They lose no customers by including it. They lose some by omitting it.
So it boils down to being too expensive to include? Hardly!
You evaluate prior decisions with posterior data. But you fail to take into account the counterfactuals. How do you know how much the FP4 would have sold with a jack?
Claiming that an increase in sales validates the goodness of the decision is not causal.
It is the same logic that would tell you that playing russian roulette is worthwile in case you win and get some reward. That’s backwards rationalization, fitting a narrative.
If market research universally showed that people don’t care about a jack then why do some phones still have it? Are these manufacturers going against the grain? Surely they wouldn’t leave money on the table if it worked like that.
The justification of “they do what sells units” is backwards. It would imply that no product would ever flop. But products regularly do. There is no telling in advance how it will perform, and saying otherwise is falling prey to the problem of induction, whether past observarions justify predictions.
The FP4 could have broke sales records for a multitude of reasons. How can you say which factor caused it when there is only one scenario that played out? We don’t have alternative universes to compare, where they released one with a jack, or another with some other altered specs.
I’m back to statistical significant data, and why it is important to have good data scientists in the loop. The idea is precisely to ask the questions you are asking. Would have been different if…? Then try to control for other variables in order to avoid the induction error. How do you know they didn’t do this with their data?
That’s why I mention other phone models. There are Sony phones with and without jacks. There are Asus phones with and without jacks. How did they perform compared to each other? How far away is that difference from what could be expected from randomness? How does that difference compare when the other factors are compensated for? How do they compare with other phones?
I assume they did their homework, and also want to sell more earbuds. They wouldn’t push for earbuds and wireless if headphone jacks were market drivers. It would be cheaper to install a headphone jack rather than updating the BT board? Maybe, I don’t know. But if other factors have a significant impact on sales while the jack doesn’t. Then they have their decision made for them. Market research is not about being right all the time, it is not magic, it is about reducing uncertainty and risk in making decisions. Precisely because there are other phone makers with a headphone jack that do worse than the Fairphone is base enough to understand why they feel safe keeping that feature out. It doesn’t add sales and its absence doesn’t reduce them significantly either. So they know they are free to keep going even if some vocal critics will be pissed, the actual buyers couldn’t care any less.
They lose no customers by including it. They lose some by omitting it.
So it boils down to being too expensive to include? Hardly!
You evaluate prior decisions with posterior data. But you fail to take into account the counterfactuals. How do you know how much the FP4 would have sold with a jack?
Claiming that an increase in sales validates the goodness of the decision is not causal.
It is the same logic that would tell you that playing russian roulette is worthwile in case you win and get some reward. That’s backwards rationalization, fitting a narrative.
If market research universally showed that people don’t care about a jack then why do some phones still have it? Are these manufacturers going against the grain? Surely they wouldn’t leave money on the table if it worked like that.
The justification of “they do what sells units” is backwards. It would imply that no product would ever flop. But products regularly do. There is no telling in advance how it will perform, and saying otherwise is falling prey to the problem of induction, whether past observarions justify predictions.
The FP4 could have broke sales records for a multitude of reasons. How can you say which factor caused it when there is only one scenario that played out? We don’t have alternative universes to compare, where they released one with a jack, or another with some other altered specs.
I’m back to statistical significant data, and why it is important to have good data scientists in the loop. The idea is precisely to ask the questions you are asking. Would have been different if…? Then try to control for other variables in order to avoid the induction error. How do you know they didn’t do this with their data?
That’s why I mention other phone models. There are Sony phones with and without jacks. There are Asus phones with and without jacks. How did they perform compared to each other? How far away is that difference from what could be expected from randomness? How does that difference compare when the other factors are compensated for? How do they compare with other phones?
I assume they did their homework, and also want to sell more earbuds. They wouldn’t push for earbuds and wireless if headphone jacks were market drivers. It would be cheaper to install a headphone jack rather than updating the BT board? Maybe, I don’t know. But if other factors have a significant impact on sales while the jack doesn’t. Then they have their decision made for them. Market research is not about being right all the time, it is not magic, it is about reducing uncertainty and risk in making decisions. Precisely because there are other phone makers with a headphone jack that do worse than the Fairphone is base enough to understand why they feel safe keeping that feature out. It doesn’t add sales and its absence doesn’t reduce them significantly either. So they know they are free to keep going even if some vocal critics will be pissed, the actual buyers couldn’t care any less.