Let's say you're asking an employee to research the internet for the the best accounting software. Let's say you are giving him some guiding criteria what should be understood by "best".

Is there a way to know if he has done all the things necessary to discover the best software, since the output of his work is fairly unquantifiable?

How could you know if used the full array of keywords to google the internet, how could you know if he thoroughly investigated all the options, if he cared to test them, if he kept his attention focused on important details, etc?

The only way to check the quality of his work, which comes to my mind, is to do the same research yourself and then compare your results with his results. Of course, this approach makes the whole point of delegating the task meaningless. A similar approach is to assign the task to a couple or more employees and compare their outputs. Cost is the drawback in this case.

In general, what are the methods to control the quality of work with unquantifiable output, such as the one in the example?


Normally what you measure in such cases is not output - since that is by your own definition unquantifiable - but the skill, motivation, and intellectual qualities of the person, and their apparent allegiance to your own goals. The presence of these things engenders what might be called "trust".

The assumption then is that if a person claims to have spent, say, 40 hours on a task, then they will have achieved whatever a person is capable of achieving in that time, or (for creative tasks where results aren't always guaranteed) have engaged in a reasonable search for a result.

You can't verify the result without doubling up on the full work, and in some cases verification is simply not possible, because whatever the second person comes up with may not be commensurate with the first, and whether that is attributable to fault by person or to the haphazard nature of the task will never be clear.

Moreover, excessively intrusive attempts to verify work may impose burdens which detract from the quality of output (for example, by competing for time or mental energy, or producing a dysphoric mood due to the nature and unpleasantness of the burdens, or the extent to which they interrupt a train of thought or course of physical activity), or undermine the trust or alignment of goals if the intrusiveness is to such a degree as to create insecurity.

There remain a large class of tasks in life where effectively the only possible measurement is the trustworthiness of the person performing the task, and trustworthiness is what must be assessed and promoted in such cases.


Ask for a proof of work.

If you want him to compare softwares ask him for a report in any form on such items.

If he can test those ask for a live presentation to prove that he familiarised himself with the task needed and have some mastery of it after putting in some work.

If the information that you are asking for cannot be retreive he will be able to say that he could not find such a thing after searching on x sites or whatever else.

It's not the quantity that can be displayed but the quality of work.

  • Let's say he prepares a table of 20 pieces of software, with some comparisons, ticks and crosses, etc. How can I be sure he has not missed another 5 pieces which could possibly be even better than those 20 presented? – drabsv Oct 3 '20 at 15:26
  • 2
    @drabsv just do a quick search afterward to see if he got the ones that you found. You can also ask him to find every accounting software of the work and compare them and their version. I'm not sure what are your expectations but you can either trust him or do the work yourself. There is also no point in being unreasonable and asking him to document every software conceived since the 1990s. – Al rl Oct 3 '20 at 16:31
  • 1
    Well, your answer is good, but creating proof on anything you do can take a toll too, making the task trivial and documentation hard. – Gintas Oct 3 '20 at 17:18
  • 9
    @drabsv if your employee creates a extensive comparison table/report where he/she compares 20 different pieces of software and you are still afraid he/she did subpar work, then obviously delegation isn't your thing. You should either fix that or do everything yourself. – thieupepijn Oct 3 '20 at 18:05
  • 2
    @drabsv “The perfect is the enemy of the good”. If his solution is “good enough”, then you should probably go with it. If there are clear industry-standard options that are being rejected, you should be clear about why, but you don’t need to, and very much shouldn’t, exhaustively investigate every possible option before making a decision. – Joe Stevens Oct 4 '20 at 6:44

Some proof of work as suggested by Al rl is a good start, and I'd recommend that too.

On top of that there is one general strategy to avoid overlooking valid options or getting an evaluation right: Let multiple people do the same evaluation independently. Then compare the results and perhaps have them discuss their points of view if they differ.

One common risk in such tasks is that someone goes into the research with a fixed understanding of the problem (and perhaps search parameters) in mind, but a slight change of that understanding could open up another branch of options. Say you ask someone to look into databases and they assume relational databases, ignoring any non-relational (noSQL) options. This is something you can either gauge yourself when seeing the list and notice a whole category is missing or protect against by having multiple people do research independently in the hope some look at the problem from a different angle. In general the idea is the same as for brainstorming etc. You get different viewpoints into the pot. Note though: Obviously this is costly, you basically multiply the cost of the task by the amount of people that you let do it in parallel. This is a common trade-off: More quality - more initial cost; and via the better quality hopefully less cost or more earnings later. Or just better sleep to have made the right choice. Still you might want to ask yourself first how mission critical the right software choice in this case is or whether it's fine to go with a potentially suboptimal option and switch later should its drawbacks ever get in your way.

