you could easily go with 10min blocks if you are a bit trained (wich I think you are right?)
I sorta doubt you could pull off 30 10s all-out sprints in ten minutes, but that's why you suggest going longer.
These are difficult to get right without a pm.
I'd start with 20on 40off for maybe 6 minutes the first time, which could very soon be 8 minutes and then see how 30/30 and 40/20 go, but don't let the quality suffer, that's your job.
The tricky part is that "on" doesn't mean all out, and "off" is not barely pedaling, ideally.
I know people that do them for up to an hour at a time, often motorpaced. They're not supposed to be maximal all out sprints, those are Tabatas you're thinking of (20s all out, 10 off).
30/30s or 40/20s done more in the saddle and gradual in your vo2 or L6 zone would be more specific for your sportive I'd imagine since it would simulate powering over rollers. A normal microburst, to me, is like 15s@200% including an our of the saddle effort, 15s at like 50% of FTP for 10min. sets up to an hour is more for punchy races, criteriums, cross, or sometimes MTB racing.
Coggan did a QA of different microbursts and found that the devil is in the details. It doesn't matter what you do as much as how you do them. 30/30s done aggressively and out of the saddle were almost purely neuromuscular even though NP and power targets were well below L7. Rolling into them gradually, in the saddle and doing them steady state was more zone accurate. His most important point was that just because AP or NP is equal to a certain power level do not think that it counts as work at that level. For example, microburst workouts often yield an NP around FTP, but its not FTP work or SST work.
With that in mind here is what I have done and why:
30/30: On is 5min. MMP, off is 75% of FTP. 10min sets for 3 total sets. These are a variable form of L5 work. If you look at a blood lactate analysis, gas analysis, and HR analysis, the first few reps bring you closer and closer to max oxygen debt. About halfway through you basically stay there as the recovery is too short to really recover. Sure, metrics drop slightly, but they go right back up immediately. Each set yields around 4 minutes of total vo2 time, but its in a more specific way since it mimics rollers, attacks, etc. After doing these for a while I will bump the power up about 20w for the on or the off depending on what it is I think needs work- the size of the surge or recovering from it.
40/20: Same thing, but harder. These tend to still be done at the same power level, but I can bump them up 20w or so to make them more L6 oriented.
1min/20s. Lower on portion, but I treat it like I'd try to bridge up to a move or close a gap.
15/15- Out of the saddle for the first 5-8s trying to hit 300% of FTP at least, then drive in the saddle for 150%+ of FTP. With an FTP of 330 I aim for a 1000w initial kick, then drive at 500w. Off is basically 50% of FTP. These are almost entirely neuromuscular and are something I use for punchier races and trying not to suck at criteriums.
5s/10s- All out for the 5s. Designed to tax the CP system. Almost entirely L7. Very criterium/cross specific.
The length of sets varies with fitness and power profile. I can't do 15/15s longer than maybe 15min. total, but my teammate does them for an hour at the end of an endurance ride to prep for cross. His on portions are a bit lower since his event requires more consistent, lower output bursts.
Don't take me too seriously. The only person that doesn't hate