A typical web development agency is not highly scalable. Even at capacity, revenue is still linear compared with how much the project is costing to deliver. This is a property of most bespoke service-based businesses, as they involve direct customer interaction and a unique set of tasks for each project.
Recently, I discussed some of the attributes of our profession with fellow web developer and CUBE member Jonathan Lister. One thing we agree on is that web design and development is a craft and will remain that way for some time [1].
The problem with crafts is that they are hard to scale. At WeMakeWebsites, that doesn’t stop us trying though.
When scaling a task, the underlying requirement is to reduce the knowledge and effort required to complete it whilst maintaining the same quality of output. For example, we use 37signals Highrise so we can keep track of the organisations communication without everyone having to communicate a summary of his or her own communications to everybody else through some other method. We use deployment scripts so that each website release can be run with one command making the process easy and relatively error-free [2]. The extreme alternative to this would be to type out every command for each release, which would require a) knowledge and b) effort from every individual required to perform the task.
A good portion of management doctrine over the past 200 years has focused on how to deskill labour, standardise it, and then make it repeatable. The ultimate aim is to replace skilled labour with unskilled labour, which is more abundant and hence cheaper. The background to this movement was the industrial revolution, which focused on how to increase the output and decrease the cost of manufacturing tangible goods.
This was the age of “the average product for the average consumer”, as Seth Godin describes it, and it was a race to the bottom. In recent years, we’ve become used to cheap food, furniture and many other commodities. Indeed, it is now often cheaper to buy an item new than to have it repaired (which would require disproportionate labour cost).
In the post-industrial society however, new jobs have been created that require vocational skill that cannot be standardized so easily. In particular, there are relatively new specialist occupations such as marketing, product design, financial engineering and software engineering [3]. Whilst the last two are questionable in their claims of engineering discipline, they are nevertheless developing professions that have created significant value in their respective industries [4].
In a service-based business model, I think attempting to make the service scalable is one of two modes of successful business model. Nassim Nicholas Taleb once wrote that he’d never be a dentist. Not because of all the grotty tooth stuff and a lifetime of staring into gaping traps, but because “At a given time in the market, the most successful traders are likely to be those that are best fit to the latest cycle. This does not happen too often with dentists or pianists because these professions are more immune to randomness.” He elaborates on this point further by describing the winner-takes-nearly-all dynamic of highly scalable industries, where the main winner is a candidate that fits the market cycle and then undergoes some favourable exposure to external forces.
Many online companies with household names are in the scalable group: Google, Facebook, YouTube. Outside of information technology, this dynamic still applies: hedge funds, fast food franchises (the parent organisations, not the restaurants themselves) and footballers are all examples of where the winners in an industry win BIG – this is because there is a non-linear relationship between skill or quality, and reward. That’s not to say that Facebook or Thierry Henri aren’t any better than their competition. It just means that the reward of their success is disproportionate because of the dynamics of their industry. The difference between the financial success of Facebook and its nearest competitor is significant; the difference between Facebook and its 10th biggest competitor is immense.
As long as web design and development remains a craft, the highly scalable route to success will achieve limited results. Nevertheless, it might be possible to standardise the web development process so much that it can be outsourced to relatively low-skilled software engineers. I think the output quality of this approach would be limited, and would involve reducing the role of a ‘good developer’ to someone that can pick apart each project in to chunks based on the development skill level required for each, and then outsource each component to appropriately skilled engineers. So the answer to the original question is No, I can’t destroy the job I love. For two reasons: I don’t want to and, currently, it’s not possible to do it, or more specifically, to do it well.
I suggest the most effective way to be more scalable as web agency, is to adopt the franchise model and create a repeatable process and excellent QA methods. An agency could then expand internationally, setting up firms, hiring people and implementing well-evolved standard processes, before moving on. This would work for the same reason that, whilst restaurants are not usually highly scalable businesses, restaurant franchises are.
Aside from scalability, I believe there is another mode of strong business model for service-based companies. It is to create a service that is scarce. How would this work for a web agency? There are over 300 million results in Google for web design [5]. That’s not an industry with scarcity. Except it is – why? Because the quality of the services out there is questionable. Or average. Quite a few are good. Then what is scarce? Premium web designers and developers that know how to understand and meet clients’ requirements are scarce. I won’t go in to detail on the method here, but the principle is sound; work on your reputation and you will be rewarded.
Service businesses are valued mainly on their intangible assets, and reputation absolutely has to be one of the important assets alongside knowledge and experience [6]. I believe the three pillars of knowledge, experience and reputation provide a stable platform for a successful web agency.
Our business is young but this has been a kind of manifesto. We believe that by offering a premium service we can exceed our clients’ expectations and build a portfolio that we’re proud of in the process. In future, we’d love to take our business elsewhere by setting up agencies in other locations that can benefit from everything we’re learning right now. Let’s see what happens.
CUBER, Alex O’Byrne is co-founder of and lead developer at WeMakeWebsites ( http://wemakewebsites.info ), a London-based web and branding agency.
Hi Alex,
I agree with you that scaling up a repeated pattern of work is not really possible. What is important is to take yourself out of the process, which as you say, is very hard, because a web agency is based on the quality of the work produced by its members – standards which are exemplified by the people who started the agency.
A couple of ideas then (not connected to each other particularly):
1) find people who are better than you but lack the desire to run an agency; make them more successful through being a part of a bigger group – a bigger group has the potential to have a reputation much grander than that of an individual
2) make products and extract a stake once they’re successful (probably easier said than done)
3) don’t focus on growing to be a Google, focus on getting as big a group of people as possible to have “enough”
J.
Another couple of thoughts:
Price by value, not by time-consumed. If you re-use effort you’ve put into a previous project, perhaps by targeting a niche (like I know you’re doing with Hedge Funds), your incremental cost of doing a project goes down.
Build higher-value things: make sure your clients can edit content without having to come back to you; add in interesting features that make them stand out; pitch for strategic work not just piece-work of this-and-that bit of web design. Through this, you’re aiming not to commoditize the function of the web designer, you’re aiming to get more out of each project because your relationship with your client develops.
J.
Totally agree with you on those three points. Point 2 could be built in to a whole strategy (i.e. a pure model of picking projects based on our belief in their future profitability, and then build them in return for a return of future revenue). This approach would add more scalability to the agency model.
I guess your second comment is the jist what I’m getting at – rather than commoditize the role, we’re looking to enrich it.