KPI exposed: The official how to guide for KPI (Part III)

Welcome back everyone for the third and final article of this series of three. If you haven’t read the first part and second part yet, I encourage you to do so. In this final attempt to draw a clear and complete picture about what KPIs are, how to choose them and generate your own executive dashboard, I will now provide you a list of the most popular KPIs used in a web context. Each KPI will be explained and provided with a specific formula.

I will also quickly explain how to set up thoughtful objectives.  After all, the logical step after tracking your performances is to set objectives for your future actions! Finally, I will rapidly wrap everything we covered in the 3 articles of this series.

DUMB Objectives

Before going any further, it is obligatory to re-state the importance of your dashboard as a tracking tool. Il will allow you as an executive to see how your company performed in the last days, weeks or quarter. With all due respect to the importance of the historical data collected, the foremost important contribution of your dashboard will be to accurately enable you to set objectives for your web marketing team. By using all your analytic raw data, refining them with your KPIs you will obtain a refine chronological evolutionary chart of your company performances. Using this chart combined with your business strategic plan, you will be in a position to make future recommendations and set business objectives.  Those objectives are what are going to drive your team to achieve excellence. I recommend you to use the DUMB methodology when it comes to select your objectives.

Here’s how DUMB objective breaks down:

• Doable
• Understandable
• Manageable
• Beneficial

KPIs, KPIs and KPIs

It took three articles to get there, but we finally did it! I will now address some of the most popular KPIs in a web-marketing context. Obviously there are many other interesting KPIs and once more used to it, you will probably even come up with your own. I used a variety of sources for this section ranging from books, personal experiences al the way to presentation material from various references. I did my best explaining the vast majority of the term used below. If you stick on a particular expression, I encourage you to leave a comment or to visit this lexicon by Avinash Kaushik.

WebSite Traffic Metrics

Take Rates

TR= # of visits performing the desired outcome / # visits

TR= # of visits performing the desired outcome / # unique visits

This metrics will help you gather more information than just the amount of newsletter registration, eBook download or RSS feed subscription. The take rate really makes sense when used in a/b testing context. It is of great help when testing eBook positioning on various pages for example. Similarly, if you are generating revenue with ads, the take rate can enable you to compare the different rates obtained when trying your ads at various locations on your page.

Repeat Visitor Share

RVS= # of repeat unique visitors / # unique visitors

This is the number of visitor that comes back to your site over time. You have to determine what period is most relevant for your business/industry in order to have comparable data. This metric is often ignored, but it is a great indicator of the quality of your content. We assume that normally more and more people should come back to your site because you should always keep on posting relevant and quality content.

Heavy User Share

HUS= # visits with over 11 pages viewed / # visits

The number of 11 pages is arbitrary and really depends on your sites and industry. In can be increased or decreased according to your needs. This indicator enables you to see the share of your user who makes a profound use of your sites. It is a great metric in order to assess the engagement level of your site.

Committed Visitor Share

CVS= # visits that lasts more then 19 minutes / # visits

This metric is based on the assumption that the longer a visitor stays on your site, the more odds it has to perform the desired actions. It highly correlates with Take Rates and sales metrics. However, there are many elements to take in consideration when using this metric.

First off all, the length of the session does not always translate in longer visitors activity. Visitor can open a tab, land on your site perform one action and then open another tab and simply forget your site for 25 minutes. A visitor can also go on your site and leave for a cup of coffee and come back 15 minutes later. Also, the interpretation of this metric varies greatly depending on your industry. If I offer self serves service on my site, I clearly want to stay away from long visit whereas if I am operating a game website, then I want the longest visit possible. Make sure to read the second part of my guide to know what type of website you are operating.

Committed Visitor Index

CVI= # of pages view in the visits that lasts more than 19 minutes / # visits that lasts more then 19 minutes

This ratio is to be used in conjuncture with the Committed Visitor Share. It combines the time spent with the quantity of pages viewed. It allows you to see if your client are just opening your site and going for a coffee! The more your site is easy to use (a great user experience), the more your site should have pages viewed for longer visits. You want to see this KPI increase over time.

Committed Visitor Volume

CVV= # of pages view in the visits that lasts more than 19 minutes / # pages viewed

This gives you the percentages of total page views on your site by your committed visitors. If it is to low, it means that you are targeting the wrong audience for your site, and you should consider reviewing your personas.

Visitor Engagement

Visitor Engagement Index

VEI= # visits / # unique visits

This KPI indicates you the tendency of your user to have multiple sessions on your site.  The larger the engagement in term of multiple visits, the further above 1 your index is going to be. Blog wants to have this index as high as possible whereas commerce site want this number to be as close to 1 as possible.

Reject Rate: all pages

RR-all= # 1 page view / # visits

This gives you the general bounces rate of all your pages. It is an interesting metrics to track. You’ll want to see this metric go down over time.

