Social Media


6
Nov 11

Can I Get Dirty With Your Data?

The initial steps of data analysis; Collecting data from various sources, ensuring the accuracy of the data and then putting it all into a legible format. Doing the grunt work of social media analytics (or really any data analysis) is not the most fun or glamorous part of the process but through my experience I’ve learned that having this level of intimacy with your data is an important  part of in ensuring the delivery of solid insight and analysis .

Getting In Close With The Numbers

Having to compile the different parts of your data puts you at a level of closeness with it that is not as easy to get otherwise. Entering in each source and dealing with all the tools you see almost instantly when there is a gap or significant change within your metrics because you see all the data points in front of you. Compare this with just getting summary data and you just get a snap shot of the entire picture of what is happening. Let’s just say this is like taking your data out for the day compared to having a quick phone call with them. One is much more effective in getting your desired result than the other.

Dressing up your data. Picture C/O Exey Panteleev

 

Having Your Way with The Data:

When all of the grunt work is done and your data set is put together and prettied up you tend to already have a sense of what you want to explore when you actually go and do your analysis. You’ll know where all the sweet spots are and try to determine why things happened the way they did. Experiencing all the parts of the data as a whole rather than in separation there’s a better understanding of the linkages and causes of change in the data. You are better able to see where something went up as a result of X or Y action. By having your hands in everything means you are able to get in there and push the right buttons to get the best end result.

Growing Old Together:               

In my work there have been some data sets that I’ve dealt with for a LONG time, sometimes since their inception going months back. Much like a long term relationship with a person you can get a read on when things are going as per usual or even great but you can also see when there is a significant drop off or unexpected event. In this sense your mental set is more finely tuned to see the trends and linkages within your data store. When your data set is mad at you, you know almost immediately to do something to fix it!

The 'sexy' end result. Picture C/O @BenLucier

 

Setting Yourself Up For The “Money Shot”

When all is said and done, the data is all in and you’re fully into writing the report, by knowing your data you are able to tell the right story and enter in the best summary metrics to share with your client. In the end, the job of an analyst is to see the data and make sense of it in the best way possible for your specific audience. Your “money shot” in all of this is when your client is sent off with the best insight and recommendations that they can take action on. When they are happy that they are seeing improvement in their operations or learning something useful that they can apply to their business this means increased results and profit for your organization. Essentially  something people will pay to see ;).


24
Sep 11

Is Sharing the New ‘Like’ on Facebook?

Getting users to ‘like’ their content has been a strategy of many Facebook fan page owners in recent weeks because it quickly amasses simple engagements and creates awareness of the page. But with users given better ability to curate the content on their news feed and profiles, being able to focus more so on app and ‘life period’ content, this mode of user engagement may not be as useful moving forward.

Engagement above the fold:

With the new OpenGraph optimized news feed, a separation has been made between content more useful to the user which is the first thing they see along with everything else further down the page. As this feature allows us to grab just quick information on the go we are less likely to scroll down the page and view everything else. This includes that post to ‘like’ carrots from the food fan page you liked a few weeks ago!

Can you ‘share’ this?

Without the full changes to Facebook completely implemented it’s hard to make an exact prediction to how corporate content will operate (sponsored posts anyone?) but with demos from the recent F8 and some functionality already in place some predictions can be made.

Prediction ONE: Posts of major announcements and big campaigns will reach higher on the news feed. Why? Much like with Twitter, people want to be the first ones to share breaking news and content with their friends. As a result this may lead to multiple content shares (increasing volume relevancy) and friend discussion on the topic ultimately driving this content higher up the news feed.

Prediction TWO:  Branded apps will become a stronger way to engage with fans. Why?  The creation of brand relevant apps where users can either create content or where users are encouraged  to return to on a regular basis is something that will either build the opportunity for a news feed impression (ex: creating a Spotify playlist) or because of usability reasons will create constant brand exposures as a result of using the app. With the new Timeline taking more of focus on life activities instead of user engagements apps are a way for users to become more active with a brand and integrate it into their online ’life file’.

Is it Time to Rethink Brands on Facebook?

All in all Facebook made some big changes to how the site will operate. Without looking into the future too much I think it’s going to be interesting to see  how brands will need to adapt to continue to stay relevant online with consumers. I don’t think it’s going to be as simple anymore as just putting out rounds and rounds of content for users to ‘like’ and comment on. I feel brands will need to create deeper relevance to continue to drive their messaging to consumers.


13
Jun 11

Writing Great Analysis (in 100 Words!)

