There's really no substitute for having in-house HT capabilities if you do a lot of one-offs and custom orders. And at least understanding the concepts is crucial to being a good knifemaker, even if you don't do it yourself.
But a kiln is only the beginning... you're going to need a table for the kiln, a way to be sure it's really running as hot as it says it is, tool wrap, racks, tongs, gloves, UV glasses, plates and/or a quench tank with good oil, a dewar for LN or at least a cooler and dry ice for air-hardening steels, a Rockwell tester/calibration blocks... not to mention space for all that stuff, a dedicated 20-amp circuit if you go 120V (or preferably a 220V circuit, it's a lot more efficient), and perhaps most of all, the time to deal with it all. To be on the safe side, I'd budget at least $2K and a month's learning curve to jump right in and get going... possibly as much as twice that.
Of course you could always stick with simple steels, a basic propane or coal forge, a can of canola oil and your kitchen oven. There's nothing wrong with that, either.
If you do batches of 10-100 blades at a time
and can plan lead time to ship them out, there's no way in heck you can match the quality, price and efficiency of having Peters' or Bos do them. I've done the math several times and it just doesn't make sense for me to HT my own batches. The electricity alone would kill me, let alone the time it would take to muck around with 5 blades at a time. As Daniel mentioned, that's time I spend grinding the next batch and/or building handles and sheaths on the previous batch.
The good news is for you is, I have a barely-used 14" Paragon digital programmable kiln I'll let you have for significantly less than retail, and you'll save a couple-few hundred bucks right off the bat... without waiting weeks or months to get it delivered. Of course, you gotta come pick it up... I am
not going to crate that thing up and ship it
