I use Buffer and their various extensions and integration’s while I’m browsing as a simple/easy way for me to share content. I love it, it’s a great tool in my opinion – highly recommend you give it a go if you haven’t already.

A little while ago, I shot the Buffer guys an email about a suggested feature, which they were good enough to respond to me on – awesome customer service. I thought I’d put it out in the public forum, maybe if it gets a little more attention it’ll make it off their ‘to do’ list onto the ‘must do’ list.

Before using Buffer, if I manually shared something – I’d try and find the social handle/profile for the author and include them in the share so they get notified such as the below if it was a tweet:

An excellent article about X by @author

One of the benefits of Buffer is that it allows me to share the content into multiple social networks in one go – awesome! However the shared snippet that goes into Twitter/LinkedIn defaults to the shared page’s <title> tag and you end up with a basic share containing a link and text description. For example, if you were to share my article about using hashed keywords instead of (not provided) keyword, it might look like this by default:

Using Hashed Keywords Instead Of (not provided) Keyword | Convergent Media http://buff.ly/S1MzQt

While you can customise the shared content, one of my primary reasons for using Buffer is about sharing it into multiple social networks and saving me time – as such I don’t spend a lot of time editing the default shared snippet. Typically I’d remove fluff, such as “| Convergent Media” in the above or rewrite it entirely if I didn’t feel it conveyed the message clearly enough but it’s typically a light touch approach.

As a result of my laziness (I’m just being honest), using Buffer has meant that I share into Twitter and LinkedIn (great) but what I share is a little less optimal than when I was only manually sharing into Twitter.

Suggested Feature To The Rescue

My feature suggestion for the Buffer guys was to look at the HTML and social widget configuration on a page that is being shared and see if they can determine the handle/profile of the author. If they can, do something smart, if they can’t – fall back to the default behaviour.

In an ideal world, if this was possible – when I share a piece of content Buffer would customise the share for each social network such that it’d use @author notation in Twitter and include the profile in LinkedIn, facebook amd Google+ if it ever arrives. Using this mechanism, the author of the content would be notified about the share in all relevant networks that Buffer could identify him/her in that the content was shared into.

This has a couple benefits:

1. It helps the person sharing the content as the author is going to get notified of the share. Often shares that don’t include a social profile for attribution will sadly go unnoticed by the author. Having the author being aware of the share will help strike up relationships with other awesome folks – network building than you very much.
2. The author is more likely to re-share my shared content with their followers as they get to show their followers other people love their product/service. A small amount of self horn tooting is okay in my book.
3. It’s good for the person sharing the content as they get their name in front the authors followers. Just like in point one above, this exposure – especially if it is semi regular can lead to the author’s followers becoming your followers – superb.

Determining The Author’s Social Profiles

This isn’t going to work all the time, in many cases it might not be possible for Buffer to accurately determine the author’s social profiles for inclusion into the shared snippet but I think it’d be worth a shot.

Following are some ways I can think of that Buffer could determine the author:

• Check if the Twitter tweet button is on the page and if they have customised the tweet text to something like “{postname} by @author”.
• Following on from the above, they could go one step further (not sure if this would be as relevant) but many websites also use the ‘related’ attribute in the Twitter share button to suggest to people sharing the content who to follow immediately after sharing the content.
• Check if the author has an author profile on the site, in which case it is likely that they’ll have linked off to their various social media profiles .
• Check if there is an author bio associated to the article, which might also include links to their various social profiles
• If Google+ sharing becomes an option in the future, Buffer could check for links to Google+ with authorship markup attached.
• Crawl all social profiles you can find associated to the author (if none are for Twitter/LinkedIn/facebook directly as an example) and see if they list their ‘other’ social profiles within the social profiles you were able to crawl.

In the case of Twitter, there are official tweet text parsing libraries provided by Twitter in various languages and there are many open source implementations of the Twitter text parsing libraries available in different languages as well.

So there you have it, a feature upgrade for Buffer that I think would be really worth while as it helps authors and social sharers alike to a mutually beneficial gain.

Do you think it’s a good idea, should Buffer add it to their queue of work or am I mad?

Recently Barry Adams wrote an article titled Social Media is Dead; Long Live SEO in which he puts forward the case that social media is a waste of time for most businesses and they should focus on what works. It should be noted that it isn’t Barry just making up sensational headlines, those comments are supported by research conducted by Forrester in late 2012 and also by Custora in 2013.

I’m not here to dispute that fact directly but I thought was worth throwing another discussion point into the melting pot for everyone to consider and that is, while the internet is a highly measurable place which marketers and businesses alike love, it does have limitations and one of those limitations is uniquely identifying a person.

While the technology wasn’t as advanced, the ability to identify an individual user 10 years ago was simpler – people had less frequent access to the internet and from fewer computers. Fast forward ten years and we are living in a multi-screen world, where an individual person might switch between phone, tablet, laptop, desktop, TV and more across the course of the day, all the while continuing what the person considers to be a single, unified experience.

All the different devices used by consumers today complicate the problem of uniquely identifying a person, as the unique identification is generally done through the use of browser cookies. That means that the same user viewing a website on their phone, tablet, laptop, desktop and TV are normally counted as separate users within web statistics software such as Google Analytics.

Barry replied to comment on his article where he mentioned that he has seen many different multi-channel attribution reports from Google Analytics that never register social media traffic sources in any significant way, even when looking at the assisted conversion report.

Google Analytics Multi-channel Funnel Assisted Conversions Report

You’re mother would have told you never to believe everything you see on TV, read in the newspapers or view in Google Analytics – okay, I’ll concede the last point. What many don’t realise when seeing a headline from companies like Forrester or a neat table like the one above, is that it is increasingly difficult to measure the impact of different traffic sources end to end due to the browser cookie issue I briefly mentioned above.

The Difficultly In Measuring Social Media Impact

For the sake of discussion, we’ll focus on both facebook and Twitter as they are the most widely used social networks. It may or may not come as a surprise, but both of these social networks report over 50% of their usage is via mobile devices.

Imagine that you’re Forrester and you’re trying to compile research about the impact of social media on businesses. When over 50% of the usage of the two biggest social networks in the world are powered by mobile and mobile conversion rates are well below their desktop browser counterparts – that alone provides a reason why it’s hard to directly measure the impact of social media.

Now consider the absurd scenario where a user returns to the same website they visited on their mobile via facebook but this time on their computer via a brand query in Google search, that ultimately leads to a conversion. It looks like search earned the conversion and they did play a role but so did facebook, however because it was across two different devices – multi channel attribution within Google Analytics fails, even when looking at the assisted conversions.

Worthy Case Study Material

Recently Google announced a major upgrade to Google Analytics named Universal Analytics. One of the big changes with Universal Analytics is that you can provide a unique user identifier into the tracking and use the identifier across devices.

The case study I want to see from someone like Forrester is a collection of big businesses who implement Universal Analytics alongside a raft of user interface components throughout their sites designed to capture something unique about those users and across all devices.

As an example, a user views your website after a referral from facebook on their mobile but doesn’t convert. The website could ask the user to sign up for an account with an incentive or to join an email database.

Now that you’ve got a unique identifier for the user, you’re now in a position to track the impact of the facebook referral if the user happens to come back, either on the same device or a different device (tablet, laptop, computer, TV, ..) and purchases using the same unique identifier they provided on their mobile, such as their email address.

I don’t know if social media is thriving or dying as far as businesses are concerned but I know that we won’t have that answer until everyone gets a lot better at media attribution across the board.