This post was a bit of an interesting challenge. If you frequent my blog you know that I tend to be a bit long winded when creating posts. Its something I’m used to and enjoy. Well in providing social media analysis, writing long essays does not flow very well. One of my manager’s  came up with the challenge of writing a post about writing social media analysis in exactly 100 words in a good exercise to test myself. Well you’ll find below that the challenge has been completed:

 

What Makes Great Analysis?

Clear and succinct communications are needed for strong analysis. Start with a brief introduction:

“Requesting user action generated significantly higher engagement”

Place key data up front:

“Asking users to click “like” for support generated an XX% increase against the campaign average”

Provide additional insight:

“ In total, the EPM for the period was XX against the current period average of XX.”

When providing recommendations, link suggested actions to outcomes.

“By directing fans to act on posted content they will be more likely to engage with subject matter”

What makes great analysis? Easily read and actionable information.

 

When writing this I found that I had to strip out a lot through multiple drafts. While I’m used to writing under a restricted word count writing in 100 words I found to be really tight.

Let me know in the comments what you think of challenge! I’m curious what people feel about its effectiveness and whether this sort of writing is useful in all cases. If you like you can also get in contact with me at@kevrichard or kevin@kevrichard.com


20
Sep 10

Do you want Foursquare with that?

The most important concept in social media marketing that I’ve learned is that for efforts to be effective they should be connected to how the firm earns money. Having fans and creating engagements is great but when it’s not driving firm revenue in some way, shape or form it’s not useful.

That’s why when I read Mashable’s recent article about the McDonald’s head of Social Media Rick Wion celebrating a 33% increase in Foursquare check ins with a pilot program they ran I was disappointed that there wasn’t more to it. While the goal was apparently not to increase sales immediately this campaign could have been much cooler in seeing if McDonalds actions actually created an increase in sales( or not…).

How McDonalds could have measured the campaign (if they had wanted to):

McDonalds could have easily done one of two tests to see if this campaign was effective. First and probably most easy is that they could have just compared sales and check-in results from previous days for an extended time period to see if this 33% increase in check-ins actually led to an increase in money into stores. If revenue saw a significant jump when foursquare activity spiked (barring other variables) Mcdonalds would have at least some proof to try the campaign again.

The second test is running the campaign on a smaller scale comparing it between two similar markets. One would run the Foursquare campaign and the other would have no promotions. In the end, sales and check-in figures would be compared between the two. If the market with the Foursquare campaign saw an increase in check-ins along with revenues in comparison to the control market McDonald’s could further test this campaign and have it rolled out on a larger basis.

While I don’t know the full goal of the McDonalds Foursquare campaign I feel that despite the celebration within the Mashable article it’s still important to clarify that social media in business is not successful by engagements alone. If I would go on twitter or another network and tried giving away something for an audience action people would have jumped on it as well. In the end social media campaigns need to be attached to financial drivers because most companies can’t operate on foursquare check-ins alone. ;)

Do you have a question or comment ? Feel free to reach me at @kevrichard or kevin@kevrichard.com


11
Aug 10

SXSW 2011: Canadian Panels To Vote For

Its that time again! As last year’s post showing the many Toronto (and eventually Canadian) panels up for voting was such a success I decided to give the list another go this year with the goal again to try to get as many Canadian Panelists into SXSW 2011! (And to make peoples dreams come true ;))

Find below an initial list of people who have submitted their panels up for review. This post is open to everybody, if you know of someone or have your own panel that you would like to submit feel free to @me on twitter or send me an email (kevin@kevrichard.com)

THE LIST:

And theres more! Came across this great listing of Vancouver SXSW sessions on Tech Vibes. Not to take away from the awesome work done in creating the profiles you can check them out and vote for them HERE.

I know there are many other panels out there so folks let me know if you want your session up! For those coming across this post please support these adventurous folk share what they have to say in Austin this spring at SXSW.


29
Apr 10

A Framework for Social Marketing Analytics

Like with everything that is done in business, it has to be done for a reason or else it’s just wasted effort. Measuring simply anything from your social media initiative without a reason behind it is just that – a wasted effort. In reading “Social Marketing Analytics” by John Lovett and Jeremiah Owyang this point is further proven. In measuring social media what you measure needs to be attached to specific business goals, not measured just because the number is available.


The paper by Lovett and Owyang starts off with a good definition of what Social Marketing Analytics is (what can I say I like when things are plainly defined):

Social Marketing Analytics is the discipline that helps companies measure, assess and explain the performance of social media initiatives in the context of specific business objectives.

Pay particular attention to the last part – “…in the context of specific business objectives.” What did your social media initiatives achieve for your company? Social media measurement is not just looking at the numbers of fans go up or down (or whatever shiny number there is out there) but looking at the end result of what your work achieved for your organization. Very simple but great stuff.

