Effective eMarketing Campaigns – Part Two

It’s taken a while but here we go…

Channel segmentation and separating responses

To track the success of each referring channel where visitors have come from it has to be explicitly identified. This is done by creating a set of parameters that are passed via the web address. The parameters uniquely identify the source of the visit.

For offline sources there is an additional stage to the process as users manually have to enter a web address.


Online links can be clicked on and are normally hidden in a banner or beneath text so the unique parameters can be hidden within the link.


Offline visitors need to manually enter a web address in order to access the appropriate web page. This needs to be different for each referral type. For this reason, any URL published should be:

  • Short
  • Simple to type
  • Easy to remember
  • Obvious

My experience has shown that using esoteric or non-standard formats leads to a reduction in visits. A common standard for published addresses is Mainsite/subname. This is well understood by consumers as it is used by popular sites such as the BBC, Dell, The Guardian and Amazon. Examples include:

  • www.bbc.co.uk/health
  • www.dell.co.uk/offers

This format helps tp reinforce the company name and allows the re-use of a standard set of URLs so that, if someone visits from an old advert, they still get up to date information.

To track each offline referral channel separately a separate URL must be used for each one. E.g.:

  • Ad 1 – www.contentsecure/offer1
  • Ad 2 – www.contentsecure /offer2
  • Direct-mail – www.contentsecure /offers
  • Door drop – www.contentsecure/special

Each address then redirects automatically to the appropriate web pages, passing the invisible parameters that are picked up by the analytics tool, in the same way as online visitors.


A standard set of metrics is required to:

  • Measure the most cost-effective and successful referral channels
  • Identify which media placements are most effective
  • Success of the creative
  • Which channels work best for each customer segment

What to measure

The exact set of statistics produced for a campaign will depend on a number of factors. These include:

  • Method(s) of delivery
  • Number of channels
  • Segmentation of customer
  • Campaign purpose/targets

For each channel, the core metrics include:


  • Number of emails sent
  • Number of emails opened
  • Time between opening and delivery
  • Click-throughs – which links did they click on?
  • Unsubscribes
  • Bounce-backs (hard/soft) – is the email deliverable (permanent/temporary)?

Web *

  • Number of visitors – split by channel
  • Visits by new and existing visitors
  • Visit from – including offline media
  • Goal conversions e.g.:
    • • Number of calls
    • • Whitepapers downloaded
    • • Trial an online demo

* For web stats, there are a number of additional statistics than can be collected (e.g. browser version) that are not relevant to campaign management. These will be used to ensure the site continues to meet the standards required by visitors.


One of the key metrics to measure is the actual cost per visitor. The exact formula for this depends on the channel they have visited from but the key metrics include:

  • Ad impressions
  • Clickthrough rate
  • Cost per click
  • Leads generated
  • Cost per lead
  • Conversion rate
  • Number of sales
  • Cost per sale
  • Campaign reporting

A standard template is used to report on the statistics. The aim of the report should be to present a meaningful summary, as well as detailed statistics to meet the requirements of all readers. This should include:

  • An executive summary
  • Results Vs benchmarks/expections
  • Ongoing trends over the year, split by month/campaign
  • Campaign-specific results and reviews
  • Key graphs
  • In-depth stats
  • Focus on core channels (e.g. email)