You can call us on +44 (0) 1 225 840 490

Name*
email*
Company*
Telephone*
Information*
  Please enter code (case sensitive)*:

Check

 
 

If It Matters Measure It - KPIs And Internet Marketing

"If you can not measure it, you can not improve it."

William Thomson, 1st Baron Kelvin OM GCVO PC PRS PRSE (26 June 1824 – 17 December 1907)

Perhaps, not a very fashionable view today, but it is true. Unless you believe that 'subjective improvement' (i.e. something based on or influenced by personal feelings, tastes, or opinions) is just as useful or valid as 'objective improvement (i.e. facts).

When it comes to people that may be the case. When it comes to business, subjective views just won't cut it.

"Yep, I think we sold a few more widgets this month and my gut tells me we spent a bit less and I haven't heard of any more customer complaints" not being a very good way to do things seems obvious; but how many businesses really measure the things that matter?

Pencil graphClearly measuring the things that matter implies that you know what they are. However, with so many things that can be measured and so many tools and "dashboards" available knowing what a true Key Performance Indicator (KPI) is more important than ever.

So how do you know what to measure and whether it should be a KPI?

Well, if it does not provide everyone who sees it with complete clarity about the objective to which it refers and how that relates to the overall success of the organisation or department it cannot be a KPI.

A little more 'teaching granny' may be required. Let's define an objective. We all know it's something that one intends to accomplish, but it should also be SMART:

  • Specific
  • Measurable
  • Achievable
  • Realistic
  • Timely

So:

  • "To achieve a 20% return on capital employedby August 2014"
  • "To decrease the time to respond to customerqueries by 3 minutes by the end of this year"

are both good examples, assuming the measurable element is achievable in the timescale and the resources to make it happen exist.

Returning to KPIs. A KPI by its very definition it must be something that is Key. Key means make or break, things that are crucial like return on capital, cashflow, number of customers and competitive advantage.

Again by definition Performance has to be something the business is able to measure, quantify and affect. You can only affect something over which you have some level of control.

KeyClearly an indicator 'points at' something. So, when combined with Key Performance it should be providing a clue about future performance as well as past results and give you some basic idea that action needs to occur and what it might be.

So, a KPI is a measure that is:

  • directly related to your business objectives
  • really "key" to what's happening and that you can affect
  • something that promotes action and gets you to improve

Internet Marketing & Website KPIs

So KPI's can help a business and departments to measure and monitor progress towards SMART objectives. But when it comes to your website and other aspects of internet marketing such as pay per click, email & social media what should you be measuring?

Circular question! What are your objectives.

Let's assume you have an ecommerce site.

Presumably you want to sell your widgets of all types in vast volumes. So given that it's a transactional site (i.e. you are selling 'widgets' online) then presumably the KPIs that count are to do with customers and sales. In that case your KPIs for web customers would be little different for those for 'offline catalogue' customers, things such as:

  1. Cost per customer
    1. Visitors by channel
    2. Conversion rate by channel
    3. Cost per response by channel
    4. Cost per customer by channel

  2. Recency frequency and value measures
    1. £ sales per customer - new
    2. £ sales per customer - existing
    3. Frequency of purchase
    4. Date of last purchase

  3. Shopping cart conversion rates
    1. Cart abandon rate
    2. Total conversion rate
    3. Product conversion
    4. Visitor conversion
    5. Lost value

Most of these measures can also be 'matrixed' so that you can get for example "£ sales per customer - new for social media" when you 'drill down'.

PerformanceOK, you say, that's great. But these elements essentially measure sales performance and whilst sales success is devoutly to be desired aren't there other things that matter?

Well yes, but one can argue they are not KPIs. They may affect the KPIs and be actionable in themselves, but if your frequency of purchase is dropping that is what you need to know - whether that's because of worse products or reduced visitor conversion is subsidiary and explanatory, not, in this case, a KPI.

What you sell will also affect your KPIs. If instead of widgets you were selling website advertising or a magazine subscription the factors that are important would change to ratios involving, for example, revenues per impression / click for the website advertising and offer types and renewal rates for the magazine subscription.

If your site is a lead generation site then the KPIs change yet again. Most lead generation sites are not truly transactional in that it is impossible to say exactly which keyword or referring site created the customer lead. For example, those cases that come in via the telephone may be the bulk of leads generated and even come in days or weeks later.

That said, making sure that those customers who do complete goals such as filling in forms are properly monitored and that the conversion path is known is clearly important.

Then it is a question of monitoring factors such as:

  1. Percent of new visitors
  2. Average number of page views per visit
  3. Time on page
  4. Bounce rate
  5. Cost per visitor

to see the (positive or negative) relationship between these factors and the number of leads generated.

Written by Richard Hill



ECRM on FacebookECRM on Facebook
ECRM on Google+
ECRM on LinkedInECRM on LinkedIn
ECRM on TwitterECRM on Twitter
ECRM on Twylah
Richard on Google+Richard on Google Plus
Richard on LinkedInRichard on LinkedIn
Richard tweeting as cassyputRichard on Twitter
Richard on Twylah
Local navigationNewsletters

Daily news

2012

April: Website design & development brief

March: Web analytics and webmaster tools

February: Choosing a web design company

January: Mobile web & search strategy

2011      See 2011 list      Hide 2011 list

2010      See 2010 list      Hide 2010 list

2009      See 2009 list      Hide 2009 list

2008      See 2008 list      Hide 2008 list

2007      See 2007 list      Hide 2007 list

2006      See 2006 list      Hide 2006 list

2005      See 2005 list      Hide 2005 list

Essays    See 2005 list      Hide 2005 list

Take the Test
Take the Test Take the test and see how hard your website is working
How hard is your website working for your business? Take the test and see how hard your website is working
Case Studies