The beginners guide to conversion rate optimisation

With so much talk about making data-driven decisions and being super analytical, it’s easy to get a bit intimidated when it comes to conversion rate optimisation. You don’t need to know about A/B testing or event driven analytics software right now, let’s start with the basics.

Related: Get your website noticed

I’ve listed the top 36 common mistakes people make and why they hinder website conversion rates. To make it more fun, I’ve split them up into different sections and liken the whole process to chatting somebody up at a bar.

As for me, my name’s James Pursey, and I advise startups and website owners on how to improve their websites and optimise for conversion. I’ve worked with over 200 companies to date so have picked up on more than a few things that happen again and again and again.

Trust

Trust is the core element of any successful website. Typical trust indicators are so obvious but get neglected by such a large volume of websites. I like to compare web design to real life scenarios, and when it comes to building trust quickly and effectively, the best example is a guy chatting up a guy/girl at a bar, choose from one of the following two options:

a) Do you introduce yourself, put them at ease by showing an interest in them, build a rapport (trust) then see if there’s mutual ground?

b) Do you take their drink out of their hand, grab their coat for them and suggest you leave?

Don’t sell to me right off the bat, this isn’t Glengarry Glen Ross, and ABC doesn’t apply here. Ease me into whatever you do, it’s really not that hard but it’ll make the world of difference:

1) What do you actually do?

How many times have you looked at a website and had absolutely no clue who they are or what they do! I see it all the time. So #1 is perhaps the simplest thing… introduce yourself.

2) Build trust, quickly

Take advantage of anything you can to tell your users you’re a reputable site. Testimonials crush it for this sort of thing, put a face to a customer and have them say something nice about you, simples! Proof through things like client logos and press features work well too, it’s all instant recognition that you’re a real person/business!

3) How do I proceed?

This one blows my mind every single time. If you want me to do something, tell me what to do! That means CTA (call to action) buttons pointing me in the right direction. Not doing this is the internet equivalent of when somebody just says “yes” in response to “got the time mate?”

4) Make your English teacher proud

I’m incredibly dyslexic, but I know how to use a spell checking tool. Grammar is important but make sure you’re at least spelling things correctly, not only will your users love you for it, but search engines will too.

5) Supercalifragilisticexpialidocious

Don’t use long words to try and sound smart, it’s the first thing people that have no idea what they’re talking about do and… it makes you sound incredibly dumb!

6) “Share To Reveal”… No!

There is a massive difference between using an accordion style feature (like this content) and making somebody share your website on social media to access content. Put simply, if your product/service/content is good, people will share it, don’t be spammy and annoying.

7) No Reviews on Product Pages (e-com only)

I understand that you might be a small site with a handful of customers, but if you don’t even have the capability of someone reviewing your product I’m going to assume it’s because you’re scared of negative feedback which in turn means your product must not be to a certain standard. You need to at least give this option even if it’s blank, or you could be smart and pull in reviews from sites like Amazon just to help your users.

8) Will my credit card be cloned?

SSL certificates are a must for all e-commerce sites, they reassure users that their data is safe.

Copy

It should go without saying that the copy on your website is how you speak to your customers (see what I did there?). You wouldn’t want to come across in person as sloppy, unsure of yourself or too aggressive, so why do it on the web? There are no massive tricks to writing good copy for the web, it’s all fairly self-explanatory – know your audience, don’t write like a computer (unless that’s your audience) and don’t be a dick.

9) Swearing, it’s not big, and it’s not clever

This is industry dependent, for example, if you’re writing a piece preventing people doing stupid sh*t on their site, then fair enough – but DO NOT swear on product pages or service sales pages.

10) Weak/thin content

This is the single most annoying thing on a site. People confuse designers saying you shouldn’t have big blocks of text, with blog posts. If you’re writing an editorial piece, make damn sure it’s fairly long and engaging. A 200-word blog is a waste of most peoples time if you’re talking about a subject you’re claiming to be an authority on. Shoot for 500-1000 words, unless you’re writing a beast of a guide like this (circa 3,000 words).

11) It’s all lies!!!

Don’t tell porkies.

12) Do you even know what a superlative is? It’s this totally like awesome amazingly brilliant thing

‘Nuff said.

13) Big boring blocks of text

See how this post is put together? Would you believe it’s over 3,000 words long? If you have a large amount of content it’s crucial that you break it down to make it more visually appealing, that can be through bullet points, numbered lists, different font weights, images and click to expand style functionality like you see here. However, long copy should be kept for blogs, articles and white papers, it should not appear on your landing pages or anything like that. Yes… I’m aware this is one of the longest points on here.

14) Don’t do this!!!!!!!!!!!!!!!!!!!!!!!

