Back to Insights
EssayLinkedIn

March 24, 2026

Why should you "Shift Left"?

By Suha Beidas Zehl, CMB®, AMSP™

Cover image for the article Why should you "Shift Left"?

Shifting left is more than a tech term – it really is a mindset

The phrase comes from software engineering. As a technologist, I am very familiar with this term. On a project timeline, the earlier stages sit to the left. "Shifting left" means moving quality checks, testing, and problem-solving earlier in the process rather than catching failures at the end when they’re expensive and hard to fix.

In AI, shifting left is the difference between adoption and abandonment

Most organizations buy the tool first and ask questions later. They sign the contract, announce the initiative, and then hand it to their teams and wonder why nothing changes. The implementation happens. The adoption does not.

That is a right-heavy approach to AI, and it is one of the most expensive mistakes I see.

Shifting left in AI means doing the readiness work before the rollout. It means assessing your workflows, your data, your people, and your processes before you commit to a solution. It means asking the hard questions about change readiness, about who will be affected and how, about what governance needs to be in place before the technology touches a real decision.

It means your people understand the why before they are handed the what.

Teams that feel like AI was done to them rather than with them become the resistance that kills the ROI. And once that culture hardens, it is very difficult to walk back.

Shifting left in AI is not about slowing down. It is about investing time at the beginning so you are not spending twice as much time undoing a failed rollout later. The audit, the gap analysis, the roadmap work: that is not bureaucracy. That is the difference between a tool that gets used and a tool that collects dust.

The organizations that get this right treat readiness as a prerequisite, not an afterthought.

In mortgage, we talk about it in income calculation

For years, the industry standard has been to calculate income at the point of decision. A file gets submitted, an underwriter reviews the documentation, and income gets calculated. If something is off, the file stalls, the borrower gets frustrated, and costs go up. The error lives at the end of the process.

Shifting left means calculating income upfront, at origination, with consistency and accuracy baked in from the start. It means the data that drives the decision is verified before the decision ever has to be made. It reduces friction, reduces repurchase risk, and creates a cleaner path to close. It also removes the human inconsistency that has quietly been one of the industry’s most stubborn compliance problems.

This is not just efficiency. It is a fundamentally different philosophy about where quality lives.

In fraud prevention, the same principle is changing the conversation

A recent piece in American Banker covered a session at RSAC 2026 where cybersecurity experts made exactly this argument about elder financial exploitation. Seniors lost $4.9 billion to scams in 2024 according to FBI data, and the problem is that banks typically only see the fraud at the very end of the process. By the time a customer tries to wire money to a scammer, the damage is already largely done. The victim is emotionally committed. The scammer has done months of work. The bank is the last stop.

The experts argued that banks need to shift left. To collaborate with telecom providers, tech platforms, and government agencies earlier in the scam timeline. To intercept the fraud before it reaches the transaction. Because by the time it becomes a transaction, it is often too late.

One presenter put it plainly: detecting a scam is one thing. Preventing a loss is something else entirely.

We do the same thing in our own lives. We wait until a relationship (business or personal) is broken to have the hard conversation. We wait until our health forces us to make changes. We wait until burnout to ask for help. We wait until the crisis to reflect on the choices that created it.

Shifting left in our personal life looks like the conversation you have before resentment builds. The healthy habit you start before the diagnosis. The reflection practice you build before you lose your sense of direction. The boundary you set before you are depleted. The clarity about your values before someone else decides them for you.

It is not about being anxious or constantly anticipating problems. It is about being intentional. About asking yourself: where am I waiting to act until it costs more than it should?

Quality, safety, clarity, and care belong at the beginning of the process, not as a rescue effort at the end.

Whether we are talking about a mortgage file, a senior customer’s retirement savings, or your own wellbeing, the question is the same: how much further left can we go?

Because the earlier we engage, the more we can actually change the outcome.

“Shifting left” sounds simple, but it’s actually a pretty fundamental change in how organizations think about work. Most teams are still operating in a right-heavy model, especially when it comes to AI. The income example is spot on. We’ve spent years accepting that validation happens at the point of decision, when in reality that’s the most expensive place for it to live. The same pattern shows up everywhere: data comes in, AI processes it, and then a human rechecks it “just in case.” That’s just acceleration with a safety net! The real unlock is what you’re getting at: moving validation, reconciliation, and clarity upstream so the system can actually be trusted downstream. We’ve been spending a lot of time on this exact topic over the past few months. The organizations that treat readiness, data integrity, and workflow design as prerequisites are the ones that actually see adoption and impact. Everyone else ends up circling back after the fact.

Link to the American Banker article on "shift left" and fraud prevention: https://www.americanbanker.com/news/banks-urged-to-shift-left-to-fight-elder-fraud

Suha I think this statement in your post matters a lot "In AI, shifting left is the difference between adoption and abandonment" -- because waiting on AI projects is a legacy habit from the SaaS mindset ... 'let’s wait until the software matures more then, we’ll buy/adopt it' ... but AI is faster and more iterative, so your statement implying 'just do it' is really important.

This article series opened with a pattern I kept seeing in different forms. The budget was approved.

My Dad loves to watch news in the morning. Today was no different.

Putting This Into Practice

This is the work I do with lenders, banks, and credit unions. If your organization is working through these questions, these pages explain how I help.