The top ten rules of renovating

I had an interesting read this weekend.

I’m in the middle of home renovations, and came across a great article from Readers Digest on the top ten rules for renovating a home. As I was reading it, it quickly became blindingly obvious to me that the same rules apply to renovating your website, or any other piece of software.

So here, for your enjoyment, are the top ten rules of renovating, applied to UX projects.


1. Factor in the budget, prepare for blowouts

Home renovations are unpredictable, and blowouts often occur. The unforeseen happens, and despite the best laid plans, most builders will factor in between 10-20% to cover these unknowns.

The same is true in UX projects. Our approach is fixed price and that does help a huge amount – but often we’ll hit similar issues. On a renovation project you may encounter a big problem you were unaware of – for example a crack in a load-bearing wall that was hidden previously and now needs fixing before you can proceed. It’s a big problem, you can’t ignore it, and you hadn’t budgeted for it since you didn’t know it existed.

UX projects can encounter the same. A part of the design that was assumed to be fine slips into view during testing, and a previously unsuspected problem surfaces like a whale. All of a sudden you have an entirely new problem to solve that was never even considered.

For the same reason as with a home renovation, we recommend budgeting for the unforeseen, and in similar numbers, 10-20% of your UX spend. If you have the capacity there to fix it, you won’t need to paper over the cracks and live in fear.


2. Choose a reputable dealer

I fell foul of this once, and it was painful. A kitchen that should have taken a week or two ended up taking nearly two months, and cost me double what it should have. I learned my lesson, and hopefully you’ll never have to learn yours the hard way.

Choosing a reputable UX supplier is just as key as choosing a reputable tradie for your renovation work. Past experience with similar projects, their customer list, and what their customers say about them are all key factors to review. You need to ensure that they know their stuff, have good people, have happy customers, and didn’t just pop into existence last week.


3. Have a watertight contract

This is perhaps the only point that’s less relevant for UX work than for building – but it’s still an important one. In my experience (and I’ve worked in UX since 1997) the most common cause of confusion or conflict on projects – both mine and competitors – has been when a customer has thought they were getting something in the price, and the UX supplier has not seen it the same way.

Make sure that everything is explicitly laid out in your proposal/statement of work/contract. If you aren’t sure, ask. Again in my experience, most UX suppliers are quite ethical and aren’t trying to hit you with hidden charges or excluded elements – but what is obvious to one person is not always obvious to another. Check check check, as my father-in-law constantly tells me (usually right before I cut the wood too short).


4. Expect timeline stretches

Just as those hidden problems cause cost blowouts in a renovation project, they also cause timeline stretches. Extra days are taken up fixing those problems nobody knew about, rain hits on the roofing days, and days are lost.

The rain rarely has an impact on UX projects, but plenty of other delays can kick in. Key team members fall sick, and are unable to provide feedback. Prototypes aren’t ready  causing delays in testing, materials coming in from overseas (prototypes, design work) get stuck at customs. In my time I’ve seen all of these and more. One project was delayed by several weeks, as a key project member passed away, a completely unforeseeable event that nearly destroyed the other team members.

Most projects we’re engaged on involve an extremely tight deadline, with little or no space for contingency. Once approved we’re asked to start almost immediately, to work non-stop and to deliver the day before it’s due. Most of the time we manage to get it all done on time, in fact we’ve made tools that help us to complete test reports faster than most others can – but we can’t always overcome the unforeseen. It’s always wise to allow some spare time.


5. Be flexible with solutions

It’s amazing how much a small decision can cost. We were discussing some work with a builder recently and he raised a recent project he’d been engaged on. In that case the owner had wanted two rooms of similar size laid out in a certain fashion. The builder told the owner they could save money flipping the rooms – which would functionally make no difference to the home. The owner however was fixed on the layout, so they proceeded. In the end, he estimated that they spent nearly $10,000 more  than they needed to.

