Are you feeling the impact of technical debt on accelerating DevOps,
delivering innovations, and driving digital transformation?
shows organizations are spending almost as much time addressing technical debt
(28%) compared to running operations (38%) and building new capabilities
You already know that I am a strong proponent of citizen development,
low-code, no-code, and citizen data science. Recent articles include
five reasons architects should embrace low-code and
why citizen development is core to digital transformation 2.0.
In this post, I connect how low-code / no-code is a strategy to reduce
technical debt – both existing debt and the risk of creating new sources of
technical debt. I share my insights and insights from other tech leaders.
And scroll to the end to see my latest
5 Minutes with @NYIke video on how
I use no-code technologies to capture all my articles, store them, and deliver
this dashboard to browse all of them. Watch the full video as I share five recommendations on developing no-code
apps, databases, and integrations at the end.
My Insights on How Low-Code Reduces Technical Debt
Abstractions in low/no-code apps help to build capabilities with less
expression and more guard rails. It requires avoiding adding custom code to
the low/no-code app and only doing so where absolutely required. So most
low/no-code apps have less or no tech debt by design.
App modernizations from code (w/ tech debt) to low/no-code
(less/no-tech debt) drives the reduction. Keep in mind that many legacy apps
often have a lot of code for implementing UX, automating integrations, and
addressing performance issues that can be out-of-the-box with low/no-code
Using low/no-code can change the conversation w/ stakeholders.
Instead of getting “must-have” requirements that often require custom code,
POCs, and using sprint reviews to demo capabilities in low/no-code, help
draw out simplified MVPs by asking stakeholders, “will this work?”
With citizen development, a business user is more likely to implement
what works more efficiently. Example – a dashboard built by SMEs rather than writing requirements for
dev or data science teams and then iteratively building one that meets the
But citizen development requires a governance model, and
citizen data science should be paired with a proactive data governance
Test automation is also code and has its own technical debt.
Automating testing in a pro-code app often requires unit tests and
functional tests that are orchestrated in CI/CD as continuous testing. But
as any test engineer will tell you, one small change in the UX can sometimes
require a full rebuild of the dependent tests. Low/no-code platforms can
help reduce testing, and they are less likely to have dramatic UX changes
version to version. This helps reduce the debt associated with maintaining
I shared these insights during a Twitter chat some time ago and knew I had to
share them here.
Insight from Tech Leaders on Low-Code Opportunities
And many other tech leaders embrace using low-code technologies.
I spoke with Rosaria Silipo, Principal Data Scientist and Head of Evangelism
at KNIME, on her
perspectives around using low-code tools. She says, “Low code tools accelerate
the development of applications. There is still a technical debt in the sense
that testing, documentation, optimization, etc., must all be implemented and
run at some point. However, the development phase itself is faster, leaving
more time for the remaining tasks, bypassing the code learning part, and
therefore avoiding some typical coding mistakes.”
Gev Hovsepyan, head of product at
Mabl, knows that testing
apps – whether built with low-code, no-code, or pro-code also need low-code
tools to build and maintain test automations. He says, “As more software
development organizations adopt agile and DevOps, manually maintaining a
spreadsheet of all testing activities is going to slow time-crunched QA teams
down. Modern test automation solutions that provide comprehensive reporting
and insights can make it easy to track and manage automated testing at a
My Demo: A Blogger’s Database and Dashboard
Years ago, I maintained a spreadsheet with links to all my blog posts. But as
I contributed more posts, articles, and videos to different websites, it
became difficult to maintain. I wanted to search the articles, plus I had the
sharing a dashboard with readers to help find all my articles.
So in this video, I share how I use
Quickbase, Tableau, Zapier, Talend Stitch, Google Analytics, Squarespace, and AWS RDS
to build an end-to-end flow. Trust me, it’s not complicated, but you see the
result of several iterations, and the MVP was much simpler.
And please watch till the end of the video to get my five recommendations on
developing no-code apps, databases, and integrations.