2. The reason engineers tend to hate tactics like this is that it shifts responsibility for fuck-ups from you, the CEO, who should be managing and prioritizing features so you get the best bang for the buck, to them, the engineers, who are now on the hook for a feature that may or may not be able to be built in the time frame you promised. Things go wrong in engineering, and there are unexpected snags and slowdowns, and a well-managed business should be able to absorb them without losing customers. There are very real costs in added stress and loss of personal time for engineers - not to mention decreased trust from customers - that come from this tactic.

As the sibling comment mentioned, just say "The feature is not ready yet but we'll take your request under advisement and let you know as soon as it is." Then prioritize the features that lots of customers want. Maybe you'll lose the sale, or maybe they'll buy anyway because it was non-critical for them. Either way, you should be operating with enough of a financial cushion that one sale won't put you out of business (as PG says, "Deals fall through!"), and being able to aggregate requests lets you build a much better product instead of continually chasing after whatever the latest customer wanted.