Priority Deathmatch: Build vs Optimise

11 months ago - 8 min read

A few weeks ago we held a Conversion Optimisation workshop along with MindArc, Bronto and zipMoney. We organised the workshop because we were seeing companies repeatedly make avoidable mistakes, each of which inevitably stalled growth.

Below is the video from my presentation, which discussed the merits of rebuilding your new website from scratch versus simply optimising it.

But before you watch it, I’d like to give a very special mention to Tom Howell from Synapsis Creative who did an absolutely incredible job putting my presentation together. If you ever need any presentation design, I highly recommend reaching out to Tom – he’s quite brilliant.

And now, on with the show!


If you can’t watch the video for any reason, below is a written version of my talk:


To Build, Or Not to Build?

Honestly, this must be one of the most common questions in the digital space: “Do I build something new from scratch, or do I tweak and optimise what I already have?” Now just to clarify, when I talk about “optimisation” I’m referring to A/B testing and the implementation of smaller design and usability modifications.

The build versus optimise dilemma spans every industry and every vertical. Nobody is safe from the horror of having to make this decision. Yet as far as I’m concerned, it’s fairly straightforward, because the answer is always data-driven.

Ideally, however, you should be doing both – and that’s the real truth.

If the data tells you that you don’t need to rebuild, then don’t. Save your hard-earned money. Put together an A/B testing and optimisation plan and keep tinkering with what you already have. It’s an iterative process, but it will deliver results if you continue to use data and commit to the process.

On the flipside, if the data says that you need an overhaul and to start from scratch, then you should definitely do that. And once you’ve built a new website, then develop an A/B testing and optimisation plan, because there’s no such thing as a perfect website and there’s always, always, always room for improvement – no matter how good anybody thinks it is.

Either way, the key is to use data to drive this decision, and I’m going to show you exactly what type of data you should be looking at, what tools to use and what the process looks like.


Pros & Cons

As a starting point, here are some of the major pros and cons to building and optimising your website.

Each of you will find that some of the pros and cons I mention resonate more strongly than others. Obviously they’re weighted differently for each of you depending on your pain points and the state of your current ecosystem.

The obvious pro of optimisation is that by testing your way into new designs and usability, you already have an idea of what kind of uplift you can expect post-implementation. This also makes building a business case significantly easier because it’s ROI-driven.

The obvious con of optimisation is that it takes much longer than a website redesign. It’s an iterative process that requires patience, which isn’t a strength for many people.

On the other hand, the obvious pro of rebuilding is that you’re given the opportunity to completely re-jig your setup. It opens the doors for new platforms, new tools, new integrations, new functionality. It’s a complete do-over, so you can wipe the slate clean of the many issues you’re currently facing and set things up as they should be. The major downside is that you have no real guarantee that your blood, sweat and tears will deliver any kind of meaningful result.

Launching a new website can be risky business, as new website builds have an enormous amount of considerations. Before you’ve even started on the wireframes, you have to select an agency, finalise budgets and timelines, manage stakeholders, select integrations, build the site structure, start thinking about content and make sure your brand guidelines are up-to-date.

And in all of this chaos the customer usually gets forgotten about or, even worse, ignored.


Using Data to Drive Decision-Making

Generally speaking designs are based on a combination of best practices, previous experience, personal preference and gut feeling. Sometimes it just comes down to the highest paid person’s opinion, which I’m sure everyone here has experienced at some stage. And there’s a certain irony behind all of this…

Websites get redesigned to provide a better experience for consumers, yet almost nobody stops to actually ask consumers what it is they’re looking for. What do they like? What do they not like? What are they expecting of you and your website?

Answering these questions is what makes the decision to either optimise or build from scratch such a consistently easy one for us – because we start with the consumer first and work backwards from there.

Below are some of the components of a behavioural research report, covering both qualitative and quantitative research methods:

  • Google Analytics analysis
  • Heat, scroll & click maps
  • Mouse movement tracking
  • Session recordings
  • Form tracking
  • Live exit surveys
  • Customer surveys
  • User testing

Now regardless of whether you end up rebuilding your website or going down the optimisation route, this has to be your starting point. If you’re not starting with data, you’re definitely ending with chaos. When the stakes are this high, you want to be making decisions based on more than “gut feel”.

We recommend you combine all of these data sources to get a complete picture of who your customer actually is (beyond basic demographic personas). What are their behavioural patterns? How do they perceive you in relation to your competitors? What do they see as your strengths and weaknesses? What usability and functionality features do they find valuable?

Once you know all of this, you simply cross-reference this against the realities and limitations within your business to determine what would be the most effective strategy to address both the needs of consumers and the business. But either way, you’ll have a strategic roadmap for optimisation – whether it be via a rebuild or an A/B testing program.

I find it genuinely strange that despite companies investing countless months and tens – if not hundreds – of thousands of dollars into a complete overhaul, they’re unwilling to invest a fraction of that into behavioural research beforehand to protect their investment.

It’s absolutely critical that you invest in this type of data analysis. It should be the backbone behind every decision that gets made internally. You’ll stop having those annoying opinion-based disputes over the strategic direction of the business, and instead develop a “data-first” approach when it comes to digital strategy.


Don’t Waste Another Minute

So here are your key action items:

  1. Install one of either Hotjar or Lucky Orange, as well as Crazy Egg, Formisimo and either Optimizely or VWO. These are the tools you’ll be using to gather your data, as well as A/B test your various hypothesis.
  2. Set all of the tools up properly and let them run for a few weeks and collect data. This will take you less than an hour to do, so get it done today. Once it’s done, you don’t have to revisit it for a few weeks.
  3. Set aside time every 4-6 weeks to analyse the data and identify key behavioural patterns and opportunities for improvement. Match this up with your own list of actionables and see how much crossover there is. From there, you can make an educated decision on whether you should be building a new website or building a testing strategy.

Whichever strategy you choose, it should be the one that addresses consumer needs as much as possible. At the end of the day we’re doing all of this for the consumer. It is their voice and their opinion that actually makes the biggest difference in your strategy, and that’s what should be driving your thinking – not the highest paid person’s opinion.