How to Avoid Costly Mistakes and Build the Perfect Website

How to Avoid Costly Mistakes and Build the Perfect Website
  • Building a great website is not about splashing an insane amount of money at your project.
  • You need to take care of 3 things: style, functionality and ongoing costs.
  • Clean minimal design works, focus on a functionality spec and calculate your ongoing costs.

Starting or about to build a website?

STOP. Read this first.

The long and short of it is, most businesses get this wrong even after the third try. You might argue that everyone else is building subpar websites and all you need to do is to sit a little above average. Well, if you believe in that, there’s no point reading this, because we are going to challenge the average and lay out a framework for building great websites. Do skip ahead three paragraphs to the framework if you’re short on time.

I was 18 years old when I first got into websites in 1995. Computer was still a mystery for most people, but I had already been programming for over a decade…talk about an early adopter. It was a freezing cold winter day and I was stuck in the computer lab, reluctant to walk into my next lecture.

After coding my first website – there were no visual editors back then and all you had was text and a black screen – I remember that first wow, or what would become the ‘www’ moment, when I hit publish and pointed a mosaic browser at my IP address. I realised that anyone in the world could see this, and the only thing that’s left is to build traffic. I added a visible counter to the nav bar (anyone remember those?), and I was officially a digital marketer.

Fast forward 5 years to the dot-com boom. I was contracted to Saatchi, an advertising agency in London that was brain-sucking in anyone who knew about websites. Hence, a handful of us from Melbourne landed gigs there over our uni break.

This during the crazy time, and we helped build a ton of sites. The contract web developers and programmers were building walls out of energy drinks and buying Ferraris as sites like MG Cars had a price tag of 5 to 10 million pounds. From what I heard, some VC firms in the UK were basically throwing money at web startups for 10 or 100 times the valuations overnight, all built on the hype of the internet bubble.

While I got the company line of 10 million pounds (that’s only a few hundred extra cars), I also knew the amount of work going into those sites that could have been done for a fraction of the cost. So, I was soon back in Melbourne, with a handful of UK clients who were delighted to pay Aussie rates. Plus, I could finish my postgraduate degree while designing and coding websites at night for the likes of Toni&Guy.

Today, where does it sit? What should you pay and how do you get the best out of your website build? The ever-changing rules and choices is a minefield for the uninitiated. Worse, cowboys who don’t understand the complexity are selling you into builds that are fibro-housing pitched as mansions. And this is everyone from DIYers to enterprise CMS companies.

Based on my experience over the past two decades, a sales cycle is always influenced by the client saying “I read this in a blog post”, or “my previous developer said this about that CMS”, which are the exact mistakes that would get you yet another suboptimal website.

Therefore, this is the triple bottom line I’d advise to anyone before we get into a sales cycle— what your site looks like, what it does and how well it works. I categorised these into style, functionality and ongoing costs.

Step 1: Style first

It may be the easiest to get right but challenging at the same time because of aesthetic subjectivity. If you’re the type of person that dresses up to go to the corner shop for milk, you probably have a good idea of how your website should look like. If not, get help from a web designer.

First things first, you must have a good, modern and clean logo. If not, everything below the nav will look ‘off’. It’s a make or break conversation we have with some clients, and often loses us work. You do not need to spend a fortune on this; just dropping your tagline will get you there. Your website should be all about making it easier for the user—every microsecond counts for user experience (UX).

And if you don’t have a clean and clear design, your website is not competitive enough.

If you’re not a seasoned web designer or if you want to save some big bucks, just go with a template that is clean and simplify it. There are millions of websites that sell themes– find one that is clean, minimal and most importantly, responsive on mobile. Your product, photography and style have more influence on your visitors than an overdone site.

An absolute must for any discussion with your digital agency about design, is references. Find sites you like, argue it out, and get expectations right at the start even when you’ve already got a quote. But if you start talking about where a button goes, how a hover works, how the CMS handles or a myriad of other tech distractions, you’re going to stuff up your build, blow out timelines and land around average at best. Don’t get sucked in. Focus on getting the overall design and style right first, and then come back to the rest.

Step 2: Functionality

This one is harder but driven a lot more by logic. There are four types of websites ‘bases’, each with their own tips and tricks. I could easily write a book about them, but here are the most important things to look out for.

1. Static website – This is a brochure website that contains information and nothing else. You can use a CMS like WordPress as it will give you longevity. If built well, you can update the site without going back to the web developer or digital agency. There may also be zero ongoing external costs as these sites should last 1 to 5 years, unless more functions must be added.

