Virality on Pinterest: Strategies of the Most Successful Retail Brands
We selected 22 Pinterest accounts for this report according the following criteria:
- Ecommerce-enabled sites (excluding the Pinterest accounts of magazines, blogs, and individual users)
- Official retailer brand Pinterest account
- At least 10,000 account followers
- First account pin on Pinterest at least 60 days ago
- At least 100 pins in the last 60 days
We refer to these retail brand Pinterest accounts as “brands”. Using the Pinsights database of millions of Pinterest user actions, we analyzed the performance of these brands. All data is for sixty days ending June 19, 2012 unless otherwise specified.
A caveat: we are not saying that these brands attracted more than 10,000 followers because they achieved the metrics included in our study. However it is useful and valuable to understand how retail brands with large followings manage their Pinterest accounts.
Summary of Findings
- Pins from sources other than a brand’s website are 50% more “viral”(1) than images from the brand’s website. Repins of images from other websites are twice as viral as pins and repins to a brand’s own website.
- Brands achieve the highest virality by consistently pinning engaging content, not through “hits”. Only 3% of pins receive more than 10 repins per 1,000 followers. Unlike many internet phenomena in which 10% of items generate 90% of results, the top 10% of pins generate less than 40% of repins.
- Successful brands employ a variety of pinning strategies — “Mavens”, “Connectors”, and “Showcasers”(2) — but Mavens have 30% to 60% higher levels of follower engagement.
Brands in this study have an average of about 30,000 account followers. Account followers range from about 11,000 (Victoria’s Secret) to almost 100,000 (Etsy).
We examined “virality”, the number of repins per 1,000 followers(1). In the table below, we broke down pins into two categories. From left to right are “Pins” (i.e., original pins made by the brand) vs. “Repins” (images pinned by another Pinterest user that is repinned by the brand). On the vertical axis we have “Own Website” (the company website of the brand) and “Other Websites”. Looking at the data, we found that original pins from other sources are 50% more viral than pins from a brand’s website. We also found that, at a virality rate of 3.7, repinned images from other websites are twice as viral as pins from a brand’s own website. Likewise, pins to own websites have the lowest virality.
Looking pins grouped by virality “bin”(3), we found that 40% of brand pins have less than 1 repin per 1,000 followers. About three-quarters of pins have less than 3 repins per 1,000 followers. Only 3% of pins have more than 10 repins per 1,000 followers.
Pin Virality Distribution
Repins per 1,000 Followers
Often when looking at distributions of internet data — product sales, search query frequencies, etc — we see highly skewed distributions in which 10% of the items can generate 90% of the results. That turned out not to be true in this case. In the chart below, we found that the top 1% of pins (ranked by virality) generate only 7% of repins, while the top 10% of pins generate less than 40% of repins. So for these brands, pinning is not really a “hits” business.
Percent of Repins Generated by Top Percent of Pins
Taking a slightly different cut of the data, we grouped brands by high virality vs. average to low virality for their accounts overall. We found that high virality brands have more pins with 2 to 10 repins per 1,000 followers. We find that brands achieve the highest overall virality by pinning consistently engaging content, not through “hits”.
Differences in Brand Virality
We found that brands employ a variety of pinning strategies. We grouped the brands along two dimensions: the proportion of original pins vs. repins and the proportion of source content from their company website vs. other source websites. Borrowing heavily from Malcolm Gladwell (2), we assigned each brand to one of the following groups:
- Mavens: brands that find and pin images from sources other than their company websites (e.g., Kate Spade)
- Connectors: brands that mostly repin pins from sources other than their company websites (e.g., West Elm)
- Showcasers: brands that mostly pin original pins from their own websites (e.g., Williams-Sonoma)
We then arrayed the brands on a 2×2 grid to illustrate their pinning strategies as shown in the chart below.
We then examined the average brand virality for each of the strategies and found that Mavens have the highest average virality:
- 30% higher virality than Connectors
- 60% higher virality than Showcasers
In the course of doing this study we conducted a number of analyses not included in this report including pins by category, board title length, pin description length, use of hashtags and @ signs, and followers vs. following ratios.
The full 24-page report is available as a free download. Click the button below to get a copy.
Download the Report
(2) We’re borrowing heavily from Malcolm Gladwell’s The Tipping Point for these strategy names
(3) Note: Virality bins on x-axis include fractions up to next integer (eg, “0” bin includes pins with virality from 0 to 0.99)