10 thoughts from 10 years close

The beginning of November marked 10 years of self employment for me. Although Codegent Ltd didn't incorporate until 1 May 2004 there was a period of time where I worked as a sole agent under this made up name that is now a company of 30 people.

In that time I've experienced a huge amount and so thought I would reflect on some personal thoughts that I hope will be of interest to the audience of this blog.

1. Design by committee never works

We all know this but yet it continues to creep in and happen. This is the number one reason for failure on any project. To achieve great results you need to empower a small, experienced team and give them clear goals. Then have the courage only to wade in and make changes if you spot a potential catastrophe. Otherwise launch it, test it and learn.

2. Take more risks

Overall the only real regrets I have from the last 10 years are when we didn't seize opportunities with greater vigour. Half hearted is worse than not at all. I say the same to clients too. The internet is a very noisy place. You need stand for something, possible at the cost of some users. Your real audience will tell you if it's wrong.

3. User testing always makes for better design

Without doubt validation by your target group cements anything you were previously 50/50 on and helps pick up any clangers. Even experienced designers get a little snow blind in the depths of a project.

It's also a great reminder of just how little people actually care about what you've no doubt been fretting over for quite some time. They just want things to be clear, simple with obvious tangible benefits to them. Then they'll give you their time and money, but not before. It's a grounding experience, and quite refreshing.

4. Stop dumping tech and starting again

By this I mean the cyclical nature of abandoning previous technology to start completely from scratch. Surely now the time has come to stop this practice. Of course those working on the new version prefer a blank slate but this seems to happen in almost every case. What a waste!

To stop this occurring in future my advice would be to take some time to invest in your business data model and backend systems and then develop an API to interface with them. Once this is in play you can refresh your site, CMS, add new channels, like mobile apps, and even syndicate your content to third parties as often as you like. You can even have different teams and companies developing simultaneously without tripping over each other.

No matter how clunky you feel your current system is there is business logic and workflow embedded in there that will not translate into a new scope of work. It will have evolved over time and not be reflected in your old docs. Be prepared for that. A complete rebuild is one step forward and, initially, two back.

5. What's hot and what's not?

The internet is very faddish and agencies are often guilty of pandering to that in order to appear cutting edge. It can feel like you need to be everywhere, which spreads resource very thin. Plus new platforms and trends emerge every week! The answer is pretty simple. Focus on what your audience are actually doing now and how willing they are to engage with your brand in those places. If the fit seems good, double down on it. A great presence on one platform is better than an average one on five.

Also always be planning and observing 6/12 months ahead so you don't get too left behind taking the advice above. Have a roadmap, revise it continuously. Visualise where you want to be and work back from there.

6. Never underestimate the importance of your route to market

Just because a competitor is playing merrily in that space it doesn't mean it will be easy for you to get there as well, even with an awesome website! Timing is often crucial to success. If first mover advantage isn't yours you will have to budget more than you probably realise to get your market share. Or partner with a company that can reach your target audience. It's often harder and more expensive than you think. The web is littered with beautiful empty vessels.

You may be surprised to hear me say this – but ship the minimal amount you can to validate your idea, then market the hell out of it before coming back and perfecting.

7. Invest in design

My hot tip on the quickest and cheapest way to add value, increase conversion and build trust with users. Head to the design studio.

But it isn't just about quick fixes. Investing in your image and user experience builds an emotional bond with your audience and will help create a strong perception of your brand. You will be judged by the quality of your web presence. Every time we have made room to invest in our brand image, it has always paid back quickly.

8. Hosting & server optimisation / config is an outrageously under estimated skill

No one ever thanks you for good hosting. Those poor engineers working on harsh shift times are the unsung heroes of this industry. Coupled with the fact that everyone thinks hosting is super cheap and easy (thanks to cloud services) they have a really tough job!

We partner with hosting companies to provide services for our clients and have built some infrastructure on top of that as well. I am rung by hosting companies almost as much as recruiters but I will always be loyal to the companies that have dug us out of a hole when we needed it.

Qualified, responsive and experienced UK based support is invaluable, when the time comes. Popularity or just random bad luck will expose flaws in code and you need to know or partner with those that can help. Likewise optimising config can have vast impacts on performance. We are so fortunate to have developers here that get and respect the work of the system administrator.

9. Optimising your agency relationship

A tricky one and a subject of endless debate amongst agency owners. Naturally not all projects or clients are the same. Some work is straight forward, quick turnaround and delivery based or quite prescriptive. I suppose what I am considering here are the bigger projects that are not clearly defined from the outset.

Where is value really derived from? To some it is from how many hours under what time pressure something can get done. Yes, there is skill in that but rarely great work. Others are looking for experience doing it before so you can roll out carbon copies again. Valuable, but are any two projects really that similar?

I believe that real value comes from a healthy mix of collaboration, iteration and transparency. That all sounds great but handled in the wrong way it is easily a road to hell (paved by good intentions).

No one wants to work with a client sat next to them all day or phoning for hourly updates. Likewise iteration doesn't mean, “I'll know it when I see it.” and if you like transparency you need to be be prepared for warts and all. It is easy to freak out if people are off sick and the schedule needs rejigging or if a server outage means the developers have to suddenly refocus. Just remember this is nothing new for us.

Agency life is hectic, dependent on numerous uncontrollable factors and often requires heroic efforts behind the scenes. Typically we keep these from clients but actually quite a lot of unnecessary work goes into that facade.

The most successful projects over the last couple of years have been with those we've worked with before and built up an open dialogue and trust with. A big added bonus is when they also have experience of delivering internet projects. Crucially they also know their own industry intimately.

By return we've let these clients come deeper into the agency and build relationships with all members of the team. People who are not typically client facing but are at the coal face of making the product. They've helped us build a greater knowledge of their business and objectives within our staff and taken more time to bring us along the journey. The results have all been outstanding and have more than delivered on their targets.

If you want to see an example of such a product check out our write up of the recent Radioplayer tablet project.

10. Simple should be the primary goal of all projects

The initial brief and brainstorm produces a flurry of features. It is normal, every solution is more complex than it needs to be in it's first iteration. It is because you do not know the project well enough yet to say for sure what is critical and what isn't.

It might be radical but I think an objective should be to take that initial set of requirements and then force yourself to only go live with half of them, maybe two thirds at a push. This isn't a tactic to screw clients out of pocket. I think proper time and consideration needs to be given to each one. But over iterations you will begin to see unnecessary requirements cloud what is actually important. Spending more time on a few things will produce better results on features that stand proud.

So how do I know this? Our own products. Every. Single. Product. I normally write up the first scope and it is invariably too verbose and complex. But it is a good stab at version one. As I focus more and work closely with the team requirements naturally drop off. We start to get to the nub of the problem and really try and solve it.

Try slimming down your scope during the project, not adding more to it, which is often the case. Make a design principle out of “What can we remove?” and then be unrelenting in getting what really matters right.

I hope this incite has been useful. It's an honest assessment of how I see digital project improving but ultimately it is just my opinion.

So finally I would like to take a moment to thank my partners, staff, clients, industry folk, friends and family who've helped, taught, encouraged, listened, advised, laboured and laughed with me over the past 10 years. Here's to many more.