SXSW day one

March 13, 2010 at 6:07 pm

A few months back I drew the lucky card (long straw?!) at 383 Project and so for the next few days I’ve jetted off to Austin Texas for this years SXSW. For those who don’t know SWSW is a huge (in every sense) film, interactive and music festival. I’m here for the interactive section of the festival and as such, will be uniting with other designers, developers and entrepreneurs for a series of talks and panels on a variety of subjects. This year there are over 300,000 people in attendance, so it makes for a pretty intense and interesting time!  Where I can find time I’ll try and pop notes up from talks I’ve been too. Apologies for any typos – most of this is being typed on my iPhone!

Beauty of Web

This first talk was by Cennydd Bowles of Clearleft. It covered the topic of ‘web masterpieces’ and explored why, unlike many other mediums such as product design, or architecture, there are still no stand-out ‘masterpieces’ when it comes to web design. With the web still in it’s relative infancy, at only 7000 days (ish), Cennydd explored whether the web was an appropriate medium for sites to be considered ‘art’ and explored the psychology of what makes a Beautiful site.

First, we looked at the responses to ‘art’. These were loosely defined as:

• Visceral/first response/precognitive response. Mainly covering colour/form etc. This approach to the web is problematic as it rewards initial attraction over ongoing usability. This is a ‘very visual’ approach to web design and often results in the emergence of following of trends and fleeting ‘aesthetically pleasing’ fads. Think of CSS trend sites and galleries which rate the initial aesthetic only and you’re pretty much there.

• Behavioural design. This is where the usability of a site comes in to the mix. Examples such as facebook etc were shown as examples of this good behavioral practice, where the ultimate purpose of why the user interacts with the site, dictates the aesthetic.

This approach I think is where a lot of division comes in to web design. There’s very much a two camp mentality of those who want a ‘useable’ web vs those who want sites to be more viscerally satisfying. The concerns here are all about user mapping and messaging and allowing that to take precedent over the ‘look’. As much as this approach works or fulfills the requirements of the user, it doesn’t always excite. It was argued that often a solely ‘useable’ approach results in a mediocre visual experience (something that by and large I’d agree with).

• Lastly, we looked at ‘Reflective design’. Rather than just combining the two points above, this approach is far more wide reaching and only really emerging in very few cases. Like art, ‘reflective’ web design can be quite subjective-’how does it make me feel?’ and ‘Do I want to share the experience?’ are good indicators. Sometimes sites this approach will be a little less useable, but the user experience is ultimately more satisfying for it. Reflective design allows users to ask ‘What does it say about me as a person? What makes me tick?. In the same way as big brands make people feel something that can’t quite put their finger on, ‘reflective design’ has a similar output.

The reflective web is far slower to emerge than a ‘trend’ and taps in to other changes-it involves wider tech development, better connections, more web fonts, a larger palette of colours etc. and a number of other shifts to contribute to it’s emergence.

Cennydd then covered a few points that he believed would lead designers more in to the path of reflective design. These were:

• Get emotional – Experience isn’t just about usability. Use stories (old media recognises the power of this) content is key.

• Think Bigger and get beyond the idea of the site existing as an ‘island’ entity within the web.

• LEAD- This I think is a big one for agencies. The responsibility to lead and not always allow design to be diluted by committee is a good thing. There can often be a ‘too many cooks’ approach to web design, which a definate differentiator when compared with art. Clients pay for a clear vision from agencies as it’s important that this is shown in the way we lead the design stages of projects.

• Think long term – When designing, think about what’s susainable beyond the visceral. Keep it interesting and surprising. This may involve rewarding visitors to keep the experience fresh upon their return, or hiding hidden treats foe those users who delve deeper in to content.

• Broaden your horizons – Again, I think this is great advice. Web designers shouldn’t just be looking at other websites for inspiration (if at all!). As designers we should often look at design beyond the web, enabling us to analyse and pinpoint what makes users tick on many levels.

• Be brave. – As well as transcending how we translate design to clients, this is often about how we translate ideas and understanding within our industry peers. We should talk about how we ‘design’ together and focus less on techniques (flash vs HTML etc) – discussions should be about deeper design philosophy and not just ‘how we will build it’.

All in all a good talk and a nice start to SXSWi.

What are Analytics?

The second talk was titled ‘What are analytics’ and was given by Margaret Francis of Scout Labs and Blake Robinson of Attentionusa.com.

(the notes for this talk are a little more sketchy as a lot of the focus was on screen visuals and Q&A).

Why bother?

• essential for understanding what’s going on.
• critical for tying social media expenditures to business outcomes
• Analytics are the key to mainstreaming social media marketing into the larger organization.

Useful analytics aren’t always about volume or regurgitation of data. Also important is the percentage of share relative to competitors. Putting things in to contextual relevance to the competition is key.

Basic metrics are loosely defined as:

• Number of mentions, by type, source or channel
• Key themes/emerging memes from conversations
• Most viral content, as measured in links, retweets, traffic, views
• Top sources, as measured by volume, influence, engagement or relevance.

Customer intimacy gives a competitive advantage it allows you to get a beat on social media.

On challenge is dow do you distill the themes? You can’t read it all! Qualitative analysis allows you to apply a weighting to filter the most important over the the least important. Scaling to match the data volume doesn’t work – analysis to filter does.

A company with lots of follows can’t engage with all of them. Figuring out what ‘influence is’ is about google searching the influencers – this gives you a handle on who in your streams has ‘influence’. This is about number of follows – it’s about blending that with job role, location, tone of voice etc.

Insight metrics

These allows you to measure your share of voice compared to competitors. How do you mainline positive data back in to your widest streams?

Top sources as measured by volume influence engagement or relevance

Sentiment – This is the great divide between companies. You can make people agree with each other about 85% of the time regardless of tools. You’re looking at the balance between useful or not useful. Is it helping you find direction? Is it informing you of swings and themes? If so it’s working. It’s NOT about 100% of tweets sentiment matching. (false positives etc – it all needs context).

Gaining a business value

- One important factor is to cross ref the foundational metrics against the company or brands own metrics (campaign data, google analytcis, facebook fans, sales data. (it all adds meaning).
- Age demographics are helpful in ascertaining  (sysomos). Info is extracted from user profiles. You cant judge age based on language.
- by job roles.
- geo data a little suspect too – are you from where you’re tweeting?

Sentiment

The great divide between companies. You can make people agree with each other about 85% of the time regardless of tools. You’re looking at the balance between useful or not useful. Is it helping you find direction? Is it informing you of swings and themes? If so it’s working. It’s NOT about 100% of tweets sentiment matching. (false positives etc – it all needs context).

Business Metrics

How do you correlate this with money>

• Social media analytics can’t measure ROI as they don’t have the R or the I. Google analytcis is more well suited to this as it allows you to set specific goals.

You can measure….

• Most viral content, as measured in links, retweets, traffic, views.
• Campaign performance: views, traffic, reach
• On site conversions, ecommerce and other
• Correlation with sales.
* Product extension ideas.