Lovett and Owyang write about a three layered approach to Social Media Measurement, which I feel is effective in further demonstrating the need for measurement:

1. Strategy: The reason behind the initiative.

2. KPI (Key Performance Indicator): What the measure of success or failure is with this initiative.

3. Granular Metrics: The information or inputs that need to be gathered to allow you to formulate the KPI.

Social Marketing Analytics Framework

In keeping the paper accessible to general business practitioners Lovett and Owyang present four major strategies (although they mention that there are more) along with suggested corresponding KPIs. I’ll leave the KPIs for more in depth reading in the paper but I thought I’d present the strategies as these apply to many businesses:

1. Foster Dialog: Developing conversation around your company or brand. This is the development of content to create interest in what you’re trying to achieve.

2. Promote Advocacy: The development of ambassadors for your organization. Encouraging people to actively discuss and promote your brand online.

3. Facilitate Support: Provide a channel where customers are able to get their problems/concerns addressed in a timely manner ( Ex: @Rogershelps or @Toronto311).

4. Spur Innovation: The development of active discussion around your brand or initiatives focusing on the topic of improvement. Essentially putting like minds together online in developing ideas to make things better.

Overall in reading this paper it really puts it out there that:

A: Knowing what you want to achieve with social media is important and,

B: Measuring your progress (and not wasting your time) needs to be followed right after.

I really encourage you to check out the paper as it’s a great first step in developing a measurement program. I would also add that while it’s a helpful framework, it’s not a complete instruction guide. Measurement, like any other business initiative, needs to be scaled and customized to fit with the specific organization. Measurement and tools that apply for a large multinational corporation don’t have the same effectiveness for a smaller, local business.


13
Mar 10

Geurilla Analytics Simplified

I think now is a good time to bring up this follow up post to the recent Geurilla White Paper posting. I’m sure for some, looking at the paper was an immediate shock of words and numbers! I hope in this small version there’s a bit more understanding of what the paper tried to get across.

So whats this paper about? Social Media Measurement and the common use of machine sentiment engines as a way of measuring effectiveness….. or how un-effective this is based on the research we’re presenting. Getting right down to the point, language is COMPLEX! Thats to say that the words and messages we share with our friends are full of double meanings, slang and one of my personal frustrations lack full context. I’m sure most people have had online conversations where you’ve had to defend what you’ve said as it was read in the wrong way.

Knowing that online messaging is really complex, often times in measuring the success/failure of social media initiatives   firms use a sentiment engine to rate mentions of their brand on a scale of Negative, Neutral and Positive. With this in mind, we the Syncapse Measurement Science team looked to see if this was an a reliable measure.

What did we find? That again language is complex and no two people will perceive a set of  messages the same way.

This research was based on an earlier survey we released on twitter  where we asked  people to rate a set of 20 messages ( all taken from a twitter search for ‘books’).  The most significant finding of this survey is the lack of agreement on the survey as a whole. With such a simple answer set  we did not find  a single pair of respondents (out of 102) who could agree on the sentiment of these messages.

To put a further test in our survey we also put a wrench in the works with a trick question. We decided to repeat a question. While a majority of people rated sentiment of this question the same on each response,  19% didn’t!  So in a span of a few minutes a person’s feelings about a message can change [ meaning you have to be particularly careful in how you time your emails! ;) ]

What in the end do the findings of this paper mean? Well to quote directly  from the paper:

If a group of 102 humans could not agree, how can a single machine output a single score that everybody would agree with?

The current sentiment score system that’s in use isn’t netting  information that should be depended upon to make real business decisions. Language is just too complicated for a human programed machine to define the sentiment of social media messages.

But WAIT there’s more! If current  sentiment scores should not be used in decision making then what should be put in its place? Well I know a team that does some mighty fine work in social media measurement ;) …..  Ok,  so there’s ALOT to measure in regards to social media measurement and no one has yet created the gold standard. As an emerging industry, measurement is still playing catchup to everything else currently happening (which keeps things interesting). What we could agree on with things as they currently stand  is a more pragmatic approach, looking at the measures and feedback that is important to a  brand( or ‘Brand Health’). This means looking at the overall online communications about a  firm and determining what aspects of these activities are most important to have knowledge of.

Have any questions/comments ? Contact me at kevin@kevrichard.com or send me a twitter message .

PS: I know I’ve been on a massive blog hiatus, I’m looking to change this ASAP!


8
Mar 10

A Guerilla Experiment In Sentiment Analysis

As tweeted a few weeks ago the Syncapse Measurement Science team put together an experiment on sentiment analysis. Here is the final result of this research:

Syncapse Sentiment Analysis White Paper

and the data set for those interested in looking at the data collected

Geurila Analytics Data Set

Feel free to leave a comment or give me a tweet @kevrichard .


