Not all work generates value.
For example “keeping the lights on” work helps maintain value. It does not generate it. In fact, by maintaining the status quo, you may actually be reducing value as the rest of the world moves on. What you do may actually mean less. It may mean quite a lot to the business at the time, but if your competitors are doing things better, cheaper, faster, then in the grand scheme of things, it isn’t much, thus a reduction in value.

You also have work that reduces value. Have you ever worked on a project that was canned, or a feature that is never used? There can be the argument that you are creating “value” as you are producing an output, but if no one uses the feature or the project is canned before anything produced is utilized, you do not get any revenue. Yes I know that value doesn’t necessarily mean revenue, but revenue can be a good measure. If no revenue is generated, the work has a cost, and the output of that cost is lost. The value of that work then is negative.

Sometimes the value can be hard to find. Training your staff on a new technology vs sacking the lot of them and bringing in new people is one way to get a skill quickly for the company, but at what cost?  The knowledge of the people. You loose whatever knowledge those workers had that were sacked. Tacit knowledge that isn’t easily transferred or learned, even if handed over to replacements. You may also gain the reputation as a company that treats workers like throwaway objects, resulting in it being harder to find people in the long term, at least not without offering significant incentives such as a very high salary, and even then you may find it hard. Increased turnaround time as new workers continually have to learn aspects outside their initial core knowledge. Every organisation does things differently. Core knowledge of a language or technology doesn’t mean that they will be able to hit the ground running straight away.

Value is a finicky thing. You know it when you see it, but if you have never seen it, it’s hard to know what it is.

In order for your work to generate value, you have to be constantly vigilant. You have to be constantly finding ways to reduce the work that doesn’t bring benefits, the time and money sucking work. Not only that, but you have to do it in a constantly changing environment. Work that generated value at one point in time, may become a burden the next. That work may be necessary, but you need to constantly find ways to reduce or eliminate that work. With the time left over, you need to find ways to increase value, which in turn generates more work. It is a never ending cycle.

So, how do you know if your work is generating or sucking value? IMO that is through measuring the outcome. The sooner you release a piece of work the sooner you can determine if it has produced value. Then you can decide whether or not to continue enhancing that piece of work, or throwing it away and loosing the cost of doing that work, and since the output of the work was measured at an early stage, that loss becomes minimal.
For the “Keeping the lights on” work, what happens if it is stopped? Does it stop the business from functioning? Does it reduce the value of the output? If not, keep it going, but look at ways to keep the same “value” at a reduced cost in both time and money. Notice I said value – not output. For example, automating a workaround to a bug that was previously done manually keeps the same work output, its just done quicker. Fixing the root cause of the bug changes the output of the work by eliminating the work, but the value (keeping the system running) remains the same in both cases.

Work has a life cycle, sometimes that cycle can be minutes, sometimes it can be years. To determine the at what stage the the work is in its life cycle, you need to look at its value. You need constantly evaluate the work. constantly check it. Can it be reduced, changed or eliminated to give value the same or more value?