Reject Rate: Top entry

RR-top= # 1 page view-top entry / # visits starting by the top entry

This metric gives you the bounce rate for your top entry pages. The rule of thumb is that your top 5 entry pages account for approximately 80% of the traffic. I would recommend you to take the Pareto law in consideration and focus on the top 5 entry pages. More than that really becomes time consuming.

Scanning Visitor Share

SVS= # visit lasting 1 minutes / # visits

Allow you to evaluate the “scanability” of your site.  Having visit that lasts a minute can be interesting in a usability context. It can mean that your visitors are able to quickly find what they are looking for. This metric gives you the share of scanner that visits your site. Make sure to use this KPI in conjuncture with the Scanning Visitor Index especially if the percentage is high.

Scanning Visitor Index

SVI= # of pages viewed in the visit lasting 1 minutes / # visit lasting 1 minutes

This index is to be used with the Scanning Visitor Share. If it is close to 1, it might be a sign of navigation problem on your site.

Scanning Visitor Volume

SVV= # of pages viewed in the visit lasting 1 minutes / # pages viewed

This gives you the volume of scanner that visits your site. Depending on the orientation of your site you might have a different perception of this metric. Service website will seek to have a lower volume whereas content website such as Blog will want it to be higher.

eCommerce metrics

Average Order Amount

AOA= Total sales / # of orders

Many web marketers are focusing solely on the conversion rate, but the average order amount is a very important metric.  There are many tactics that exist to drive your average order amount up. Free shipping over a certain order amount, upsell attempt, similar product and others are all tactics that can be used to leverage your order size.

Conversion Rate

CR= # of order / # visits

This now famous KPI is the one that everyone talks about and with good reason. A conversion is nothing less than the realization of desired outcome on your site; it’s a win for you! Many people think that only sales are conversion. In fact, there exist many types of conversions. Avinash wrote an article about micro and macro conversion that I encourage you to read.

Sales Per Visit

SPV= Total sales / # visits

This indicator tells you how many sales you are making per 100 visits. It gives a good perspective on the overall efficacy of the website.

Cost Per Order

CPO= Marketing Cost / # of order

This will give you the marketing cost per order. It is a critical measure to have in order to assess the profitability of your web marketing efforts.

Cost Per Acquisition (CPA)

CPA= Total cost of referring source / Number of confirmed conversions

Shows the cost of obtaining one conversion. It is a very interesting KPI to track in order to evaluate the performance of your PPC campaign.

Repeat Order Rate

ROR= # of client orders / Total order

This measure enables you to see the % of your sales that is made by repeat business from your existing clients. Again, it is important to establish a time period for this metric in order to have relevant comparison. This measure can be used with the client retention rate.

Client Retention Rate

CRR= # of client orders / # Client

This KPI gives you the rate at which your clients are coming back and making other purchases. This KPI is not used as often by the web marketing community but is a metric that I really like.

Cost Per Visit

CPV= Marketing cost / # visits

This KPI allows you to calculate how much each visit cost you in term of marketing. It is interesting to track it to keep control and monitor the marketing expenses. Depending on the data you have access to you can also calculate the CPV client and CPV new clients

Order Acquisition Gap

OAG= CPV – CPO

This number is going to be negative and will represent the cost of not converting a potential client. Your objective will be to reduce this number.

Order Acquisition Ratio

OAR = CPO / CPV

This ratio is going to give you a positive number that represent how much more it is going to cost you to acquire a new customer. An OAR of 25 means that it is going to cost you 25 time more to acquire a new client than a new visitor.

Return On Investment

ROI = (Net profit / Investment) × 100

Another KPI that is well known in the business community is the ROI. This formula will give you a % of return on investment. This measure is very broad and encompasses many source of investment but does provide a financial incentive to justify your web marketing expenses!

Conclusion

KPI enables the transformation of raw data into meaningful information and by having a dashboard, it enables you as web marketing managers or executives to have quickly grasped the essentials of the web performance of your business. We talked about how to develop KPIs, how ratio and percentages where preferred to simple un-processed raw data, how to select SMART KPIs, how to categorize your business questions using the REAN model, the different kind of website that exist based on their core motive, how to generate DUMB objectives and finally, we covered a whole list of web marketing KPIs.

As managers or executives, we often seek to obtain the purest data possible. You have to keep in mind that the accuracy of your data in web marketing never really is of prime concern. What matter the most is the consistency of those data so that you can accurately highlight the various trends that take place over time. And it is with those trends that you will be able to plan your action further down the road.

I hope that this series of articles helped you to understand what KPI are, how to select them, how to set objectives and how to build your own dashboard. If you have any comment or recommendation, I will gladly answer them.

Source:

Cult of Analytics by Steve Jackson. Book available here

Drilling Down website

Web Analytics: Nicolas Malo, Jacques Warren. Book available here

Print PDF