3
Sep 09

Corporate Twitter Accounts: Balancing Conversation and Frequency

So this is a bit of a long time coming but part 2 of my twitter ‘experiments’ can be found below:

Experiment 2:

My second experiment involved messaging and how much interaction cans a personality or twitter brand can have with its followers. Yet again I decided to take a twitter alias(@socialmediabot2)  as I’m far from the average twitter user. Going to wefollow I decided to follow 50 people (close to the average of 69 )  and posted intermittently hoping perhaps for some interaction with who I’m following … a bit of a downfall of this experiment lack of true connection with people.

Afterwards, starting on the Monday morning I started posting messages from another account. 1 for every half hour for 7.5 hours for a total of 14 with the messages ( TEST #_ )  to see how this would be viewed on my feed. I wanted to see how the everyday  messaging from the people I followed would fit with the planned ‘corporate’ messaging and whether it would overload my feed and possibly cause me to want to unfollow the account

The Result:

Twitter Overload

Twitter Overload

As you can see there are portions of my feed where the messaging overlapped, for a casual user of twitter this may just be too much!  Especially for accounts with little customer interaction (  just sharing links, company news etc) a strong consideration will have to be  made regarding how many times to post daily will work to allow you to build and retain your audience.

Overall Takeaways:

Twitter is still a difficult area to get solid data from and you can’t just jump on twitter hoping for success in your campaigns or awareness.It takes planning and organization but even then trial and error is needed. I think something that is often missed with twitter  for heavy users like myself  is we forget that the vast majority of people don’t follow hundreds of people or have the equivalent # of followers .

They follow friends and organizations that they are close to and that needs to be considered when planning out a campaign. While heavy twitter users may be able to withstand messaging it probably would be seen as spam to these casual users. I think an excellent example of how to avoid this sort of twitter spam is from Best buy and their Twelpforce where they have once central account but many separate support accounts.

Have any questions/comments ? Contact me at kevin@kevrichard.com or send me a twitter message .


19
Aug 09

How I didn't get 100's of twitter followers!

Having a  traditional marketing background,  for promotional actions I always consider the end result or goal. In terms of twitter and social media  though there are a lot of unknowns and what ifs . Overall its untested territory for most people. Steming from a conversation I had recently I decided to do a bit of testing of the twitter platform and see if I can build a bit of test case.  This is part 1 of 2 twitter experiments I completed, stay tuned for the second one to come soon! * Note these  experiments are definitely not experimental and would probably not be replicated if redone, take my conclusions with your own judgement.

Experiment #1:

I think a lot of twitter users have come across the messages “Get more followers now” or ” Get X amount of followers today” and despite the general sentiment of  quality of followers over quantity I’m sure everyone even for a moment  where slightly tempted by this thinking that that more followers equals more people to get their messaging out to.  This is especially relevant  to  traditional marketing mediums as  more eyes =more marketing awareness= more customer action = money!  So I decided to put this to the test, can someone get tons of followers overnight using these follower programs.

The Scenario:

Starting off with an empty account ( @socialmediabot1) I joined the following  social media following programs:

Initially I started following 19 people and I noticed with all 3 of these programs that there were paid for premium options ( so basically paying for followers)  I also noticed that I almost immediately started sending out auto posts from these programs.  From there I left the account completely alone to let the followers come in, opting not to send out anything to not influence the result based on my content.

Result:

#FAIL

#FAIL

I started @socialmediabot1 on a Friday night, deciding to leave it  over the weekend. Checking it once I noticed that I was following over 100 people and had just 19 followers, no where close to the promised massive amounts of followers.  At the end of this experiment I come to above picture, no followers and 12 auto tweets and a suspended account.  I wasn’t completely surprised.

Conclusion:

I am by no means a great twitter user but my account certainly didn’t popup over night and I certainly didn’t expect this one to. A major marketing/branding implication of this is communities need to be built and  its not a matter of massively following people or hoping that  people  run to your brand. Relationships need to be formed  and this takes a lot of time and effort.

Often times I feel that platforms like twitter are advertised as a panacea for marketing and that suddenly you’ll see huge results. FALSE! Like all marketing efforts hard work needs to be invested and careful planning and organizing should be done.  Social Media is just another touch point to your customers, the only ‘magic’ there is the opportunity to have further communications with customers and the general public.


I’d like to hear  what you  think about twitter and  any of your thoughts on  best cases or usage scenarios for companies and organizations so please feel free to leave a comment, send me a twitter message or email me at kevin@kevrichard.com .