Recommended reports
State of Social Report 2011
Econsultancy's State of Social Report, produced in partnership with LBi and bigmouthmedia, is the most comprehensive study of its kind around the strategies, tactics and websites companies are using to harness social media for marketing, sales, customer service and other business objectives. The research, based on a survey of more than 1,000 companies, benchmarks budgets, resourcing, measurement and barriers to success ... plus much more.
JUMP Magazine: Issue one
JUMP is Econsultancy's multichannel magazine in support of the multichannel event, JUMP. In this first issue of JUMP Magazine we focus on a number of areas that will help you to join up your business, including why brands should launch print catalogues and magazines, how to track inbound telephone calls with web analytics, emerging multichannel job roles, and why retailers need multichannel returns policies.
CEO at Econsultancy
16 March 2005 16:14pm
It’s now 7 years since I wrote my book ‘Web Project Management’ which put forward a method for delivering larger scale web projects. Happily it continues to sell well and, indeed, the majority of it I still think is valid and useful.
Sometimes I’m asked whether I should be updating and revising it. The answer is yes, but I’m not sure I’m up to rewriting the whole thing. So what areas would I think need updating?
As I wrote back in 2001 in a forum post here ("Web Project Management" - what do I think 1 year on?”) the biggest missing component is the need, in my view, to integrate a user-centred design approach.
User, or human, centred design has a recognised process. Indeed, it has an ISO standard (ISO 13407 which you can buy from ISO). Interestingly, at our recent Usability Supplier Showcase all of the leading agencies presenting were agreed on their methodologies, even if they are graphically represented in a different way, and all follow the UCD approach. (If you download the presentations you’ll see these processes).
So, user-centred design is the right approach for producing a web site that works. But I think there is still something missing if you *only* use UCD in your site build process. What about search engine optimisation thinking? Accessibility? Web Analytics? Copywriting etc.?)
Is anyone aware of a web design process which incorporate all of the above? In fact, is anyone aware of a generally accepted web design process at all…?
I’m not. So following are some thoughts on how you might get from ‘paper to page’ i.e. from a paper prototype to a live web page whilst integrating not just UCD but some of the other important thinking now required of successful sites.
I’ve somehow managed to crow-bar the titles into all beginning with P…
[NB the process below only concentrates on the page/site design element of a much wider process. For example, I’m assuming you’ve got your value proposition, brand, commercial strategy, target audience etc. properly defined…]
1. PAPER
At this stage you’re scrawling things on flip charts, on pieces of A4. Sketching out possible information architectures / sitemaps, navigation, page templates and so on.
UCD techniques you’ll likely be using:
# Personas, Scenario Modelling, Contextual Enquiry, Card sorting…
Specialist people you should definitely have involved:
# Information architect, Customer Experience architect (if you have one…)
Other specialist people you should involve (and may not have thought of):
# SEO specialist – so they can start to get a feel for key site themes / content areas
# Content Management / Tech infrastructure people – so they can start to get a feel for how content and metadata might be architected
# Web Analytics / Measurement lot – so they can start to figure out likely tagging / grouping / segmentation models based on user needs and business goals
# Online marketing people (PPC, Affiliate, E-mail etc.) – so they can start to understand opportunities for targeting, content personalisation, partnership strategies and so on
For the above you don’t really need these people to do that much at this stage but just make sure they are in the loop. Not least because they might help influence the process at this early stage with some good ideas or head off some really bad ones.
2. POWERPOINT (or Visio, but that doesn’t begin with P…)
At this stage you’re building Wireframes of page templates (example of these in the Template Files for Web Projects pack). Essentially you’re beginning to fix more firmly the sketches you refined in the Paper phase, but still not going into expensive design and coding.
I think adopting the ‘T’ prototype approach works well. If you imagine looking at a site map, the shape of the ‘T’ means covering all the top-level section homepages and then drilling down through to the bottom level content in one section. This allows you to go “broad and deep” and cover most aspects of possible customer journeys.