2. CMS – Content Management Systems are ever evolving and there are hundreds of choices, most of them with far too narrow a usage to consider. For open source, we recommend WordPress— as long as it is maintained up to date, you can always switch agencies or developers if needed. For enterprise platforms, we recommend Adobe or Drupal 8 for their availability of internal and external resources. A good rule of thumb is to look at what you will be spending on the site; choose an enterprise platform if you’re spending over $500K and opt for open source for anything under that amount.

3. eCommerce sites – Use Shopify or WooCommerce for small proof of concepts, rebuild in BigCommerce as a step towards enterprise, and Magento 2 for anything close to enterprise. Don’t look at Magento if your 24-month investment is under $300K, unless it is used out of the box, configured only (including the design template) and you’ve allowed for maintenance, your minimum spend will still be circa $100k.

4. For those thinking about building a CRM as your base, definitely keep HubSpot in the mix. Look at open frameworks and custom builds if your budget is over 500K, since website maintenance may be far less than licensing costs (plus, you will own the IP)! Most CRMs come out of the box with hundreds of features that have to be removed for them to work optimally and without additional staff resources for training and support.

What about special features on your site?

Everyone has specific requests. This bucket includes any platform that offers additional features outside of the key configurable parts of the system. This could be anything from higher than normal traffic usage and content requirements through to drop shipping global logistics or event/ticketing sales. You can either look to bolt on something configurable that already has a robust roadmap of successful builds with your core platform or build it yourself.

There is no hard and fast rule here, but as you step up from simple to complex, you should consider the benefits of a clear functional specification. The devil is always in the details. It often sounds simple to bolt on a ticketing, email, social or event system (the list is endless), but anything you synchronise in would require constant attention and ongoing costs.

There is virtually no set and forget site anymore. This is why I recommend you get a clear, functional specification that tackles what you want and how to achieve them before you let anyone start coding. This should include a layman’s summary and image references to every function. It’s costly, but a stitch in time saves nine.

Sign off a clear functional specification or bail out at this point. Do this in detail, so you know what you will be getting in advance and what the ongoing costs will look like. It is very easy to be swept up in a sales pitch and be dazzled by features, neither of them represents the complexity of the functionality you are trying to achieve.

I’ll give you two examples. Once we had to convince a client to bail out on a national learning portal (LMS) project, as their expectations of what they wanted was around 5 times of their price expectation. Another time was when a client signed off on an event and ticketing platform, but later changed their mind and ended up complicating the build to the point of financial failure. We eventually helped simplify the project back to its initial recommendation, configurable out of the box. This is like one of those 80/20 ideas – 80% of the cost can come from 20% of the requirements.

Aside from functionality, site speed (scored by online tests like Google) is another essential item to tick off. We measure in milliseconds what people do and don’t do on sites, and if you have a 1 second latency, you lose a huge chunk of engagement. This ties functionality back to user experience.

Step 3: Ongoing costs

No commercial website is without capital expenditure (capex) and ongoing costs. Capex is easy as you can get a quote from digital agencies, web designers and developers. Ongoing cost, however, doesn’t consider your resources, time and effort, and other minefields in costed maintenance.

Websites today are serious pieces of software. All software must be maintained and monitored. This will solve simple problems like site errors, speed deterioration, or being hacked. Staying ahead is like getting your car serviced. If not maintained properly, you’re going to be stuck on this side of the… wait for it… super-highway! And frantically calling your last agency for help.

Lastly, don’t believe your ‘work friends’. Getting sold on something because you’re looking to hang out means your website could land well below average. On what is becoming a huge part of a business success, this could even get you fired since boards and business heads are getting more digital savvy. Imagine them picking up a financial article and ask, “Why are we still spending X on our website?”

This is of course merely scratching the surface, and the variation we see in our clients’ understanding is zero to hero. But to build a great site, remember at least 3 key things – use references to discuss expectations, simplify where possible, and figure out the costs for the next 2-3 years. If your site can last 5 years and still deliver strong business outcomes, you’ve done exceptionally well.

Michael Simonetti

Founder and Director at

Michael is the Director of AndMine, a digital agency based in Melbourne, Sydney, and Asia. He has degrees in Computer Science and Software Engineering from Melbourne University. He was hired as a postgraduate lecturer and curriculum writer for the Marketing Faculty at RMIT. He has also been featured in articles and stories for; Fairfax, The Sydney Morning Herald, ABC, Channel 10, The Age, to name a few.

Comments (1)
Steven Dixon

Steven Dixon

It really happens, it is not necessary to spend a large amount to create a site. You just need to highlight key aspects and focus on them.