The same is true on UX projects. In terms of solutions and design approaches there are often shortcuts that functionally make little difference. Just as flipping two rooms can save thousands in pipework and cabling, opting for a slightly different UX approach can save thousands in effort and time coming up with new designs. Don’t be precious; sometimes 95% is good enough, if you’re looking to spend money wisely.


6. Expect surprises

When we fixed up a bathroom a few years back, we discovered that the previous owners had used a form of concrete to cover the walls. Where we had assumed we would spend half a day stripping the room ready for plastering, we ended up spending closer to a week. We had to buy a huge hammer drill, and it nearly killed us. Nobody would ever have guessed what lay under the paint until the job started.

The same is true on UX projects. Expert reviews are a great way to spot what is most likely going to go wrong in user testing, but it’s never 100%. Throughout my career I’ve constantly been surprised by the issues uncovered during testing; they almost always make sense once they come out, but sometimes you’re blind to them until they appear. Sometimes they are an easy fix, sometimes you’re left hammering and chiseling for a week – either way, expect the unexpected.


7. Don’t overspend

This is a bit of a no-brainer when it comes to renovation work, but it’s still surprising to hear the horror stories. It’s very easy to spend more on the renovation than you’ll recover in the value of the home, so watch the budget. Know up-front what you want to achieve, and spend wisely.

Again, this applies perfectly to UX projects. I once had a customer ask me to price up a usability test for 200 participants. My mouth fell open, and I explained that most user tests are between 5 and 15 people strong, for several good reasons. The customer was adamant however that 200 was the number. So we priced it up. Unsurprisingly, the project didn’t proceed.

UX work scales well, in most cases. Testing can be a few people, a day, two days, or three. Interviewing two or three people will help, if interviewing 50 is too expensive. There are almost always options for scaling so know up-front what you want to achieve, and what you have available to you. Most good UX agencies will help you to identify what you need and what it will cost, and then scale it down to fit if it needs to.


8. Visit the site often

If you’re watching the work progress, you can spot mistakes before they become costly. If you’re watching the stairs be put in upside down it’s simple to correct it, not quite so simple once the rooms are all built.

In the same way, regular checking of the design process is paramount. The sooner you find the mistakes, the cheaper it is to correct them; that’s the basic concept behind the entire User-Centered Design process.

A good UX supplier should offer a more or less transparent process. At every step of the way you should step in and view the progress, and check the design. Prototypes are easily placed online so that you can verify the direction and the progress regularly. And if the navigation goes in upside down, you’ll spot it instantly.


9. Book jobs in sequence

If you book work to fit into a sequence you streamline the process – as long as you allow for the unforeseen. Tradies are still on-site and available to fix/help with issues that might arise. For example the Electrician may turn up to do the wiring and point out that he needs access behind a wall. If the plasterer is still on-site cleaning up, he can help to provide it.

In the software/interface design world the same rules apply. We often work on projects where the next in line (usually the content and creative teams) are involved and looking over our shoulder during the research and design phases, and are booked to start once we complete. Most importantly however, they’re seeing our work as it progresses. When a problem creeps up we’re still involved, and can easily amend and compensate.


10. Budget for landscaping

It can be surprising how much landscaping can cost. We recently looked at a piece of land to build on, and a quick estimate of landscaping came in at nearly $30,000 – mostly for retaining walls and leveling. Sure it was a steep block, but the figure flawed me.  The point is, don’t discount what it costs to completely finish the job.

The UX process involves the same kinds of finishing. It creates the content structure, the flows, the architecture of the experience. It creates the controls, the layouts, the functions. But to finish the job you need the content, you need creative, sometimes a good copywriter. It all needs to be implemented too.

The vast majority of our projects get the finish they need, but we’ve certainly seen some where the ‘retaining walls’ have been rushed into place or put off entirely. The result is always a let down, and usually leads to failure down the line. The message is simple; don’t skimp on the work that’s needed to finish the job.



Oh, and my own personal number 11 – always keep a good relationship with your builder and tradies, that way they’re always happy to work over the weekend and come round at night to fix a leak!