4 ms·
It's a confirmation bias problem. Customers who fill out surveys and give direct feedback back to vendors are frequently lower-usage enthusiasts that aren't rep
by jyrkesh 7y ago
It's a confirmation bias problem. Customers who fill out surveys and give direct feedback back to vendors are frequently lower-usage enthusiasts that aren't representative of your overall userbase.
I work on a product, and we include some telemetry. I'm also a strong privacy advocate, and I believe I've done my best within the corporate realm to ensure that the data we're collecting is extremely scoped AND useful for decision making and prioritization. In my experience, there aren't that many of me, but I implore folks to realize that as PMs and engineers, we absolutely do have a say in making sure that blanket data exfiltration and aggregation doesn't happen in our products.
Communicating with your customers proactively about what you're collecting and why is important too. And not buried in some privacy policy legalese: publish a blog post, explain what you're gathering, give examples of how it's driven decision-making for you in the past, and what you're hoping to learn in the future. It goes a long way.
- fl0wenol 7y agoAs a user, even if you're up about these things and why they're important for product development, my answer is still "No." Because I will only say "Yes" if I'm being paid or otherwise compensated specifically for that input. I think it's important companies pay for usage testing so that they value that information and are more likely to hold it closely since it represents an investment and is perceived as competitive advantage.
- Piskvorrr 7y agoThat's why a sane data collection is aggreggate, upfront about the data it collects, avoidable, and most of all, explicitly opt-in. Alas, this requires eternal vigilance, as the pushback is neverending.