Methodology:
I analyzed the portfolio and download data of the Top ~2000 contributors (based on Total Downloads) and plotted:
- Histogram of Number of Photographers versus Portfolio Size
- Average Annual Downloads per Photographer for Each Portfolio Bucket
Chart 1: Histogram of Photographer Count vs. Portfolio Size
- 82% of the photographers in the Top 2,000 have portfolios of 2,500 images, or less.
- The last category is larger than the others because there are so few contributors with portfolios over 10,000 images.
Chart 2: Average Annual Downloads by Portfolio Size
Conclusions:
- I was very surprised to find that less than 20% of photographers in this dataset had portfolios larger than 2,500 images.
- I found the download numbers made more sense since all things being equal, more images should lead to higher earnings.
- I’m curious about average downloads per image for each of the above bands — I’ll post on this subject later this week.
Methodology:
I took the same dataset I used for last week’s post on Rank vs. Tenure, I plotted the following:
- Contributor Rank vs. Portfolio Size
- Average Portfolio Size by Contributor Rank
My goal here was to study the relationship of portfolio size & performance.
Chart 1: Top 2000 Contributors: Rank vs. Portfolio Size
- The average for the entire set was 1,580 files, and the median was 1,067. The spread between these two numbers means that most people have less the mean, but those with very large portfolios are dragging up the average.
- The smallest portfolio in the Top 100 contained 233 images.
- The largest portfolio in the Bottom 100 contained 4,350 images.
Chart 2: Top 500 Contributors – Average Portfolio Size by Rank
- There is a clear break between the top 100 contributors average portfolio sizes and the rest.
- The middle of the range (101-400) are relatively uniform.
Conclusions:
- More images online is correlated with higher rank, but it’s a weaker relationship than I expected — quality plays a major role too.
- Productivity counts — The Top 100 Contributors add 724 files per year, over double that of contributors ranked 400-500, who add 333 files per year.
NB: We can only talk about correlation. You can’t say that if you add 700+ images per year, you’ll be in the Top 100.
Methodology:
I got public data from iStockcharts for the Top 2000 contributors, ranked by Total Downloads. I then plotted the following:
- Rank vs. Years contributing of the Top 2000
- Average & Min Years contributing for the Top 1000
- Histogram of Years Contributing for the Top 200
My goal was to examine tenure vs. performance to see if the number of years you had been at iStock made a difference.
Chart 1: Top 2000 Contributors: Rank vs. Years Contributing
- The average for the entire set was 5.4 years.
- The newest member of the Top 200 had been contributing for less than 3 years.
- The entire set was in the range of 1.5 – 9.0 years.
Chart 2: Top 1000 Contributors — Average & Min Years Contributing
- The most notable thing here is how little variation there is — for the most part, there is no difference in experience between those ranked 1-200 and those ranked 800-1000. I think this is amazing.
Chart 3: Top 200 Contributors by Years Contributing
- 48.5% (97/200) contributors had between 4-6 years of experience.
Conclusions
- I didn’t expect to find this, but there is very little correlation between your iStock performance and your years of experience at iStock.
- The R-squared coefficient is 0.03. It’s range is between 0 (no correlation) and 1 (perfectly correlated).
- Chart 1′s distribution is very uniform for each ranking band.
- No difference in avg & min years contributing for the Top 1000.
- It takes about 2 years to build up a portfolio large enough to break into the Top 2000 at iStock.
- After that, quality and consistency in uploading is what counts.
Methodology:
I conducted searches for the keyword ‘vacation’ and each continent, on iStock. I summed the downloads for the first 100 Best Match results for each one. If you’re interested, you can check out the searches yourself: Australia, Africa, Europe, Asia, Antarctica, South America, and North America.
Results:
Points to Note:
- Australia accounts for 30% of the total DLs for the seven continents.
- I was very surprised by the low DLs for North America. It may be because people don’t search for ‘North America’ but rather search for ‘USA’ or ‘Canada’ (more to come on this one.)
- Editorial is live but sparse — 13 out of 700 items were editorial images (all in North America, South America, & Antarctica.)
Methodology:
I ran the searches below on iStock and summed the downloads of the top 200 Best Match results for each one:
Results:
Points to Note:
- 45% of the Total Downloads in this set of searches were generated by Valentine’s Day AND Roses.
- Valentine’s Day AND Roses generated more cumulative downloads than just Valentine’s Day.
- Jewelry and Chocolate are tiny niches compared to roses.
To see how images were slotted in iStock’s Best Match search results, I divided 2,100 results into two groups:
- New Images: Images with 0-1,000 DLs
- Proven Images: Images with 1000-10,000+ DLs
New vs. Proven Images by Search Position
The chart below shows the number of new vs. proven images for the first 25 positions.