I’m guilty on this on everything I send, even emails. My own faults aside, the more you use exclamation marks/points, the less they have any effect, and the more you sound like you’re trying too hard. It’s tough when you’re very passionate about the topic you’re writing about, but try to limit yourself to 1-3 per 500 words or so.

15) I’m sorry, who are you?

Testimonials are very important for the vast majority of websites. I mentioned earlier that you should use them to build trust, but make sure they’re genuine. A great testimonial should have a person name, a headshot/picture of some form and a positive message that isn’t over the top. If you miss any of those things, you run the risk of it seeming fabricated and having the opposite effect. Do not use stock images for head shots.

Design

If copy is how you speak to users, design is how you dress to impress. A really well-designed site will ooze confidence and look like it’s an established business even if it’s only just gone live. It’s important to get the balance right between simplicity and minimalism and something that gives enough information to enable the user to make a decision, get the balance right and you’re on the fast track to conversions.

16) Was it acceptable in the 80s?

If your site was built outside of the past five years, you almost certainly need a complete redesign. Retro may work in nightclubs, but it certainly doesn’t hold true here. Old sites are ugly, difficult to navigate and were created for an entirely different audience… web users move quickly!

17) Too much stimulation

An overly busy site is one of the first indicators of a badly designed website. The Americans have led the way here, with big elegant images and sometimes less than 50 words on a landing page. If your site only has one goal, strip out all the rubbish and just concentrate on the core purpose of what you’re doing. Busy sites are ugly, they put people off, and they tend to have a very high bounce rate.

18) Auto playing media

Landing on a site that instantly plays background music or video content is the single most frustrating thing in the world of web browsing. Users could be out in public, or in a quiet area with their sleeping kids, or any manner of other places where they don’t want unexpected voice overs or seedy background tunes playing, especially if they’re using a bunch of different tabs and don’t know where it’s coming from. When this happens, you find the culprit page, and you close it, without paying any attention to what the site does. When I see this, I want to punch my screen, so don’t do it.

19) Stock images

Stock images can work for a bunch of different scenarios, but they have to be contextually relevant and tasteful. If you put a stock photo on your team page, prepare to have a bunch of users not believing a word you say. You can normally tell when a photo is genuine for that purpose and when it is not, so regardless of what you may think, you’re not fooling anyone.

20) No custom 404 page

A 404 page is where the user goes when they navigate to a page that no longer exists, or never did exist. They’re normally a bit on the funny side (ours says you blew up the internet), then they try to help you find your way. An e-commerce site could use this to point people towards a product, or a blog can simply let you search what you want. Anyway, if you don’t have a custom 404 the user goes to a browser error page doesn’t exist page, meaning they don’t get back onto your main site and they invariably leave forever.

21) No About Us page

I’ve seen this crop up a lot, and I’ve told people off for not having one in the past, but I think it’s a little industry dependent. One thing you certainly do need is some form of explanation of who you are somewhere that people can easily see, that could be through an About Us, or it could even be on a side panel like I’ve used on this site. Bottom line, people like to deal with people, not machines.

User Experience

So now you’ve got your tone right, you’re dressed to impress, and it’s time to head to the dance floor and show off what you can do and how you do it. Have you seen Hitch with Will Smith? When he’s telling his client how to dance he tells him to keep it simple and stay in his comfort zone… UX is no different. There’s a great book about UX called Don’t Make Me Think by Steve Krug, the concept as you may have guessed, being that as soon as the user has to think of what to do next, you’ve lost them.

You get about 2 seconds between a user hitting your site and the user choosing whether to stick around or to bounce (leave without clicking anything). If a user clicks something and it doesn’t work, or it doesn’t do what they expect, or they can’t even find something to click… they’re not coming back.

Sites are very rarely faultless, though. I personally hate the accordion functionality of this post because it doesn’t push the screen back up to the top of the section, but I can’t find an alternative, and I can’t code a fix. I figure if you’re reading this point, it hasn’t annoyed you enough for you to bounce.

22) Umm… how do I use this website?

No surprise that this is in here. If you go to your own website and look at it through the eyes of somebody that has never seen it before, can you honestly say you’d know how to use it? If the answer is no, you probably need to gut it and start again. It’s far easier to design a website with specific conversion goals in mind than it is to tweak a poor user experience site into the same end goal.

23) Overly complicated mouse-over navigation

I see this all the time, cascaded mouse-over navigation is a terrible user experience and if your site isn’t mobile optimised it’ll be even worse on mobile/tablet devices. Keep things simple, it’s better to take people through a well-designed page laying out different options, than to funnel them through overly complex navigation functionality.

24) Exit intent

This can be done exceptionally well if you use something like Bounce Exchange, a great example being QuickSprout by Neil Patel, but if you code it yourself and use standard browser dialogue blocks, people will hate you.

