Is it time for CRO (Conversion Rate Optimisation) to GO?
Improving conversion is appealing and necessary to drive success but how many look beyond the initially defined goal?
If you work in eCommerce, especially retail, when you embark on a CRO (Conversion Rate Optimisation) programme, what KPIs (Key Performance Indicators) do you use to monitor and analyse to declare success?
If you are working higher up in the sales funnel and are testing different copy within your PPC (Pay Per Click) account, naturally you can monitor impression levels and CTR (Click Through Rate) to check that the copy is doing its job. That it is getting the attention of the user and persuading them to click the advert to come to your site. If your advert sees a high CTR but you do not see an uplift in the number of users taking the next step, it would suggest your landing page has an issue.
The same can be true when working on a product page. If you develop a new product page and the A/B test proves the new version increases the number of users adding product to basket. If you don’t see an increase in sales, then the basket or checkout pages probably have a problem. Perhaps there is a hidden shipping charge? Maybe it’s the fact there is not a guest checkout? Who knows, but you can investigate and test to identify why and continue on the CRO (Conversion Rate Optimisation) journey.
However, if you are working on a CRO project that is more proposition based, that is perhaps site wide. How do you monitor and declare success?
Scenario 1
You currently offer free delivery on orders over £100. You want to test whether you can increase conversion levels by offering free delivery on all orders over £50. You configure your test to focus on your organic traffic (for new visitors only) and you will run the test with 50% seeing the offer, 50% not. The test is put into action and the number of orders received during the test period increase. Therefore site conversion rates for the test are positive.
So, the sale conversion rate on the website has increased. Do you declare the test a success and deploy across the whole site?
Scenario 2
You currently offer free delivery on orders over £75. You want to test whether you can increase website conversion levels by offering a free returns service. You configure the test to focus on organic traffic (new visitors only). The test will split the traffic 50 % seeing the offer, 50% not. The test is put live. It runs until statistical significance is proven. The website conversion rates increase.
So, the sale conversion rates on the site have increased. Do you declare the test a success and roll out across all visitors?
If it wins, should it be deployed?
In both scenarios we see conversion uplifts. But, is that enough or do you ask to look at other data points as well? Do you also look at the average order value levels and the number of transactions? Do you then look at sales data, delivery data, exchange and return rates to see if, whilst conversion rates have increased, have the profit margins increased?
An increase in site conversion on initial inspection is always good but, in this circumstance, increasing conversion levels by increasing the number of smaller transaction being taken means there will be less profit once the cost of picking, packing and despatching the order are taken into account.
The profit margin can be further eroded if there is an increase in exchanges or refunds as a result of more orders being processed. Again, these elements can increase running costs of the business and stock holding levels if stock levels are replenished. Thus, profit margins are further reduced.
So, is it time for CRO (Conversion Rate Optimisation) to GO?
Don’t worry. I am not advocating we all stop working on CRO (Conversion Rate Optimisation) projects. I am simply asking, is it time to evolve from CRO to Growth Optimisation? CRO is a highly valuable discipline and most successful when developed as part of a culture for a company. However, where CRO tests are devised that are focused around changing a business value proposition e.g. free delivery (saving money) rather than adding a guest checkout (removing friction), it may be useful to not only review the conversion increase but also define and monitor the actions that can occur after the conversion uplift.
Great article, it's the whole quantity vs quality debate. Define 'success' before testing.
I agree. CRO has become a buzz word in some places, and in other places it isn't understood at all. The article hits the nail on the head as to why. So many decision makers in companies are confused about how to include testing in their marketing plans or budget because they don't see the value. I agree with them! CRO generally has a one-track 'mind'. As Lee Duddell says, conversion optimisation shouldn't be about how many tests a company, or an agency runs. I also believe it should be a holistic approach i.e., strategic, 360 degree marketing. We have almost forgotten about integrated marketing. Companies would be far better hiring strategists with experience in analytics, integrated marketing strategy and yes, dare I say it, hiring marketing generalists! A marketer may not sound as sexy as a 'CRO expert', but they will know far more about how to optimise businesses and grow revenue. Plus they will look at the big picture, not just a spreadsheet with a list of tests on it. At the end of the day, conversion/business optimisation should be about meeting customer needs first and foremost, and improving overall net profit for companies via a data-driven, strategic approach to marketing.
Lee Duddell, Thanks for the comment. I think #CRO has, in certain areas, become another buzz phrase and is therefore project orientated and does not deliver its full potential. As you say, if we simply called it "optimisation" and utilised principles from CRO, UX & CX and embedded them as part of an "Optimisation" culture, the long term benefits for both business and customers (or clients) are quite impressive.
Yes. CRO does need to mature. We are still seeing people boasting about the number of A/B experiments they're currently running as if that in itself is some form of result! Conversion Rate is but one factor, as you rightly say, so maybe it should be "Business Optimisation" or just "Optimisation".
Very good points. 100% agree the focus should be the bottom line. True that the "free delivery" conundrum has to be handled with care. My view on this is that if you (and your customers) attach any importance to the quality of delivery, then this should have a value attached (regardless of it being offered or not). Then it is up to you as a brand to decide what the threshold should be (if any) to give it away for free. Conversion optimisation is not the only metric here. Altogether CRO is probably about getting the "quick wins" right in the first place, then looking at impacts across the value chain. Growth management is a wider issue, which is about making a business scalable.