Twitter is to Testing Social, What PPC is to SEO

Posted in Blogging, Social Media, Twitter

tweets

The social Internet is about conversations that propagate virulently and index at various speeds.  Twitter, the fastest feed community of all,  can be easily leveraged to “proof” marketing concepts for slower social channels the same way SEOs use PPC (pay per click) to prove organic landing pages.

Read on for a tactical starting point to use Twitter’s flashpoint sharing environment to test your linkbait, promotional messages, calls to action, friend-making and branding.

To gain perspective, set a Google Alert a couple of days early in an emerging dialog, and marvel at the pure genesis of a few blogs spawning organic global dialog. Monitor the topic’s Twitter stream at the same time for a jolt. The speed of conversation is radically faster.

What’s interesting about the blogosphere and some other social channels, is the relatively slow speed at which content propagates.  Though it only takes a few minutes for posts from authority publications to be be indexed by Google,  that’s a long time…too long for effective multivariate message-testing.

Over minutes and hours people share, react, engage, bookmark, think, write more blog posts and the whole thing moves along over days and weeks. Later sympathetic viral chatter can flare about, banging around various channels.  Blog buzz, though much faster than traditional media channels,  is not a real-time machine.

While it’s possible to test individual blog posts and make changes as viral waves wash over time, anyone who’s ever tried changing-out blog post headlines, layout and copy in reaction to to well…non-reaction, knows that blogs and other web pages are clunky–hardly platforms for real-time social media marketing tests.

PPC Testing to Prove SEO Funnels
Like slower social channels, SEO is also difficult to run multivariate tests on because pages get indexed relatively slowly. That’s why it’s great, when possible, to test organic landing pages with PPC. Overcome the relatively slow GoogleBot organic process with PPC testing, in order to gain a large enough sampling quickly enough to preview organic conversion.

PPC artists are famous for guerrilla ad-message and landing page testing. They grind every shred of possible conversion out of each paid click by quickly driving users past a grid of different landing pages. Volume and speed matter for testing, because marketers need a large enough user-sample, driven from specific keywords in order to gain enough insight to sharpen conversion.

Why Use PPC For SEO Multivariate Testing?

sesnypost3

  • PPC moves much faster, is precise and controllable.
  • Quickly funnels, conversion, design, etc…
  • Validates marketer’s instincts or proves pages, action calls or messages that do and don’t work
  • Quick method to get the user-volume needed to proof concepts and conversion

How to Test Social Using Twitter

sesnypost9

That fact that a channel is “social,” does not mean it necessarily moves faster. StumbleUpon, Delicious, Digg, Facebook, Twitter, LInkedIn, Reddit, Flickr, YouTube all have their unique viral rhythms, all slower than Twitter. We’d rather plan for slower channels by testing the populace’s reaction in Twitter’s petri dish and refine the approach prior to broader sharing.

With a little technique, some first-degree-of-seperation-tweeps and a ton of respect for community, feedback to test headlines and messages is immediate.  Twitter allows social media marketers to parade test messages past a larger sampling of users, very quickly. The speed of conversation provides any number of opportunities for savvy social media marketers to test.

Viral proliferation essentially occurs when others traffic your message because it’s somehow provocative. Friends of friends of friends rebroadcast to others and recommend content where they have authority. It’s the power of one to exponential results.  So what do you want to know when testing social?

  • Does anybody care? Is the subject matter or presentation interesting to others topically engaged– or only me?
  • Will my friends, their friends and folks I don’t even know rebroadcast? We’ve all recommended content that went thud. I want to know if the content or message we offer is worthy.
  • Is there are combination of words and/or concept, more likely to catch attention?  I’d much rather find that out prior to making a large investment of time sharing in different places.
  • Does my avatar’s circle of friends have enough authority to zealously get a message they care about, out to a broader audience?
  • What do various interest segments respond to? How can we tap into the interest group to make friends and influence others by inspiring with brilliant perspective, wit, insight, etc…
  • Are the headline and description I wrap my content recommendations up in, likely to do the trick? Sure the content has to be great, but what packaging works when I recommend.

Recently aimClear Blog featured a post about creating a reputation monitoring dashboard which got a ton of attention, links, rankings,  buzz and resulted in a cool new client. The inception of the buzz was in a series of retweets, the headline concept of which we had tested days before, on another writer’s blog posts about reputation monitoring. Using one of our avatars, we did multivariate headline testing to determine what types of messages inspired rebroadcast.

Once we understood that bookmarks like “Don’t get caught with your pants around your ankles” or “bend over, the TwitterJacks are coming” worked we settled on the “wrapper” for release to slower channels. We already knew it would work, like dry tinder for lighting a fire, because we had proven the idea. It was easy to come back a week later, armed with a well thought out post, match to fuse.

Be Creative In Testing
Micro-blogging breakout Twitter is an incredible incubator for testing social messages.  There are many methods to probe and test ideas. Using tools like Tweetdeck, it’s easy dial into conversations surrounding just about any keyword combination, second by second, real-time, as opposed to hours and days. Use your participation in focused conversations to refine  instincts surrounding what’s worthy viral material.

  • Retweet others recommendations, testing different wrappers
  • Reply using the @ sign to folks you just met, add perspective to the conversation and measure how the approach you take affects reactions in the topical community you’re participating in.
  • Make lists of Tweets that are massively rebroadcast and analyze what works amongst various interest groups.
  • Be willing to throw ideas, even those you’re attached to, away once empirical evidence proves it won’t work.

Twitter is one of the ultimate social research tools, providing incredible insight as to a segmented communities likelihood to rebroadcast. Much the way search engine optimization jockeys look to PPC to achieve necessary user volume to quickly test, leverage Twitter’s hair-trigger rebroadcast paradigm to vet social media marketing message prior to broader sharing, blogging and bookmarking.

  • Britta

    excellent article and useful suggestions to using twitter for testing purposes. thanks!

  • jeremy

    Great article. My guess is you will find this blog post fairs well in the twitter “petri dish.” I think it’s ready for the “other” social media.

  • Michael McDermott

    While changing the headline in a blog and leaving it to Google Search to find may not work effectively, I have found that changing the blog title and tweeting the messaging out allows similar multivariate testing results. Some titles garner zero RT’s, others become quite viral.
    Leading with the headline (dry tinder) I then publish out for outpost and bookmark indexing. Stand back and watch it go crazy!