Finding What Terms Are Converting Into Sales/Tracking Keywords to Conversion with Weighting

1. Finding What Terms Are Converting Into Sales/Tracking Keywords to Conversion With Weighting
   

   Having 100,000 unique visitors a day really doesn't matter in the end if you aren't getting any conversions (new members, info requests,
   sales).
   
   Measuring successes and failures for landing pages, on-page content like CTA's, and especially keyword to sale are some of the most important pieces of information that you can gather and use to improve and optimise your overall website.
   

   Here are two scenarios to better illustrate this point;
   

   1. Paid Advertising – A car insurance company starts a paid advertising campaign on Google and after a week or so they see that the name of their company or their 'brand' seems to be converting the majority of their sales. Because of this discovery, they target the majority of their budget on their brand terms like ABC Insurance and ABC Insurance Company.
      

      A week later they see that their CPA (cost per acquisition) has sky-rocketed almost two-fold and can't figure out why this is. When they look at Google analytics and other third-party tracking software, they both say the same thing.
      

      So why is this?
      

      Let's take a look at the buying process (also called funnel tracking) to see where they went wrong; Mrs.INeedInsurance hopped online while enjoying her morning java to look for insurance because last night when Mr.INeedInsurance opened his renewal notice he got a significant premium hike. At dinner they decided to start shopping around for insurance. Mrs.INeedInsurance searched 'car insurance' between 6-8am that day, going in and out of different companies websites, learning what she was up against…tens of 1000's of results. So at work (11a-2pm is the #1 time people shop online – not necessarily making purchases) Mrs.INeedInsurance has learned a bit about search and decides to add her city in the query. This time she searches 'car insurance London', and still gets several thousand results, but at least they are localised, and there are a few that she recognizes from this morning so she goes in and fills a few of the forms out to get quotes. Throughout the rest of the day she gets the quotes either immediately from the website or via email. Now she's getting somewhere. Jump forward to after dinner that evening. Mr.INeedInsurance looks through the notes his wife brought home and decides that ABC Insurance offers the best deal for the money, then goes to Google and searches for ABC Insurance and makes the purchase.
      

      See what happened here? I use this as an example because this is exactly what I identified for a client a few years back that inevitably led to changes that doubled their conversions.
      

      The problem is that all the data pointed to ABC Insurance's brand name as being the top converting term, so that's where they concentrated the bulk of their budget. In actuality, 'car insurance' and then 'car insurance London' were the terms that actually led up to the sale.
      

      The reason that this is important for PPC campaigns, or any paid advertising, is that many will allow you to do keyword weighting. This is where you increase your bids or decrease your bids by a percentage according to day parting. Day parting is turning your ads up or down according to the time table that you put in place.
      

      In this instance I would turn my bids up to 125% on 'car insurance' and 'car insurance London' in the morning and afternoon, then down at night. On 'ABC Insurance' I would turn the bids down in the morning to 50%, and then back up to 125% in the evening.
      

      Keyword weighting also allows you to weight your keywords and track them to conversion. It places a cookie on the end-users computer to track what keyword brought them to the sight, what keyword resulted in a quote, and what keyword resulted in a sale.
      

      This is beneficial because I can further adjust my bidding strategies according to demographics and geographical metrics.
      

      With these cookies I can also successfully measure and establish LTV (Lifetime Values) of the average customer. This allows me to adjust the conversion value, which allows me to go back to my company/client and potentially get a higher advertising budget.
      

      Using this same insurance company as an example; initially they gave me a conversion value of $25. Now, since we were able to identify other sales made by this customer, the conversion value is $40.
      

      Offline this company spends 100,000 on advertising through different venues, acquiring customers at a cost average of £/$56. Guess what happened the next month? They increased the budget by 100,000.
      

   2. Organic Advertising – Same scenario as above, except ABC Insurance Company identifies through log files or Google Analytics that his top converting keyword that is getting sales is car insurance.
      

      In light of this, the decision maker decides to create a landing page that is fully optimised so that the relevancy grade that all 3 search engines use will increase their organic positions, which it will.
      

      The problem here is that the term that was actually bringing them to the website to buy was 'cheap car insurance'. If they had identified this they could have built the page around the term, 'cheap car insurance' rather than just 'car insurance'. This would have served double-duty and acted as a great landing page for both keyword phrases.
      

   This is why tracking your keywords to conversion is so important. It can save thousands on paid advertising and identify the actual keyword phrases that need pages built around for improving organic rankings.
   

   If you are experiencing a high bounce rate or what you feel is high cart abandonment, you might be surprised to find that many didn't buy elsewhere; they actually came back to you and bought.
   

   This is also helpful in refining your stats. Rather than show this customer as 3 separate visitors, it identifies (through the cookies) that they were actually just one visitor, and the bounce rate or cart abandonment is significantly reduced.
   

   This information can very invaluable as well.
   

   For instance, maybe I was getting high unique cart abandonment from unique users that was significantly higher once they went to checkout. I know that happens when I add shipping costs into the total. So I might try to do some A/B testing with and without shipping costs listed separately, added into the price initially and adding it during checkout and see which converts better. Or I may set the website up to recognize the cookie and create a drop down that offers free shipping today with any purchase over $/£XX.XX.
   

   There are endless possibilities to use this information for.
   

   There are many good tools out there to measure these variables and that will let you set up rules and keyword weighting, as well as many other great features. Some of these are;
   WebTrends
   DirectTracks
   BidBuddy