I've seen a few people recommending calorie counting here but haven't seen anyone mention Macrofactor, which seems weird considering how often I see people recommending Kagi. I draw the parallel because, while it's a paid product, I find it significantly better than the competition.
I started using it at the start of the year and have had steady progress. Foremost, it is extremely snappy and easy to log food. The database is fairly expansive without having poor quality user submissions. The real win of the app is the feedback loop. Rather than estimating calories expended using formulas meant to be accurate across a population (but not necessarily accurate to each person), it uses your calorie intake data and your weight data to derive your expenditure.
This, to me, helps reduce the stress of tracking significantly. Reason being, if you habitually do not track something like small bites during cooking or condiments, the calculation will take it into account and reduce your calorie target accordingly.
It also doesn't take into account data from activity trackers. Instead, your exercise is essentially smoothed over the following weeks. It helps psychologically to break from "I exercised so I get a treat" mentality, where you 1: immediately eat back whatever you've burned (or more) and 2: are telling yourself a reward for good behavior is calorie-dense food.
The website has a lot of data driven articles.
It also has a bunch of neat graphs. Anyway. Would recommend it. Obviously there's a LOT of different ways to lose weight, but for me it starts with understanding what I'm putting in my body. Can't outrun a bad diet.
Previously worked for Amazon data labeling. The priority was largely having everything done in-house due to privacy concerns. It's a lot easier to act on privacy leaks coming from within. That said, the long term strategy is using crowdsourced labeling for anything not having to do with customers or customer data. So looks like you're both right :)