Points to Note:
- For popular searches (business, medical, lifestyle), the top positions are reserved for images with staying power.
- iStock is dedicating half of the first page to New images (assuming 10 images per row on a typical screen.)
- 80.9% of the first 5 search results were Proven images even though they only made up 40.8% of the total results.
- Digging deeper, there are no images with less than 100 DLs in the first five search positions.
To see how much iStock favored exclusive content in searches, I analyzed 2,100 results which I got by recording the Top 100 Best Match results for three searches — ‘business’, ‘medical’, and ‘lifestyle’ — for 7 days.
Percentage of Exclusive vs. Non-Exclusive Images by Rank in Search Results
The chart plots the % of results that are exclusive vs. non-exclusive for each search position. For example, an image in position 1, is twice as likely to be exclusive vs. non-exclusive (66.6% vs. 33.3%).
Out of 2,100 total search results; 1,687 exclusive vs. 413 non-exclusive. This is a ratio of 4.1:1. This surprised me because I expected a ratio closer to 2:1 in favor of exclusives based on anecdotal information I had heard.
Conclusions:
- It’s great to be exclusive — you’re a 4 to 1 favorite for some of the highest earning searches in best match.
- You may want to consider exclusivity if iStock is at least 25% of your earnings. (You’ll have to make your own decision of course.)
Overview
Earlier this week I posted an analysis of Lifestyle, Business, and Medical searches by Collection. As a follow-up, I analyzed each contributor’s share of search results by counting the number of images they had in the Top 100 best match results for: ‘lifestyle’, ‘medical’, and ‘business’.
Top 25 Contributors by Share of Search Results
Observations:
- 129 contributors accounted for 300 results; average share: 0.7% (2.3 images) per contributor.
- 10% (31 images) of the results are Yuri’s. This is over 14x the average for this set of three searches.
- 42.3% of the results belong to the Top 10 contributors.
- Small improvements can make a big difference — moving from 2 to 3 images in best match matters.
Questions
- What do you think of this analysis?
- Are there other searches that interest you?
Sign up for free today and start using LookStat to get insights that will make you more money.
Methodology
I searched on istock and recorded the top 100 Best match results for the following searches:
I analyzed the Top 100 images returned for each search. This post analyzes the results by collection. Future posts will look at contributors, downloads, and search placement.
Search Results by Collection
- 76% of lifestyle search results are either Agency (50%) or Vetta (26%). This is double what you see in medical and business searches.
- 29% of business search results are non-exclusive images. Of these 29 images, 20 belong to Yuri Arcurs. (The average is 2.1 images per contributor).
- 8.3% of all results were video files and only 1 search result was an audio file.
- 40% of all results were either Agency or Vetta files. However, they only account for 4.8% of downloads. This is most likely because the files in these collections are new.
Conclusions:
- Agency & Vetta results skew towards lifestyle. Keep this in mind when shooting & submitting.
- Audio & Video are niche business for iStock (less than 9% of best match results). Invest accordingly.
- It’s great to be exclusive. 80% of search results are from exclusives, much higher than I expected.
Which searches are you interested in analyzing? Let me know in the comments.
Sign up for free today and start using LookStat to get insights that will make you more money.
Do you know how the ethnicity of your models impacts your microstock earnings?
You can find this out using LookStat collections. I’ll walk through how to do this for the following groups: African American, Asian, Caucasian, and Hispanic.
1. Create a Collection for the first group — ‘african_american’ in this case:
2. Enter the following keyword sets and add all results to the collection:
- HAVING: african%
- NOT HAVING: hispanic%, caucasian%, asian%, %indian%, divers%, mixed%
(We’re use both fields so we can isolate each ethnic group)
4. Repeat steps 1-3 above for each ethnicity using the keyword sets below:
- African American:
- HAVING: african%
- NOT HAVING: hispanic%, caucasian%, asian%, %indian%, divers%, mixed%
- Asian
- HAVING: asian%
- NOT HAVING: hispanic%, caucasian%, african% %indian%, divers%, mixed%
- Caucasian
- HAVING: caucasian%
- NOT HAVING: hispanic%, african%, asian%, %indian%, divers%, mixed%
- Hispanic
- HAVING: hispanic%
- NOT HAVING: african%, caucasian%, asian%, %indian%, divers%, mixed%
5. Analyze the Results:
- LookStat will update your collections within an hour.
- View each collection and compare their relative performance.
- Combine other keywords with ethnicity to analyze impact by category – for example: ‘african%, business%’ vs. ‘caucasian%, business%’.
Sign up for free today and start using LookStat to make decisions about which models to use in your next shoot.