25) Sooooo sloooooow

Load speed is important for SEO, why? Because it’s directly related to user experience, people hate waiting for a website to load just like they hate queuing tickets or an ice cream.

26) Horrible appearance on mobile & tablet devices

Ideally, you’ll have a fully responsive design, but if you don’t, that’s no reason just to let things lie. Meta viewport tags allow you to set the zoom level and position of a page, to minimise the annoyance of your users.

Conversion Paths

Tone tick, nice clothes tick, great dance moves tick, now it’s time to checkout, but what’s the goal? Do you want his/her number? Do you want to call a cab? Do you lean in at the door? You need to know what your goal is before you try to realise it. I promise you that’s my last cheesy analogy, but this section is all about knowing what you want and how you plan on getting it.

27) Do you know what you’re doing?

I see this all the time, it’s like a website that has an identity crisis. When designing a site the first thing you need to do is decide on its purpose, is it to get users to a contact page, to sell a product, or even to get someone to call you? The whole site should then be built with the view to triggering this conversion. I see a lot of sites that have 50 different things the user can do, including the one thing that makes them money, but there’s no direction or purpose or funnel. If you’re looking to improve your site and convert more, then you need to know what it is you’re trying to improve.

28) Not enough direction

This is related to the above point. Once you know what your conversion is, you need to craft a path to it! That means pushing me back onto the funnel at every opportunity. A blog, for example, isn’t just to let people read stuff and leave, it’s an acquisition tactic so use your side panels to push people back on route.

29) Don’t throw me off

Continuing the theme and working from the point above, don’t link out to your blog from the checkout cart of your site, it works both ways. Push users onto the conversion path but don’t kick them off it before they convert.

30) Don’t push too hard

Nobody likes a pushy salesman, if your product/service is actually good, then it’ll sell itself just by you explaining its benefits. If your product/service isn’t that good, stop reading this and either make it better or think of something else.

31) Not testing things

If you’re reading this blog, I can assume it’s because you want your site or your client’s sites to convert better. You can have the best plan in the world and execute it to perfection and then have a 0.00001% conversion metric… when that happens don’t come to my address and beat me up, just try different things, test the outcomes and then repeat, Rome wasn’t built in a day.

Search Engine Optimisation

You may think SEO isn’t a direct reason that users don’t convert on your site, and you’d be right. The specific art of optimising a website for search engines won’t impact your conversions, but some technical website elements will make a difference (and benefit your SEO in the process).

32) Spammy descriptions & titles

When you Google something the two elements of each site that you see are the Title and the Description, the latter being the longer bit underneath the title (obviously). The title is super important for SEO, and you should have your target keyword for that page in the title, preferably near the start. But beyond that, it’s all about the user, not the search engine – meta descriptions have little to no income on your ranking, they’re designed so that you can convince the user to click on your link rather than the others. In summary, write your copy for people but with search engines in mind… not the other way round, and definitely don’t ever keyword stuff these things.

33) Over complicated URLs

Having a URL like domain.com/main-category/sub-category/random-attribute-filter/attribute-filter-2/attribute-filter-3/ … /product-parent/product-name is really ugly, and that’s exactly how a user will see it. Keep things simple, if you’re running an e-commerce site go with domain.com/category/sub-category/product-name and only if there are enough products to actually warrant a sub-category. If you’re running a blog then maybe segment by content type (list posts, interviews, etc.) but again, only if you have enough content to warrant it.

34) Poor internal links

From an SEO perspective, internal links are crucial to allow crawlers to get into the deepest layer of your site. They also increase time on site, by helping your users find other interesting content on your site that’s relevant to what they’re looking for. Having poor internal links will decrease a user’s time on site, and it doesn’t take a genius to point out that it will also be detrimental to your conversion rate.

35) Rubbish keyword consistency

This relates back to trust issues. If you’re claiming to be running a site about dogs, and the word “dog” doesn’t appear in the title, URL, description or content… the chances are that your content isn’t really about dogs at all. Keeping your target keywords for a page appearing consistently throughout all on page elements, will not only help your SEO dramatically, but it’ll also reinforce the purpose of each page to your users.

36) Broken links

Not much of an explanation needed here, linking to an internal page that no longer works and fires a 404 error screams that you’re not professional. You should run a full content & link audit on your site every six months or so.

 

Key Takeaways

  • CRO stands for conversion rate optimisation, it’s got nothing to do with birds.
  • If you’re not super data driven, don’t freak out – start with the basics and use this as a bible.
  • Just like flirting with somebody at a bar, these things get better with practice.
  • Feel free to change your LinkedIn to “Growth Hacker” “Conversion Rate Expert” “Optimising Optimiser”.