A few things to keep in mind.
Generally, you would never expect a long power reserve movement, such as the 7-days 5001, to be as precise as a movement with a regular power reserve. Why? Because the long power reserve is driven by an extra-long mainspring, or more than one mainsprings that must be switched between using a transmission. In the first case, isochronism error becomes a big problem, as there is an exaggerated difference in tension when the mainspring is fully wound versus unwound. In the second case, you get "lumpier" results and introduce added mechanical complication and future maintenance challenges. Most modern long power reserve movements use the multiple mainspring solution. The single mainspring solution is mechanically more elegant, but makes regulation very challenging due to isochronism error. In the instance of the 5001, I suspect the single mainspring solution was absolutely necessary since it is an automatic movement (I believe it remains the only automatic long power reserve movement in existence).
Anyway, isochronism error is the problem faced by the 7-days 5001. To regulate the movement optimally, the watchmaker must account for how it is likely to be worn. If worn everyday, it will stay fully-wound and the watchmaker will bias the regulation toward high amplitude. If it will only be worn every few days, and the mainspring stays in a more relaxed state with less tension, he will have to regulate it for more moderate amplitude. Anyway, that's why you can get such wildly variable timekeeping with the 5001 Portuguese. You need to tell your watchmaker how you wear your watch in order for it to be correctly regulated. Also, if you change the way you wear it frequently, it will be near impossible to regulate it that well.
Is this a flaw with the movement? Arguably. However, I like the single long mainstream approach and think it is at least a more innovative and elegant one. Originally it was thought that they could overcome isochronism error by simply never letting the mainspring fully unwind (there is a cutoff point)--that is why you get 7 days as opposed to 8 or 9. Yet, that clearly didn't do the trick.
High-end watches like the 5001 are all about heritage, design, and horological interest. I find that I care much less about accuracy/precision when I understand the movement better, including why it was designed the way it was, what its advantages are, and what trade-offs had to be made. The truth is, there is no other movement that does exactly what the 5001's does, and it was designed specifically for application in the first ground-up modern Portuguese. That's a story. It's a "pure-bred" watch. Hence, despite its temperamental nature, it will likely always retain value and desirability. Like a Ferrari.
For certain other watches, an in-house movement is not as important. Take Panerai, for example. Why? Because Panerai was not originally a manufacture. Until recently, they had always used ebauches from outside sources. So, it really doesn't damage the purity of a Panerai to not be in-house. It's just not part of the story.
However, IWC, along with a handful of others (Patek, JLC, Audemars, Vacheron, Rolex, etc.), were the only manufactures to survive the quartz revolution intact and continuously produce in-house mechanical movements. As such, they hold a certain pedigree other brands don't, and it will always matter to collectors whether their watches have in-house movements. Patek/Audemars/Vacheron famously relied on the very high-grade Lemania chronograph ebauche for decades. Yet, over the past few years, Patek has striven to bring chronograph design in-house. That is a very costly move and no accident. It's a matter of reputation.
Though various versions of the Portuguese have not always used in-house movements, the original Portuguese always did, and was special for the size of the movement it contained. So, it's arguable that for any Portuguese to stand up to collector expectations, it must have an in-house, pocket watch-sized movement. The prices on the secondary market are demonstrative.