(Sorry for the length, I am in an avuncular and effusive mood, so longwinded advice is where anything I write is headed a the moment.)
I will answer the question, but I do not think your example is an example of the problem you are claiming.
Slippery slope is a version of a broader category: correlation does not imply causation. And the solution is to require a theory that establishes the causation.
Every real prediction requires an observed correlation and a theory.
If you don't have one, it is possible that the correlation implies in the reverse direction, or that it is linked to a common factor and there is no direct causation applicable to you argument, or, though more rarely, that the observed correlation itself is simply the result of random coincidences.
Slippery slope arguments lack a theory that makes the likelihood of valid extrapolation logical. So when you find this kind of argument in progress, seek that theory.
If the likelihood of progression actually is logical, this is no longer a slippery slope fallacy, but a weak scientific assertion.
You can make it a strong scientific assertion by establishing its falsifiablity: by pointing out what further facts it predicts, and whether those seem to also be true.
So, where is the slippery slope in your argument? You are claiming that dismissing a minimal ROI leads to dismissing greater ROI's until ultimately the only alternative is to have infinite ROI by having a zero denominator, by making zero investment. So what is the theory that makes you believe he will reject any intermediate ROI? Is it already falsified by his previous behavior? Probably. So this is an unsound argument. My tactic works, but the application seems odd.
The idea there is a 'slope' determined by a single point already makes it a pretty silly argument, and that is a more straightforward way of looking at this weakness. Prediction from a single data-point is already far worse science than jumping to conclusions from a trend. If you had a list of things he had rejected with various levels of ROI to predict his behavior from, it would at least qualify as a slippery slope argument.
As it is, the real fallacy here is more basic. It is 'splitting', 'black/white thinking' or catastrophizing -- a reflexive blindness to the relevance of any middle ground. Seeing it that way more directly motivates checking whether an intermediate ROI would equally be rejected, and might keep you from putting forward your untenable position (probably insulting to his wisdom.) (Just looking up Godwin's Law, splitting seems like a much more likely relative of an argument lost to "forfeit 'by Hitler'" than to a slippery slope.)
Primarily, you are not predicting progression, you are ignoring degree. He probably has some minimal notion of an acceptable ROI, and if you can look at the company's history, you may be able to find an example of a lower ROI which he found acceptable and a theory for why yours is better than that past investment, or you may be able to suss out what that minimum ROI projection needs to be before you offer future projects.
Many people safe from a slippery slope argument are still given to making outright splitting arguments. In psychoanalytical terms, the former is an aspect of worry, which comes from a fairly nuanced level investment. The latter is an aspect of panic or anger from an excessive or unstable level of investment. Even though they are logically quite close, getting good at avoiding one of these will not help the other very much.
Avoiding slippery slopes, as I pointed out, answers well to the basic aspects of trying to make the argument into good everyday science (from a Popper point of view) since there is data.
To learn to avoid a habit of splitting, you might want to look and the layperson's equivalent of the therapeutic approach known as Dialectical Behaviorism. In that state of mind, you try to develop a habit of knowing when to abandon confrontational stances and reframe arguments dialectically in the sense of Hegel or Marx.
To proceed dialectically, you have to clearly identify your position (have a grounded Thesis), develop empathy for your adversaries position (frame its Antithesis), and purposefully search relevant ground unaccounted in either position (approach a Synthesis). Either his position or yours will better stretch to incorporate the uncovered ground, and that position will often be the better one to pursue.