The Gap Between Intention and Impact

Contingent responsiveness is a concept from developmental psychology that describes the difference between a parent who provides and a parent who listens. The research is fairly unambiguous. Children who feel genuinely heard develop more securely than children who are simply well provided for. The gap between those two outcomes is not effort. It is direction. Whether the care is oriented toward what the parent believes is needed, or toward what the child has communicated they need. Intention rarely maps onto impact as neatly as we assume. Sometimes a child just needs a hug. Not piano lessons. Not a structured weekend routine. And no amount of correct provision gets you there if you never stop to ask.

Output is what you deliver. The outcome is what changes because of it.  These two are not the same thing, even when they look identical from the outside. You can do everything right and still miss entirely. The gap between what we intend and what the recipient experiences is one of the oldest problems in human organisations, and it has a way of hiding behind very impressive-looking numbers.

Let’s cast our minds back twenty years. No Chief Data Officer. No data engineering practice debating the merits of a data build tool over stored procedures at 2 pm on a Tuesday. The work existed, but it was tucked inside IT or finance, handled by whoever had the closest thing to a spare hour. Then came the insight, correct and important, that organisations were sitting on genuinely valuable information and needed specialists to unlock it. Data was a resource, just like capital or talent. Colleagues of mine have written about this at length, and the framing still holds. What nobody quite planned for was the function slowly turning inward. The goal quietly shifted from delivering value to the business to maintaining the infrastructure that might one day deliver value to the business.

Which is how you end up with thirty-four dashboards shipped and four opened. A pipeline running at 99.9% uptime. Hundreds of Jira or monday.com tickets closed. Five hundred catalogued assets that nobody is using to make a single decision. The scoreboard is green and has been green for months, and somewhere down the hall, a senior leader is making a call based on gut feel because the data they have been handed does not quite answer the question they are asking. This is not because the work was bad, but because the wrong question was answered very, very well.

The honest reason this keeps happening is not incompetence. Data teams came out of IT, and IT taught them something valuable and something limiting at the same time. Valuable: ship clean work, close the ticket, maintain the system. Limiting: the job ends at delivery. Whether anyone used it, whether it changed anything, whether the person at the other end got what they actually needed, that was never really part of the loop. And going out to find out is uncomfortable. Business stakeholders are direct. They will tell you something is not working in a tone that does not invite discussion. They will change what they want mid-build. They will occasionally treat the data team like a vending machine. Most people find creative ways to avoid that dynamic. So, the gap widens, quietly, one green dashboard at a time.

The quarterly earnings call version of this problem is well known to anyone who has sat in a room and watched a deck of metrics that technically tells the truth while practically obscuring it. Everyone knows. Nobody says so. The quarter closes. The underlying thing gets deferred to next quarter, where it will be deferred again with equal professionalism.

Closing it takes two things, and neither involves buying new software. The first is measuring what actually matters. Not how much was shipped, but how much was used. Not necessarily pipeline uptime, but how often did the work change a decision. Not tickets closed, but insight adoption. These numbers are harder to get and more awkward to present, but they are the only ones that tell you whether the function is doing what it was created to do.

The second is shared ownership. The DMBOK, which is as close to a definitive text as the data industry has, speaks at length about this. It draws a clear line between data stewards, responsible for quality and definition within a domain, and data owners, accountable for data as a genuine business asset. The distinction matters because ownership without engagement is just governance theatre. A title on an org chart does not make someone accountable. Business leaders need to co-author the questions being asked, the decisions in scope, and how success gets measured after delivery. Not as a courtesy. As a condition. This is where governance stops being a compliance exercise and becomes the thing that actually holds the strategy together. A dashboard that does not change how someone acts is not an asset. It is inventory.

At Robinson Ryan, we ask the tough questions. As consultants we hold ourselves accountable not just to what we deliver but to the change that delivery makes possible inside your business. Output matters. But outcome is why we are here.