14 Nov 2017
AKA their site conversion rate has run out of fuel.
What you can expect from this post:
First things first. If you are not sure what ‘conversion rate’ means, then take a look at this guide from the NN Group. It pretty much explains everything you need to know. For the sake of keeping it simple though, we can summarise this as follows:
The conversion rate is the percentage of users who end up buying something from your site.
An e-commerce site is visited by 1,000 people during the month of October. If you add up your sales orders and arrive at 100, then your conversion rate is 10%.
Let’s take that a stage further. If your average order value is £50, then you made £5,000 during October.
The art of conversion rate optimisation is, therefore, to turn that £5,000 into more money - without spending or generating more traffic.
If you take my simple example and give it context, a site generating 10,000 orders per month stands to gain huge increases in revenue if it can improve site conversion rates.
But where do you start?
This is pretty obvious but not everybody is that savvy with their own conversion rates or where to find them in analytics.
If you fall into this category then you need to head into your analytics data and start to identify how you are currently doing. We will look at that in a second.
It's worth pointing out that the worldwide average conversion rates for e-commerce stores is around 3%. So don't stress immediately if you see a low number.
I would suggest that you look over a significant period of time to get a true picture of where you are. So as a suggestion, one year is the minimum I would recommend. Any shorter, you may be seeing seasonal drops or something relating to a specific site change or shift in the product range. So whilst that may be a cause for jubilation (or concern) it is too short a period to assess whether you are in a growth or stagnation period.
So the initial view in analytics is fantastic and as it says on the tin, will give you an overview of what is going in. In here, you can assess some really key metrics including site revenue, and most critically, your site conversion rate. If I had to choose the one report I could keep in GA, this would be it. The site conversion rate is the ultimate yardstick and health check on any store.
An improving line demonstrates you are increasing profit and growing your business. A flat or declining line is your first indicator that something is wrong.
Here is the report in GA:
In the above example, you can see the green line indicating where the Adwords campaign for this client was overhauled. You can see an improvement in conversion rate as a result of that work.
Whilst the line only looks marginal, we are talking about a full percentage point improvement on a store that generates over 2.5 million pounds annually. This is, therefore, a lot of extra money. In fact, not money, profit.
Conversely, if your graph shows a relatively ‘flat’ line or worse, a declining line, then you could be looking at a stagnating business.
At this stage don’t panic. It could be anything from a competitor price cut to a site bug or payment gateway tracking issue. It could be a poor quality traffic increase impacting conversion rates and making the numbers look bad. Maybe you hired an SEO agency and they have improved traffic markedly but not impacted sales.
In short, it could be anything or nothing.
This report is your early warning system that something somewhere could be wrong.
In my experience, conversion distress comes down to three main areas:
It remains a fact that a large proportion of traffic comes from Google. You have other channels of course but generally speaking, Google converts better than most other channels apart from direct and maybe email.
Let’s look at both organic and paid search and how you can evaluate performance.
Organic search traffic from the big G is an absolute must for any healthy store. The attraction is simple. It’s free and often long tail targeted i.e. product focused.
So any traffic you grab can result in more sales and with a higher profit margin than its sister Adwords.
With many stores, there is an initial growth period where product pages get indexed and traffic begins to build. At some stage however you will hit a ceiling and this dictates your next move.
So firstly let’s take a look at your organic profile from G. This can be viewed by going here:
Behaviour > Landing Pages > Organic
The last bit is a filter you can find here:
When you first look at this report, you will see a blue line covering the date range (top right) specified.
This is ok but lacks the ability to cross-reference performance.
You can either a) extend this over a longer period or b) compare against the corresponding period last year.
The above graph has a 2017 line (blue) and 2016 line (orange) so you can quickly see if you are going forwards, backwards or just stagnating. The top arrow is how you filter for organic only traffic. Generally, a comparison against last year is the easiest way to spot performance issues.
Its a great initial indicator as to how your organic strategy is going.
Unfortunately, organic search development often has no quick fix. It takes time and effort to build expertise and authority.
