Sometimes life throws a curveball into the game, and it’s no more business as usual. My dad died seven months ago with the love of his life, my mom, at his side. He had a very long battle with Alzheimer’s, which had already changed him profoundly — and us, too. Now he’s at peace. In the fast few weeks, as my family faced the inevitable, we were constantly reminded of what kind of man he was: a true leader, fair and decent and genuine. Anyone who had the good fortune to work with him responded to that.

As I was developing my own knowledge base on what galvanizes talent into something called employee engagement, I was constantly struck by my dad’s example. Inspired by his example, his colleagues and teams worked harder, stayed later, treated each other better. They were never under any pressure to do so: My dad’s culture was not “my mission must be your mission.” His people did it because they wanted to. What motivated them: feeling valued.

Of course there was no way to measure this, but he had radar, forged from his own intelligence and experience. No surprise that it correlates with some of the best ideas on engagement out there now.

So here are four meaningful ways to measure employee engagement:

1) Stop pretending it’s about satisfaction. Ask what your people need and want from their employer. You are not selling something to them, you are collaborating with them — and you depend on their interest in collaborating back. So ask what your employee needs to stay. Don’t use a multiple choice, closed survey question to find out; allow them the room to articulate what they’re after. Again, this may be a disconnect related to our lack of imagination about the essence of an employer brand.

2) Ask appropriate and targeted questions. Here’s an obvious irony: Surveys take time out — from projects and tasks and the usual requirements of an employee’s day. Which means you’d best make it worth your employee’s while. Make sure you’re asking questions that are relevant to your employees and their work. There are an infinite number of possible questions to ask about engagement, but research is showing that perhaps a dozen are truly useful.

Consider the nature of each department’s work and the kind of performance it requires. If you are wasting their time on irrelevant questions, they might just decide to embark on a job search instead. And time isn’t what it used to be. A DICE survey found that in high tech, two-thirds of all workers believe they could land a better job in less than 60 days.

3) Don’t take your talent for granted. High performance comes from those who feel highly respected. Gallup found that as of January of this year, some 32.5% of workers are engaged in their jobs, but 51.9% are not. Worse, 15.7% are actively disengaged. So if your workplace is already struggling and it’s clear there are problems with performance, check your survey at the door. First, do some initial repair work to even get the car on the road. If you are already dealing with a disengaged workforce, throwing a survey at them that asks for tangible data on engagement will seem even more inauthentic – even tone-deaf — and only serve to send them on a search for greener grass.

4) Frame it in the positive. Two possibly undesired byproducts of an engagement survey: 1) that you reveal to the employee just how disgruntled he or she really is in the process of answering those questions, and 2) that the survey feels so non-transparent that it further disengages. But the goal of the survey should be clear to all — aka, transparent — that the employer’s intention is to improve the workplace, and support the workforce, in order to make them feel better about working as hard as they do. Present the process in the spirit of collaborative purpose and a generous workplace culture, not a standardized test, and you may just wind up with a great win-win.

Here’s to Dad, a sage judge of character, who would doubtlessly approve.

A version of this was first posted on Forbes.