ALNA
EN
English עברית
BOOK DEMO
IMPLEMENTATION & ORGANIZATIONAL LEARNING

Why a Good Recommendation Can Still Fail at Implementation

What stands between a recommendation that's right on paper and a change that actually happens

Shani Shimshian, social worker and therapist, founder of ALNA · Updated September 2026

The recommendation was approved. Six months later, nothing had changed

A program director receives a clear diagnostic report: the problem is identified, the recommendation makes sense, and leadership approves it in that same meeting. Everyone leaves feeling good, there's a direction, a decision, a way forward. Six months later, when someone checks what actually happened, it turns out almost nothing changed on the ground. The same people are still working exactly as they were before, and no one can quite pinpoint the moment the recommendation stopped being a priority.

That does not necessarily mean the recommendation was wrong. The gap may lie between the decision and its implementation. These stages are easy to collapse into a single meeting, even though each requires separate attention.

A good recommendation is not a guarantee of implementation

In the field of Implementation Science, a professional discipline concerned with how sound decisions turn into change that actually happens, it's standard to distinguish between two entirely separate stages: reaching a well-founded decision, and then implementing it on the ground. These are two different tasks, with different difficulties, and often different people responsible for each one. Whoever is skilled at diagnosing a problem and building a precise recommendation isn't necessarily the person who should, or has the time and tools to, oversee its day-to-day implementation.

This gap is sometimes known as the "know-do gap", the distance between what experts or leadership know is right to do, and what actually happens day to day. It isn't a sign of failure on anyone's part; it's a natural result of the fact that decision-making and implementation happen in different contexts, under different time pressures, and often among people who weren't even present when the decision was made.

The professional literature suggests: the larger the gap between who makes the decision and who is expected to carry it out day to day, the greater the chance the recommendation stays on paper, even if it is entirely sound professionally.

A few common reasons unrelated to the quality of the recommendation

1. No measurable target behavior was defined

"Improve coordination" is a direction, not an action you can observe and check for. Without a concrete definition, each team member interprets the recommendation slightly differently, and sometimes does nothing with it at all.

2. Ownership is diffuse

When a recommendation belongs to "everyone," in practice it belongs to no one day to day. Without one clear owner, no one is actually responsible for making sure the change eventually happens.

3. The change adds load without removing existing load

A stretched team tends to revert to the old way under pressure, even after agreeing the new way is better. A change that expects staff to "do one more thing" without giving something up tends to erode first.

4. Feasibility wasn't tested against real work

When a recommendation isn't checked against real working conditions before rollout, it's only on the ground that it turns out to clash with an existing process, another system, or an already-packed schedule.

5. There's no ongoing feedback

Without a simple way to know what's happening on the ground, no one notices implementation has stalled, until someone checks the outcome months later, by which point it's hard to know exactly what happened along the way.

The difference between "the hypothesis was wrong" and "the change wasn't implemented"

When an outcome doesn't improve after a change decision is made, there's a natural tendency to conclude that the recommendation, or the diagnosis behind it, was wrong. In organizational behavior management (OBM), standard measurement checks a different question first: was the change actually carried out as planned, how often, and how well. An implementation failure is a completely different explanation from a hypothesis that wasn't supported by evidence, and the distinction between the two determines whether it's right to abandon a direction or simply fix how it's being carried out.

This distinction matters especially in care and human service organizations, where a change decision is often made at the leadership level, but its day-to-day execution depends on dozens of small interactions between staff and clients or service recipients. Without checking that execution layer separately, it's very easy to attribute to the plan itself a failure that actually originated in how, or whether, it reached the ground.

At ALNA, here's what we do: every recommendation is accompanied from the outset by a defined target behavior and implementation metrics, so it's possible to clearly distinguish "not implemented" from "implemented and didn't work", before deciding on next steps.

What helps a recommendation survive implementation

No formula can guarantee successful implementation. A few conditions, however, make it more likely. Start with one observable target behavior. For example, replace a broad instruction such as "improve communication" with "update case status within 24 hours of any change." Name one owner for day-to-day implementation, even when the decision came from a wider group. Check feasibility with the people who will carry out the action before it becomes an official instruction.

The fourth condition, and perhaps the most important, is planning in advance how and when to check whether the change is actually happening, not only whether the final outcome improved. An early implementation check makes it possible to course-correct within weeks, instead of discovering six months later that the whole effort went off track.

Combining human and technological support during implementation, not only during diagnosis, is what makes it possible to spot early when something stalls and fix it before the momentum is gone.

Why this happens, and why it repeats

Implementation failure often repeats because a decision meeting and day-to-day work have very different conditions. The meeting offers time to explain, discuss, and approve. Daily work brings a ringing phone, an urgent case, and a form that still needs to be completed. A recommendation that sounded clear in the meeting now competes with everything else that feels immediate.

That's why the distinction between diagnosis and implementation isn't a theoretical matter. It determines whether an organization actually learns from every improvement process it goes through, or repeats the same pattern again and again, a good decision, initial enthusiasm, and a quiet fade with no one quite sure why.

Sources & Further Reading

Have a recommendation that isn't taking off?

In a short introductory call, we can help figure out whether the gap is in the diagnosis or the implementation, and what's worth doing in either case.

Also worth reading: How ALNA Thinks · How to Tell If the Change Didn't Work, or Simply Wasn't Implemented