But not always. Sometimes you can establish ‘quick wins’ through a few technical improvements:
Ensure you are providing a fast UX to visitors – try running one of your key product pages through the Google Page Insights tool for example. Benchmark this against other sites in your industry. The faster your site runs the better. It will assist conversions and allow Googlebot to index more of your site. This is commonly overlooked but has obvious benefits.
Don’t have or move product content and reviews behind tabs. For desktop, that is a bad idea. Make sure your product descriptions are visible without having to click between tabs. If they are currently stuck behind tabs, move them.
Cover the Basics
Check the basics are covered – titles, descriptions & headings. Have you used this space effectively?
Check your mobile site is functional and is free from buggy interfaces and general quality issues. Is the content as you expect it to look? if not, what is missing?
Are you running over https yet? If not, then really the site should be by now.
Have you got lots of boilerplate content floating around? If so, maybe think about dropping that to a single page.
Parameters are an easy way to burn your crawl budget. Luckily Search Console gives a really nice interface to allow you to instruct Google what to do when they encounter URI's with parameters attached to them.
Beyond these quick fixes - we are looking at content and link development. Not a quick fix but committing to these areas will yield results long term.
I am yet to work on a successful store where the client doesn’t rely on Adwords to one degree or other. It’s a powerful platform and handled with care, a very profitable one.
However, I have also seen client accounts get out of control. As the business grows, so does the Adwords account and often it gets worked on my multiple people too. In short, it can unravel into a bit of a mess and depending on how well it was set up initially, actually, it can end up a big mess. To decision makers, wading through the AdWords interface isn’t an option.
You need to cut through this and see what is happening in reality:
This is a great report. Two clicks will give you a window into every Search Adwords Campaign performance:
The report can provide a top-level view that can quickly identify where you are struggling to break even or worse still, losing money. These campaigns need weeding and maybe even closing. In my experience, there is a story behind most underperforming ad groups and campaigns.
Now before we move on, a quick bit of maths here. The calculation for this report is as follows:
Ad revenue / Ad spend x 100 = ROAS
What we need to do therefore is establish performance categories for these campaigns. Here is how I do it:
1) Anything with a ROAS over 500% - mark in green. It's probably making money.
2) Anything with a ROAS between 300% - 499% mark in orange. Depending on margin, it could be a positive contributor.
3) Anything with ROAS lower than 300% - mark in red. It’s tanking and probably costing you money.
This has served me well over the years and really helped focused minds. All we do is take the above ROAS criteria and apply it to the data.
This requires you take a CSV export of the report which can be found at the top right of the GA screen. Make sure you have included a sensible date range - again 1 year is a solid period to use. Once you have this data, simply order by the ROAS column and either use some excel wizardry or manually colour each row.
Here is what it looks like once you have colour coded the rows:
How useful is that! No need to wade through Adwords. You can have this data in minutes really.
Once you have this format, you can ask for regular updates from your team:
The missing (and vital) bit of information in the above examples is margin.
ROAS is fine but without context it’s meaningless. Margin is the context.
You can factor this in by heading towards an ROI calculation. So let's assume that your campaign is in the red zone and generated £3,000 sales for a spend of £1,500. So its ROAS figure is 200%. At a margin of 30%, the campaign has actually cost £600. This is why it's essential to work on the red zone campaigns.
Now at this point, I can hear the agencies jumping up and down about assisted conversions. If you aren’t familiar with these, I wrote about them way back in 2012.
Never turn off Ad campaigns without checking if they are assisting sales in some other capacity.
But a loss is a loss and you need to start working through the lower end of the campaign ROAS numbers to identify why. This is potentially very complex, but some basics can be applied.
Things to look at:
When you have checked these areas, then you may suddenly have a lot of things to look at. Your store may have hit a glass ceiling and the question for you as the site owner is whether you have the in-house skills to address these areas.
If you are reading this and already deploy agency support, then, by all means, take your findings to a review meeting. Agencies need time to improve things, but have they taken their eye off the ball or hit a knowledge ceiling?
Hopefully, this may have inspired a few people to make a start and admit they need help to push sales forward. There are of course lots of more advanced things that can help, but let's keep it simple, to begin with.
Important: If you mainly run shopping campaigns, the process is exactly the same. Simply head to the Acquisition > Adwords > Shopping report in GA and follow the above steps.