UCD techniques you’ll likely be using:
# Expert Reviews, Card sorting, Usability Testing…
Specialist people you should definitely have involved:
# Interface Designer, Content Management Specialist, Copywriter (as well as the Information architect and Customer Experience architect)
Other specialist people you should involve (and may not have thought of):
# SEO specialist – so they can help input on element like navigation, headings, footers, site inter-links, page content, page structure and so on from an SEO point of view
# Web Analytics / Measurement lot – as previously, so they can start to figure out likely tagging / grouping / segmentation models based on user needs and business goals (e.g. content affinity groupings, key site processes to measure)
# Online marketing people (PPC, Affiliate, E-mail etc.) – so they further understand opportunities for targeting, content personalisation, partnership strategies and so on. Also to look more carefully at this stage at how screen real estate is used for cross-promotion, cross-selling, internal promotions etc.
# Customer Service specialists – so they can input on where, and what, forms of online customer service are embedded into the site architecture, and relevant web pages (e.g. chat and co-browsing only through the final stages of the checkout process vs. an online knowledge base on all pages…)
# Content Management people – so they can define and refine the content taxonomy, and associated metadata, to manage the content as smartly as possible
Accessibility consultant – to make sure your proposed site / pages will be accessible
3. PHOTOSHOP
Making it look pretty… Well, that’s a bit of an understatement. However, this is essentially about adding the visual skin to the skeleton (information architecture) and bones (page templates) that you’ve already built. Clearly branding and the more emotive aspects of design are important here.
UCD techniques you’ll likely be using:
# Expert Reviews, Focus Groups, Usability Testing…
Specialist people you should definitely have involved:
# Graphic Designer, the “brand guardians” (Brand Manager, Marketing Director etc.)
Other specialist people you should involve (and may not have thought of):
# Accessibility consultant – to make sure your proposed site / pages will still be accessible once the designers have had their way…
If you’ve done phases 1 and 2 properly I don’t think you really need re-involve all the others at this stage as the content and structure shouldn’t change appreciably.
4. PROTOTYPE
This is where you build a working HTML prototype which people can interact with as though it were the real web site. Again, you can follow the ‘T’ prototype approach described earlier to minimise effort without sacrificing thoroughness.
UCD techniques you’ll likely be using:
# Usability Testing
Specialist people you should definitely have involved:
# Well, pretty much everyone really… I’m not quite sure how best to manage this potential bun fight (ideas welcome…) but really this should be the last chance for all the specialists to have their say and plan their activities for when the site goes live.
5. PRODUCTION
This is where you actually roll out the design and build of the whole site.
UCD techniques you’ll likely be using:
# Usability Testing
Specialist people you should definitely have involved:
Again, pretty much everyone, although now in a more doing / operational role rather than an advisory / consulting capacity. So, for example:
# SEO specialist – will be helping define actual file names, directory structures, internal linking, tweaking the code, checking the site will be properly indexable, preparing for directory submissions etc.
# Analytics person – will be configuring reporting against key metrics, ensuring tagging is embedded as required, defining important site processes and setting up reports for those, setting up user segments, content segments etc.
# Copywriter – will be very busy actually creating the content, whether body content, navigational or instructional content, as well as refining the calls to action, the user journeys etc.
# Content Management – will be busy creating, storing, managing content in its various forms, adding metadata according to the defined taxonomy, refining workflow processes, training content editors and managers in how to add and update content etc.
# Accessibility Consultant – just making sure that all is still on track to provide an end result that is properly accessible
# Customer Service - making sure that the necessary systems integration (e.g. to call centre) is in place, that people are properly trained, the required processes are in place etc.# And obviously the coders, designers and project managers who will be busy knocking out templates and pages…
Other specialist people you should involve (and may not have thought of):
# QA / Testing folk (all sorts of testing – compliance, tech infrastructure, security, functional, accessibility, code validation etc.). This all needs to happen before ‘go-live’.
Above is just an initial schematic of how things might work. The important point is a) to use a user-centred design approach which begins with “lo-fidelity”, iterative, design techniques but b) to build in SEO, accessibility, analytics, content management thinking from the start.
Any thoughts?
Ashley
MD at Screen Pages
18 March 2005 14:32pm
Its probably true to say there isn’t a single way to approach any project and it is dependant upon the task owner as to how processes are undertaken and therefore upon result.
As a designer I am shocked to find my own inclusion within the project cycle beyond mid-way and my inclusion as being to ‘add a visual skin’ or ‘make it look pretty’. When applied correctly design makes a company or product more than the sum of its parts – it becomes a brand. A brand sets off many psychological triggers in an audience and be used to communicate much more meaningfully than to sell or promote product or message. It creates desire.
Through brainstorm and questions I would suggest it would be a designer’s role to assimilate how the CEO, the marketing team, the sales team etc. see their company from the outset. What is important for them to communicate, how they communicate at present and in the past and how the public perceive their service. Find the strengths, the flaws and then unite the ideas discussed into themes on which to build the web site.
With a coherent set of themes or principles to build upon the designer would then sit down with specialists brought to the project including online marketers, content managers and information architect. This would then set out the structure (skeleton) of the site while always keeping an eye on the original larger, targeted goals brainstormed in the first meeting – you should then have a set of clear messages expressed creatively and clearly throughout a sites structure.
Then the physical design can be started – and the design process can be quantified non-emotively against original objectives and create a coherent whole of marketing, content and navigation. This process will have a number of revisions and further analysis via first stage usability testing, accessibility, content management etc. before going to a build and production as the article lays out.
Mark Spires ()
Head of Design, Screen Pages (www.screenpages.com)
MD at Screen Pages
18 March 2005 14:51pm
There are two issues here:
1) Make sure that the site is indexable. This is about robot-friendly building (flat URLs, use of Flash etc). Getting this right is mechanical and only requires technical savvy. As navigation here is critical and, in an e-commerce context, navigation is informed by product categorisation, some thought needs to go into making navigation accessible & comprehensible to spiders.
2) Making the content friendly/"relevant" (possibly at the expense of branding). Internet buyers do not search for "yak yak jackets", they search for "ski jackets". Product names/promotions - as SEO assets - must be relevant to searches and readable. An understanding of categorisation & product naming in an SEO context is therefore vital at the design stage.
Roger.
E: roger@screenpages.com
W: www.screenpages.com
CEO at Econsultancy
18 March 2005 15:07pm
Hi Mark
The 'make it look pretty' was designed to provoke.. ;)
I agree with you on many points (iteration, brand importance) and, of course, different projects will indeed need to be approached in different ways - if you're doing a brand-led micro-site or game it's likely to require a different bias in skills during the project from a large scale retail site.
However, the way you describe the process, it sounds like it's the designer who is leading the process. I don't think this is correct. At least, it would depend on the nature of that designer and skills he/she had.
In an ideal world l think I'd like to see a "commercially-minded customer experience architect" in charge of my web design project. This person would be a jack of many trades, a great communicator and project manager, and would be able to walk the line between the needs of the user and the needs of the business.
In my experience a web 'designer' tends to have a bias towards strong skills in either *interface* design (and usually client-side coding, usability, accessibility) OR the more "creative" / "art direction" / "ideas" side of things (often including illustration, animation, copywriting, straplines etc.). Some designers do combine both, but not that many. Each has its role to play and there's no reason why a designer should have to excel at both. But I don' t think either are in a position to lead the project all the way through?
Ashley
-- at --
18 March 2005 16:50pm
I havent got an incredible lot to add to this.
My comments would be that I see website builds treated as too much of a focus on architecture component before customer facing component.
I would always focus on getting the presentation layer correct first. Even if that means the designs are fully finalised and a full dummy site is built from scratch. Once that is nailed down then the business rules feed off exactly what the customer sees/needs.
The business rules dictate the content engine you need and the content engine dictates the architecture you will build it on. The architecture dictates the technology you use. The technology dicates the personel you need.
I see way too many companies saying - "this is the CMS we have chosen and this is the architecture - now lets build a website."
The beauty of getting the customer facing part right first is that its usually the cheapest to build and the easist/cheapest the change. Once that part is right - getting the rest is a helluva lot easier
Maybe im trvialising the process :)
Digital Marketing Consultant, Trainer, Author and Speaker at SmartInsights.com
18 March 2005 17:14pm
That's a very good question - every agency has some sort of process, but its interesting to ask what is best practice in terms of activities for the different stages.
Today, the majority of site builds are re-designs of existing sites, this means that there is a lot of learning we can get from existing sites - so I think there has to be a stage 0 which is Analysis or Research, but to keep with your mnemonic Preparatory Research or Preparation.
Here are three key aspects of Research and analysis.
1. Web analytics.
--------------------------
When I work on re-designs one of the first questions I ask after what outcomes are you looking to achieve,? and from whom? is can we use your visitor tracking? (to work out what works and what's popular). I am AMAZED by the number of often large companies who say , oh-yes, i think we have something like that, but I can't remember its name or I never look at that. This is understandable due to the quality of some tools, but running an analytics tool like Clicktracks against existing log files is often a great way to learn about customer needs and behaviours (I know log files aren't accurate, but they don't need to be too accurate for this)
2. Search term research
-------------------------------------
Another aspect of Research which I use to get in the mind of the customer is to look at the search terms tools to understand the way the customer thinks when researching a product or looking for information.
These are the tools I use: http://www.davechaffey.com/Internet-Marketing/C8-Communications/E-tools/Search-marketing/Search-marketing-keyphrase-tools
Apart from being useful for SEO/PPC, it also helps show how well the current site delivers against visitors information needs in its nav and content.
3. Benchmarking against competitors/portals
----------------------------------------------------------------------
Then there's benchmarking against competitor sites. I have a standard procedure for benchmarking commercial sites (and related e-communications) in terms of their effectiveness/efficiency for Acqusition, Conversion and Retention which I run through before I even think about a new design. Quite a few agencies now offer this type of Online Marketing Evaluation as I call it.
I know your looking with your Ps at the narrow aspects of the Design proces, but the model we use for the design process also has more emphasis on aligning design with business goals, so its something like:
In the new IDM Digital Marketing qualification the broader process we suggest in the module on web design has these stages:
1. Evaluate performance.
Use web analytics and analysis of marketing outcomes to see what existing web site is delivering, if the web site design project is a redesign or relaunch.
2. Research market.
Understanding different site audiences and customer groups within the market including their characteristics, needs and the barriers and facilitators to using digital media. Competitor web sites and intermediary web sites will also be reviewed at this stage.
3. Set goals.
Setting SMART objectives for online marketing, for example Sell, Speak, Serve, Save and Sizzle.
SMART stands for:
• Specific;
• Measurable;
• Actionable;
• Relevant;
• Timely or time-oriented, a historical analysis.
4. Define communications strategies.
Define strategies to achieve the objectives through web site design. i.e. Which content, experiences and offers will be offered to different target audiences.
5. Design and implement site.
This is the aspect you were referring to in this post Ashley?
6. Promote site.
Attracting relevant visitors to the site. Often there will be re-launch communications around a site re-design. From a web site design perspective, it is particularly important to accommodate search engine marketing.
Dave Chaffey
============
Internet Marketing trainer, consultant and author
eResources and Books: www.marketing-online.co.uk
Blog: www.davechaffey.com
Managing Director at Panlogic Ltd
18 March 2005 22:03pm
Ashley
Was pleased to read your post and the considerations you have raised on User Centric Design (UCD).
There are three core points I would add to your comments
1. Brand Considerations
Whilst I of course agree on user centricity being at the heart of good design I’m also of the belief that consideration of the specific audience type being targeted need to be considered.
Recenntly we went in to talk to a major car hire group. Their concern was, yes of course about accessability and usability in all its manifestations (layout, DDA, navigation etc.) but also considering the brand in how the site communicated with their audience. Their concern was that the current site appeared too corporate and formal for the growing number of consumers that were using their product offering.
In order to accomodate the leisure traveller, usability would need to be different to the current site. Yes, certain elements would remain the same but certainly not all. The T shaped approach is a good benchmark but then it is not a panacea for all deliverables.
2. Behavioural Targeting
Panlogic has over the years developed research looking both at various age groups and their differing behaviours and attitudes to the web.
