How low can you go? Part 2

How low.jpg

A short introduction to low code platforms

Low code platforms and solutions are attracting lots of attention, and span integration, database applications and app development platforms.  They promise to bring technology closer to the business, removing the need for costly development projects and empowering end-users.  But can low code really deliver on these ambitions? 

In the first article in this series I looked at integration tools. Now I’m turning my attention to what is perhaps a more fundamental building block - low code data platforms.

Before I start I have an admission to make- I’ve become a big fan of one offering in particular!

Spreadsheets or Databases?

I was recently discussing digital workplace offerings with a non-profit client of ours. Today most of their activities are run from spreadsheets. The question that they were grappling with was a familiar one - how should they scale?

A CRM system may be the answer – and there are plenty of good, and even free offerings available. But moving from a spreadsheet – with full access to all data – to a CRM system where the only data access is either via the GUI or via a CSV extract can be a difficult transition to make.

But sticking with ever more complex spreadsheets is probably not the right approach - while spreadsheets are an incredibly powerful tool, they aren’t great at relational data. Large data models with different hierarchies and relationships can end up as a tangled mess of “VLookups” and redundant data.

Databases, by contrast, are (obviously) better at managing relational data, but most remain the preserve of application developers, taking control away from end-users. Traditionally the “gateway” database for such users was MS Access, but over the last few years several new SaaS options have become available and have the potential to be game-changers.  Stackby, Airtable and Ninox are a few notable options, but there are plenty of others.

A migration tracking tool in Airtable

Airtable has Unicorn status, as a start-up recently valued at over $1bn. It combines a familiar spreadsheet-like GUI with the ability to create truly relational database models along with data entry forms for basic applications.  Out of the box the product supports the creation of dashboards with native charts, summaries, and pivot tables, and it includes integration with Vega-Lite should you have the need for more complex graphs – although frustratingly these dashboards can’t be published to external sites or users.  Automation, including scripting, is also supported and adds to the power of the product.

When SharePoint lists proved unworkable for tracking a client’s migrations, I turned to Airtable and have now created a multi-user migration scheduling and management tool, which provides Gantt chart views, Kanban board views, and reporting capabilities to easily understand the status of more than 1,000 application environment migrations from seven on premise datacentres to the cloud.

Similarly, I have a financial tracker that allows us to manage budget, forecasting, and actual spend from vendor proposals to purchase orders to invoices.

Low code won’t be the answer for everything

Building similar migration and financial tracking models in a spreadsheet would have been extremely complex, requiring duplication of data which may have compromised the reporting, and almost certainly employing some of the more exotic features available in Excel.

But spreadsheets remain powerful tools, and where your underlying data model doesn’t have one-to-many or many-to-many relationships, then they may still be the most appropriate way of meeting your calculation and reporting requirements.

In that sense, the usual rules apply: spend time thinking about requirements, come up with a design (however basic) and then select the best tool for the job.

Finally, as with low code integration tools, put some thought into who will maintain the system longer-term, particularly if the models being created are put in place to support an operational rather than a purely project requirement.

Part 3…

Having provided a brief introduction to the capabilities of low code platforms in integration and data, in the third part of this series I will be looking at how far low code can go in providing a comprehensive application development solution for your business.

Previous
Previous

How low can you go? Part 3

Next
Next

How low can you go? Part 1