both my wife and I have finally gotten enough photos uploaded and are getting enough sales to begin to make tracking meaningful.  I do, however, have a question about obtaining a meaningful RPI calculation, and that is what to use for portfolio size.

We both uploaded a fair amount this month. I probably increased my port size by 30% and my wife by 50%.  If I use the portfolio size as it is now, end of the month, I actually see a fair sized drop in RPI over last month, even though my sales are much higher.  I'm wondering if a more meaningful approach would be to use an average between first of month and end of month portfolio size for the calculation.

What do you think?

I know I only seeing this discrepancy because I don't have that many files on-line (about 150 or so), but even with mid-sized portfolios a busy month of uploading could skew the RPI results, right?

Personal opinion, and I'm sure someone will disagree and have a better idea, but here's what I'd do.

Don't get all caught up in the details, it isn't a financial report for a Fortune 500 company. It's just something for yourself to give and indication of how things are going. If you wanted to be accurate you would have to track how many photos for each day of the month, because it's changing. With that every site will approve different photos on different days. You would go insane or have some complicated database with all kinds of complicated variables to account for daily changes in your photo count across all the sites.

Easy way. Count your photos at the end of the month, see how much you made that month. Kind of like gas mileage on the car. Divide and there you have it.  

As your portfolio grows, the changes in total photos will be less significant, and your averages will start to smooth out.

I started out tracking which photos were accepted on which site. Spreadsheet with thumbnails and "X's" for rejected in red, Green fill for accepted. Yellow for submitted (so I wouldn't send duplicates by accident). With some sites taking one image and rejecting others, and another site taking ones that were rejected, but rejecting some that were accepted at another site, it was pretty funny looking. Very soon, I was just looking for was a trend, not to track every image. Shortly after that, I just uploaded and gave up trying to track each image.

Lets say I had nine Microstock sites. I'd create a batch of photos, then make nine folders on the computer, one for each agency. One by one I'd upload to that agency from that folder. I might add that some sites require different sizes, so all photos went to all the sites that took the smaller size, then the mid-size and larger went to other sites, and the few that required the largest minimum photo size, got the larger ones.

If I had 50 photos, some got 50, some got 30 something, some got 10. There goes the whole plan to track RPI by image, because they are all over the place, and never the same photos or same number of photos.

Do I count from 50, 30 or ten? I don't know?

That leaves us with RPI by agency, based on total photos at the end of each month, which is the simplest reasonable solution. In my opinion.

I'd rather spend time on something else rather than complex data tracking. Maybe taking pictures or editing.  

Thanks for all of your input.

I, too, believe one can get caught us with too much detail.  However, one can also over simplify.  What I am thinking of doing, is carrying forward to closing number of images online for each agency to the beginning number of images for the next month.  Then, at the end of the month, I enter in my closing number of images, and use the average of these two numbers to calculate RPI.  Using excel, it would mean no additional entries, just complitact the RPI formula slightly.  Seems to me this would be the best all things considered, and once I get more and more images online the error would continue to decrease.

1. Monthly income  / total no. of images - does not make any sense cause it will keep going down when you add more photos

Why? I think it will go down if you _do not_. This number is more or less reliable if "average" MSG member keep in pace with other stock submitters. If we submit less - they get more, and opposite.

I think astrocady's variation is a good way to go (if you increase in a port size < 10%, even this is too complex, use either end or beginning of month data, does not make almost any difference)