MicrostockGroup

Lowest prices? N
Aggressive discounting? N
Best customer service? Y, both for customers and contributors
Niche content? (military, medical, etc) Y, with good understanding of each niche (e.g., proper understanding of copyright and property laws in architectural photography instead of rejecting everything modern just to stay on the safe side; same is true for every other niche)
Best technical quality? Y, but in a smart way: being obsessed with even lighting without shadows and noise is not looking for quality, just looking forward to have a boring collection
Most images? Y
Most aggressive sales team? Y
Tons of marketing and advertising? Y
Nothing (build it and they will come) N
Other? Y: No LCV check, as a way to create an alternative library and let buyers decide if I must choose just one

I wrote this some time ago. I'm not sure about the exclusive thing any more...but...

http://www.microstockgroup.com/general-stock-discussion/if-i-had-to-start-up-a-new-microstock-site/

don´t know how much buyers think of contributors cut but I would say a very fair agency, something like at least 50% with very good communication and impartiality, that would give higher motivation to create better content, keeping pictures fresh and strong.. too bad to achieve this some millions are needed

I think you left off one of the biggies - best search. Other than the quality and usefulness of the collection (niche or general) the ease of finding what you want is a huge deal.

I think this clearly does not mean a CV but something closer to google-like searching with very smart "more like this" "none like this" tools to help narrow the search based on image selections as well as keywords. I think it also means being really hardcore about keywording standards so you don't have the kind of spam or missing keywords that just ruins things for buyers. Case in point some of the Getty stuff on iStock: this image is of an elephant but doesn't have the keyword elephant. It does have rainforest which is wholly inappropriate.

There was a startup that did initial search results as an amazing grid of tiny squares where you could pick the ones you liked that would be used to refine the search. It was amazing that you could pretty quickly rule some things in and out from such a small thumb and refine the search in a better way. Not sure what happened to that technology.

If facebook and cameras can handle face recognition, how about stock engine search that can handle the people, how many, no people thing without keywords? That sort of thing

Another reason I've not wanted to deal with Getty other than the double-tax-whammy and the 20%. They don't let you keyword your own files, so you get cr*p like this.

Which means, yes, great search would be the Killer App for any new site. BM2 had such great potential, but they abandoned it. Maybe it wasn't as great as I think it would have been, but I don't know why. It rewarded good keywording, punished spamming and the above image would soon have sunk to obscurity. EdStock's queleas are still 48 and 50 in the best match for African Elephant (and quelea isn't in the keywords), so the current best match would seem to have no bias towards keyword relevance.

I'm starting my own next month my differentiator is regional content and personal contact.

Is the correct answer. The one thing buyers value even more than their money is their time. They don't like having it wasted by poor search results.

Spaces is a great site Jonathan. I hope you are gaining some traction.

Christian is contributing at Warmpicture with several more of us. Very happy to have a portfolio as strong as Christian's representing us.

The first 7 months of Warmpicture have been grueling. We are almost at break-even per month  , but the battle of getting us search position is one of the toughest I have ever fought online. I guess I have already put my money where my mouth is with regard to what I would like to see in a RF site:

1) Premium Collection. There are too many images on stock sites. A concentrated premium collection is what I am after.
2) Simple RF Licensing. Price the images on resolution, and include unlimited usage with the price. Since our collection is priced per the Corbis and Alamy models, we are able to completely remove the need for extended licenses.
3) No exclusivity, favoritism, or any other shenanigans. Image rankings within search are determined by performance, period.
4) Small roster of portfolios. We need to get larger, but I don't ever want ours to be huge. I like that our numbers are small enough that we can keep in close touch with each other through message boards and email, and discuss strategy. It keeps everyone on the same page. And anyone who has a good idea can contribute it without being lost in the masses.

We're going to make it. The portfolios at Warmpicture are so strong, it's just a matter of getting enough eyes on us.

Hi Dan,

 So great to hear. I find if you can keep your overhead down and offer your photographers, strong market information, support, and respect and I think you will have a great collection. This is the direction agencies need to turn. Constant feedback and direction when asked for, art directors to help plan shoots, editing that is done asap and the best key wording money can buy with a 50/50 split. Add that recipe with a strong distribution model, solid niche collection plus excellent photographers and it is only a matter of time. Best of luck in 2012.

Cheers,
Jonathan

I think most contributors prefer sites that avoid excessive 'distribution'.  Better to have a strong core site than worry about having your money siphoned by distributing to distributing distributors.

I would choose a site where all the contributors are owners at egal level, so the company cannot be sold to some big shot corporation...

I'm not Sean, but taking your question at face value:
1. No, because when Morrison's bought out Safeway, they stopped stocking some popular lines that Safeway sold. Nada the suppliers could do about that. All eggs in one basket = precarious.
2. Yes, because if you sell the same stuff everywhere, there's a supermarket price war, or price-matching deals involving "exactly the same product bought elsewhere".

Conclusion: from the contributor pov, the best model would be 'image exclusivity'.