Don’t shoot the messenger. Adwords is often blamed for ROAS issues when in fact, the campaign is helpless to things that have changed in your market. Namely, things your competitors have done….which you may have missed.
My top tip here is to look carefully at your product performance report. This is also located in the e-commerce section of analytics and can yield very useful data. Critically, you can query some of your better selling product lines using the search functionality.
Does anything stand out when you do this?
Let’s say you sell fire prevention equipment online. One of your key lines is your range of fire extinguishers. Now unless you are closely monitoring your competitors' prices, you may have missed something that could be hurting your sales. The product performance report is a great way to check into this.
All you need do is type in a keyword in the filter box:
If you want to be more specific than this, you can use the advanced filter. Maybe you want to check a specific size (for example you may sell mainly 5kg extinguishers) or you only want to check a specific range because that represents over 50% of line sales (for example ‘water’). The advanced filter will help you refine the search for just the products you want to review.
After doing this, if you see any drop in sales you may have identified an issue. This could be a competitor cutting prices or maybe your landing pages contain out of date information. If you do this across your major product lines, you will quickly identify problem products.
Common issues with competitors:
UX problems often go unchecked. You may be spending small fortunes on traffic development, yet UX is often given no budget or worse still thought. If you are spending thousands on ad clicks, yet are not UX focused, you are literally throwing money away.
This is a real mistake. Making improvements that help people convert more easily (aka UX) is an area where you can drive up conversion rates pretty consistently. Two scenarios usually exist:
A third scenario exists. This goes along the following lines:
3. We are conversion focused and 'think' our website is ok.
The third scenario is by FAR the most dangerous. The first two are pretty normal, UX doesn’t get the budget it deserves and that in many ways is ok. At least you know where you are.
UX is not about your opinion or your web designer’s opinion; it’s the people that use your site that matter, with a foundation in heuristic analysis a given foundation.
'I think' statements usually damage clients the most.
Sometimes people just get defensive and feel threatened by the possibility they are making errors or outsiders coming in who know more then they do and essentially saying they are making mistakes. That can damage internal kudos and create resistance.
Site owners need to push through this.
E-commerce UX & conversion reviews + testing are the only way to really address conversion/UX issues. However, you can also get some clues by having a good look in GA.
If you are plugged into the newish enhanced e-commerce analytics, then you should be able to see where people are falling out of the checkout process. This can provide valuable data that otherwise could be much harder to track down via a typical Goal Funnel setup.
If this isn’t set up yet, you won’t see these enhanced reports and will need some help getting the checkout labelled up correctly.
Once you get some data, this is really useful and gives a great visual clue as to where people may be struggling in your store.
Let’s assume that people are abandoning their carts at your delivery page. Why?
Have you thoroughly tested the checkout to identify any UX issues? Did you skip the cross-browser testing or maybe you are simply asking for too much information and the customer is bailing out? Maybe it’s a bit of functionality people are struggling with?
Again this is no more than an early weather warning. If you see an unusually high abandonment rate, then you will have to look into the issue and see what can be found. If you haven’t spent much time testing your checkout, then this should be the incentive you need to start doing so.
Here are three common areas that normally fail users. If you are looking for some ideas and don’t want or have the budget to get pro support, try these:
Usually has some issue or other. Commonly it doesn’t work under scrutiny and also can often not be being tracked in GA. Back to throwing cash away. Site search really helps users but designers can often hide it or make it hard to spot thinking that is ok. It isn’t. Ever. Test it thoroughly.
Often these are badly produced and don’t help users. When we test this functionality, we usually find that competitors have way better interfaces and/or our clients doesn’t work very well, especially on mobile.
Often images are hacked by CSS / Jscript and badly bloat category and product pages. Very few sites have switched to http/2 yet and in many cases, achieve very poor results under testing. Site speed in 2017 is a given, but often is badly overlooked by the build team – everyone has broadband right?
So having looked through all your data - how does it look? are you growing your store or stagnating? Hope you find some of this useful – if you would like to chat about your project in more detail let us know!
The article was written by Jon Colegate. Connect with Jon: