I preface this response entirely on that it depends completely on your source water profile. My experience is mainly coming from treated groundwater via municipal water company. Average 350-420 TDS coming into my house.
Also, sorry for the long response.
I am starting to believe that most of us hobbyists are wasting a canister on cation. As a general rule, industry retailers tell everyone to go cation, anion, mixed bed for the cleanest output. Reason being that the cation exchange can prevent precipitation in the exchange of the anion process. So it makes sense to have it first... maybe. Except, folks consistently run single mixed bed stages all the time, with no known precipitation that they are aware of. And the majority of that single stage is wasted and they would never know.
For many of us, we have similar source water chemistry and profiles, especially the groundwater characteristics. The amount of time it takes for the cation to get used up, is months. Mostly because of how effective cation is to removing the positively charged particles. It doesn't require much cation resin to be effective.
IMO, that is a waste of a chamber. I've got some experiments that I will be trying soon with different ratios for my 1st chamber.
Here is my thought process.
Stage 1: An inch or two of cation and the rest MB
I was going to weigh the total volume of resin to fill a 10" canister and then start with only 10% cation resin.
Stage 2 - Anion
Stage 3 - MB
I believe that if that first stage is changed out as soon as you see 2-4 TDS into the Anion chamber, you could see significant savings in your anion usage versus only running a cation in stage 1. As soon as you see the color changing in the anion, refresh chamber 1. That would leave you with a 3rd stage of MB for final polishing and truly a backup for preventing TDS creep in your reservoir.
Another idea I had was to go
MB - Anion - MB
Mixed bed in chamber one. That is going to get burned up very quickly. Basically at the same rate at which your anion chamber would typically go, if you have it after the cation as is typically recommended. But that mixed bed will capture the initial positive ions no matter what. Effectively acting as a cation chamber anyway.
That initial chamber will likely produce 100*ish gallons (maybe more or less) of clean 0 TDS water before entering the anion stage, allowing the anion to last longer than normal. Especially, if you are consistently changing out your 1st stage before the tds gets too high. By the time the water hits stage 3, there is little to no particles to use up that mixed bed. Again, reducing the chances of TDS creep in your reservoir. You're biggest resin consumption will be mixed bed in both these experiments. It's possible to see a significant savings in resin cost with the understanding that your stage 1 mixed bed will get consumed quickly. But it is still less expensive than anion by itself. You could even go with stage 1 and 2 mixed bed and save the anion as the final polishing stage, and deplete it at a far slower rate, albeit you are consistently changing your resin.
A little side note to past experiments I have tried.
I have run a mixed bed cartridges until full visible exhaustion and got 0- 2 TDS for at least another 50*ish gallons. I feel that the resin does NOT need to be changed out when it's an inch from the top as we are instructed to do so. IF you cannot monitor it closely, then the advise to change it before complete exhaustion is valid IMO.
Ok... I have had to go back and forth for the past few hours in between water changes and chores for this response.