In the 17-22 year olds we identified three separate behavioural groups (Fun Seekers, Enthusiasts, Focussed). Without going into the detail of the research we pointed out that well designed sites would cater for all three of these behavioural groups.
Incidentally, powerful marketing would also result from targeting these groups through behaviourally-led creative being placed on media sites that talked to such groupings.
3. International audience
My final point would be the age old maxim "think global, act local". Invariably what is good advice in the offline world is equally sound (though with added elements) in the online world.
Panlogic opened an operation in China last year. What became abundantly clear very early on is the different approach needed when talking to that audience compared to our UK mind set. As one example, e-commerce in China is at a very early stage. Most sites are there for branding and information purposes and with an age-old distrust for the government, users expect differnet stimuli and therefore differing design led considerations.
Yes, this means being user centric but it also demands a wider culture-centric approach as anyone will know who has arrived at a US site of a European brand.
William
On 16:14:05 16 March 2005 Ashley wrote:
nothing at nothing
22 March 2005 16:46pm
I see so many companies focused in on strictly the web right now that I would like to suggest a better method of web page design processes. Being focused on only the best practices for web page design really misses out on larger opportunities in today’s cross-channel communications. Before spending any time on web page design, I think you (as a company, etc..) really need to do an analysis of what the best ways of reaching out to your customers are. Each is going to be unique and there is no longer one optimum way of achieving this.
Let’s say your customers use your web site to do research and then take that information into your stores to make the final purchase. (which in fact happens a majority of the time right now).
Is it more important to spend your time designing the optimum online shopping experience or is your time better utilized by esploring more efficient ways to or incentives to transition them off the web and into the store to make a purchase.
It’s just something to consider in this discussion. In my book, a thorough analysis of your market is crucial for any online success. Human Factors Analysis -> Design -> Focus Group -> Redesign -> Focus Group -> Build -> Test -> Release
Director at Foviance
07 April 2005 16:40pm
Ashley,
An excellent posting - perhaps you should write a "Dummies Guide"!
I have some comments regarding the User Centred Design aspects of the "Paper to Page" Guide.
Even before Stage 1 Paper, if there is budget and real User Cented design is being followed, then Requirements Gathering is the first step for a User-Centred Designer. As most UCD Specialists are trained somewhat in Psychology, they are expert as sifting "real" user requirements - what we refer to as preference data. According to HCI standards this stage should incorporate the users end "context of use" into the equation.
We also believe at The Usability Company that it is never too early to be thinking about the website metrics - you place the analytics work later, however if there are business metrics to be collected through a proposed site, then they should be captured as early as possible - again before paper is even touched.
The final point I would like to make is that keeping Accessibility Issues at the forefront of the teams mind - even at paper stage is important - true, the real work comes when the site has been coded, and then automated tools, and disabled participant user-testing can take place, however we must consider that those users with dyslexia may need to have their needs considered when at first the text is being written!
Catriona Campbell, The Usability Company
Managing Director at Progenit
26 April 2005 16:14pm
The process is still as sound as ever but what has changed is the business and user context.
Seven years ago the web brief is was most likely authored by a marketing professional. Now that brief is just a likely to come from any other part of the business.
Advice to those managing user-centered design today:
1. Establish hard business measurements for the site such that they can inform the design process. Don’t worry if they can’t be directly measured on the site - design to them regardless. The fanatical championing of the user needs is of less value if the site fails to deliver or be operable by the business. The trick of the designer is to reconcile the two.
2. There aren’t many green field sites on the web anymore. Even if there were you need to consider the broader customer experience. What you are building is adding to that experience. Use the techniques Ashley descirbes to consider the journey before and after the site (or bit of it) that you are building.
3. Talk to stakeholders. Regardless of the process you use web project documentation is often too abstract or doesn’t communicate the subtleties. Supplement deliverables and process with ’shuttle diplomacy’
4. Don’t be afraid to challenge existing business practice but don’t expect it to change just because of the way you’ve design the website. Changes in behaviour need concious effort, justification and management will. User-centred web design can be a powerful tool to cause change - especially when its driven by direct customer insights.