In addition: If you somehow doubt the honesty of your employee, don't let them do mission critical tasks (alone).


You are literally asking how to quantify the unquantifiable. The simple answer is that you can't, by the very definition "unquantifiable".

How could you know if used the full array of keywords to google the internet

You can't prove a negative. It's impossible to prove that someone exhausted all possible search queries. The onus is on you to put forward a search query that (a) is meaningfully relevant and (b) has not been tried yet.

At best, you could log everything they've looked up. But it is literally impossible to prove that they've exhausted every possible option.

how could you know if he thoroughly investigated all the options

You can't. And even if you could somehow know that they investigated the right options, how would you determine that they understood the information correctly?

if he cared to test them

... I mean, have you tried talking to your colleagues?

But again, even if they tested, how would you determine that they performed the right test, and drew the right conclusion from it?

if he kept his attention focused on important details, etc?

You do realize the outlandishness of what you're asking, right?

You're literally trying to pick your colleague's brain, bypassing their entire consciousness and communicative ability in favor of essentially tracking their every move and their conscious understanding so you can judge their work without having to do any of the legwork to gain the needed topical knowledge.

The only way to check the quality of his work, which comes to my mind, is to do the same research yourself and then compare your results with his results.

Take a step back here and think about what you're suggesting. You're implying that you're guaranteed to do the research correctly. If that is the case, why didn't you do it yourself in the first place?

I can even express that into quantifiable boolean algebra. If A is your colleague's research and B is yours, then (A and B) or B always reduces to B. In other words, no need to involve A to begin with.

If you do not trust people to do a good job, then don't rely on them to do that job.

A similar approach is to assign the task to a couple or more employees and compare their outputs. Cost is the drawback in this case.

This issue boils down to how you can have multiple people working on a given task without the cost of having multiple people working on a given task. Again, that's impossible by the very nature of what you're trying to achieve.

If you mistrust your colleague's work quality so much that you'd rather have it double-checked by others, then you have to put forward the effort/cost of having the work double-checked.

You can't eat your cake and have it.

  • " You're implying that you're guaranteed to do the research correctly." - not exactly. The difference of me doing the research is that I am at least motivated to do it thoroughly and I won't slack. Therefore, that would be a test to check not to check if the employee performed well or not, but a test to check if the employee decided to slack and put as little effort as possible. – drabsv Oct 8 '20 at 6:51
  • @drabsv: Right. So if you can do it better, and you're going to do it anyway, why not just do it yourself then and not bother people whose input you clearly don't trust? – Flater Oct 8 '20 at 10:22
  • Too many wrong assumptions here. I do not know whether I can do it better or not; only thing I know for sure is that I would not do it carelessly. Thinking how to control the quality of employee work is not the same as not trusting them. As soon as you have employees, your job is to control their performance. Performance may be poor for a myriad of reasons, not related to trust. Your question is like asking why are there QA departments in companies instead of managers trusting their employees. – drabsv Oct 8 '20 at 12:47
  • @drabsv: You're not asking how to have these employees report their findings to you, you're asking how to confirm your second-guessing of their findings. Specifically for an unquantifiable task, the resolution of that task, and therefore its provable correctness, cannot be quantified. If you were to write down your acceptance criteria (e.g. must have tested X, must have searched for Y), then you have inherently quantified your task, therefore meaning that the task was never unquantifiable and rendering the question moot. – Flater Oct 8 '20 at 13:19

I came to a conclusion, quite close to Steve's answer - what should be monitored in this case, is not the output, but rather the dedication and the narrow research skills:

(1) is the person intelligent and skilled enough to perform such a task? This is fairly easy to check by assigning a number of smaller and easier tasks. Basically the skilled one would spend time refining search phrases after the initial search; the poor performer would use whatever results have appeared after the first search and would stop there. This task definitely requires good categorization (information architecture) skills;

(2) does the person have the habit to go the extra mile to perform well or just on the contrary, relentlessly seeks to do only the passable minimum? This one is tricky. Even generally dedicated employees may not pay attention to detail and add extra observation and thought, to a task they do not find interesting. Maybe bringing an element of challenge could help - e.g. put forward an inadequate piece of software and list its top ten flaws, and have the employee use it as a benchmark. Discuss those top ten flaws with the employee, watch if that discussion sparkles his interest. I would expect naturally curious people to perform better, but even that is tricky - someone might show high level of curiosity in sports, let's say, while being generally lazy and apathetic about most other things in life.

And once you have made sure (1) and (2) are ok, I can hardly think of anything else that could be done. Maybe small checks such as taking a title of a researched piece of software, typing "xxxxx's alternatives" and checking if titles not present in the employee's list would appear.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .