| Tweet | Buffer |
Much has been written about how to maximize follower growth, interaction and click-through rates on Twitter; what days of the week to tweet and what time of day, as well as the structure and content of tweets themselves. However, few studies exists on the optimum volume and frequency of tweets. Assuming that best practices and etiquette are being followed, what impact can suddenly doubling the volume of tweets have on a business/organization account?
The Data
In September 2012, we increased the amount of scheduled tweets from 6 per day (seven days a week) to 12.
Using data gathered from Crowdbooster and Google Analytics, here is a comparison between a four-month period at 6 scheduled tweets per day and a four-month period at 12 scheduled tweets per day (data does not include replies and retweets).
Original Cadence (May-Aug):
- 7am (Industry article)
- 9am (Slingshot SEO blog post)
- 11am (Industry article)
- 1pm (Slingshot SEO landing page/offer)
- 3pm (Industry article)
- 5pm (Industry article)
Content: 33% Slingshot SEO, 66% industry
via Crowdbooster
New Cadence (Sept-Dec):
- 7am (Industry article)
- 8am (Industry article)
- 9am (Slingshot SEO blog post)
- 10am (Industry article)
- 11am (Slingshot SEO landing page/offer)
- 12pm (Industry article)
- 1pm (Slingshot SEO blog post)
- 2pm (Industry article)
- 3pm (Slingshot SEO landing page/offer)
- 4pm (Industry article)
- 5pm (Industry article)
- 6pm (Slingshot SEO blog post)
- 7pm (Industry article)
Content: 42% Slingshot SEO, 58% industry
via Crowdbooster
The Impact
Doubling the frequency of scheduled tweets produced some interesting results, the least surprising of which was that overall impressions more than doubled from about 6 million to 14 million. It should be noted that an increase in overall impressions can be in part attributed to the increased amount of followers enjoyed during the latter half of the study.
Speaking of followers, the Slingshot SEO twitter account only gained a couple hundred more followers during the increased tweeting period. Tweeting twice as often seemed to have only a small effect on follower growth. Could the increased volume of tweets have led to more unfollows, thereby counteracting the potential gains in follower growth?
In terms of interaction, tweeting twice as often produced about 38% more retweets, about 14% less mentions and about 70% more favorites.
Referral traffic was impacted significantly:
via Google Analytics
Tweeting twice as often produced a 124.60% increase in referral traffic, no doubt as a result of the increased volume of tweets containing links to Slingshot SEO blog posts and landing pages.
As a baseline, here is a look at overall website traffic during the two date ranges being compared:
via Google Analytics
You can see that the increase in referral traffic from Twitter represents a significant percentage (nearly one-third) of the increase in overall traffic to the website from September to December.
Takeaways
For B2C companies focused on conversions to leads, an increased volume of tweets may be appropriate – especially for those who rely on shared thought leadership to build brand equity. But be warned: an increased volume of bad content won’t get you anywhere, and may actually prove harmful.
If you do decide to tweet as often as the Slingshot SEO account, be sure to employ the following best practices.
- Avoid more than one link per hour. Dan Zarrella has reported that as the speed of link tweeting increased, the CTR decreased.
- Follow the “rule of quarters” – 25% your content, 25% interaction, 50% others’ content
- Share only what your community will find useful
- Attribute content to their authors (example: “article title” + “link” + “via @author”)
- Don’t sleep on nights and weekends
- Bulk schedule tweets in advance to save time
- Analyze and adjust
tl;dr
Tweeting twice as often had little effect on follower growth and slightly increased interaction while more than doubling referral traffic. Tweet as often only if you maintain a high standard of content quality and usefulness.
Having trouble organizing and scheduling your social media updates? Download my Editorial Calendar Template!
Steven Shattuck
@stevenshattuck
Latest posts by Steven Shattuck (see all)
- “Past” Contributorship Has No Impact on Authorship Annotations or Rankings - April 23, 2013
- Want To Increase Conversions? Focus On Referral Traffic - April 18, 2013
- Google+ Adds Current/Past Option to “Contributor To” Field - April 17, 2013
Great stuff, Steven!
Very transparent! The takeaways at the end of the article drive home strategy is necessary to succeed in the social world.
I have unfollowed users on Twitter based on increase content without increased value.
We try to be transparent around here. More doesn’t always equal better on social media.
Great test! I think it might be interesting to look at this on a per Tweet basis.
Because while traffic was up 124% the number of Slingshot Tweets was actually up 150% between those two periods, right?
Here’s my quick math. 120 days with 2 Slingshot Tweets per day = 240. 120 days with 5 Slingshot Tweets per day = 600. So (600-240)/240 = 1.50 or 150% increase in Slingshot Tweets which produced 124% increase in traffic.
So overall traffic increases while each Tweet was potentially less effective.
2,533/240 = 10.6 Visits per Tweet
5,689/600 = 9.5 Visits per Tweet
Or a 10.4% decrease in Visits per Tweet.
Hey AJ! I hadn’t thought of looking at it like that. You’re correct, Slingshot-centric tweets went up from 2 per day to 5 per day (150%). Looks like less really is more in this case!
Unfortunately, I don’t have solid conversions data from the May-Aug period, or else I could look to see if the 10.4% decrease in visits was worth an increase in conversions (assuming an increase occurred).
I love the transparency of your data set here, I’d be interested in seeing your assisted conversion data to see what the impact on conversions was from the shift in social strategy?
Thanks, Jeremy! Unfortunately, my conversions data from Feb-Sep is a little suspect. I can say that the conversion rate from Twitter between Oct-Dec is right around 1.9%, while month-to-date in January is 2.5%.
Pingback: The Best Expert Advice On Content Marketing This Week Jan 30th