database, allowing you to filter, sort, group, and organize data into digestible components. You can have views for each group of users, allowing them to see just the data they need.

Form/Interfaces: It has an interface designer to give you greater control over data input and edits. The interface designer has continued adding features, allowing a high-degree of flexibility and rapidly generated forms.

Automation: It allows no-code automations that let you generate emails and other updates based on conditions. Should you need to go deeper, you can add javascript automation to connect with external data and automate additional elements of your database.


All this simplicity has a dark side. Because anyone with reasonable technical ability can create a database, ANYONE with reasonable technical ability WILL create a database. Because of this, many (most I've come across) organizations suffer from two major issues with their Airtable database.

1) Database Design/Schema

Yes, you can easily create links between tables. However, for many organizations the person who builds the database may be technical but has little, to no, knowledge of database design.

The result is either repeated data or several (I've come across dozens) unnecessary fields. This makes standardizing data entry and reporting nearly impossible.

Spreadsheets often contain column after column of fields that, with proper schema design, should be in a separate, related table. But the spreadsheet-esque nature of Airtable - while simple for most users to jump into, lends itself to spreadsheet-esque challenges with the data.

In short, because it looks and feels like a spreadsheet, it is typically treated like a spreadsheet.

2) Standardized Naming Conventions

Again, a byproduct of rapid implementation, column and table names are often random and follow no standardization. Similar to bad schema design, this can be confusing for those using the database. Additionally, should you implement automation down the road, non-standard naming becomes a chore.



Hire me! Done!

Okay.. you are NOT required to hire me.

However, it is likely that for good database design, you would do well to get input from someone with expertise in data design. I've taught database design to several developers over the years and it is not a trivial skill.

Someone who has had to correct bad design - their own bad design - views data differently than others. Once you've had to work around the headache's of bad data design, you approach it thoughtful. Taking a little extra time to create the right schema will save you hours and frustration down the road.


Also, I strongly suggest you create some mock-ups of reports you believe will be important to you.

Don't think in terms of what you've done so far and certainly do not use a rough Airtable design to create your report layout.

Rather, think in terms of what an ideal system would tell you about your customers, your business, your sales, etc. Whatever it is that your Airtable database will track. Create a few report ideas (hand-sketches is fine) and then, looking at the report, dissecting data that should be separated into related tables becomes a bit simpler.


You can certainly learn more about your data and how it will need to be structured by working with the data. However, taking some time to list out all the types of data you need to capture before you create any tables, will mean you spend far less time correcting mistakes down the road.

In the mad rush to create a way to store our data, we often build a system, any system. Later, we know we need to fix it but find that we are now knee-deep into using the data. Fixing our data becomes a "we'll get to it" proposition.

Truth is, many organizations never get to it.

Take time and front-load the work. You'll be happy you did.


Airtable is an amazing tool and application. You can do a lot with it. I have a few clients and their Airtable systems perform virtually all the heavy-lifting for their business.

Yes, I find it's built-in reporting & document assembly to be limited. We often integrate our client's Airtable with Google Workspace and Google BigQuery for true, high-performance dashboards and more robust document or spreadsheet creation and distribution.

But, if your growing organization or department needs a rapid development relational database, you might want to check out Airtable. You can get started for FREE - which is a pretty good price.

Posted in Automation and tagged , , .

Leave a Reply

Your email address will not be published. Required fields are marked *