How To Calculate My Website Value?

How do you calculate the value of increased website performance?

• For a pure media website, such as http://cnn.com or http://huffingtonpost.com, how do you place a quantitative value on increased website performance? For example, if a site initially has a load time of 11 seconds per page and has the opportunity to pay an CDN a X amount of dollars that guarantees the load time will drop to 5 seconds, what decrease in bounce rate/increase in traffic/change in some other quantifiable metric can be expected to justify the dollar investment?

• Answer:

  If understand your question question right, you want a Cost Benefit Analysis for a web performance optimization project. I do not have any public real examples for this, but I am not even sure they would matter as they will vary by site and also by what the project entails! Two sites are not the same, even if they are in the same business. To run such analysis you need to look at what is impacting your revenues, current costs, costs associated with implementing the project, and any on-going costs (like CDN fees). Going from serving static content in house to serving from a CDN will be pretty straight forward, I am sure any CDN can share information with you on how similar sites calculate theirs. In this case you are just outsourcing part of the infrastructure, so you can analyze what your current cost for serving such content is (servers, datacenter, bandwidth, operators) and just compare the cost of serving from the CDN, plus the one-time cost of implementing it (this could be 0 depending on the platform you are using or built). You might see right there a benefit, without even looking at the revenue improvement. In the case the project is more complicated, than you would need to gather more data. Starting with revenue, and moving to cost. The primary source for any media business is Advertising. Therefore their major goal is to increase the number unique viewers and pageviews (media consumption). The larger the number of unique viewers (or reach) on the site, the wider variety of advertisements they can sell and the more campaigns they can accept. That said media sites might have secondary sources paid subscription models. This case is no different than selling products or lead generation site, you care about conversions. It is key that you build a baseline of metrics before the web performance optimization happens. It is also important to take in account any other activities in the company that could impact the numbers (advertising campaigns, partnerships, etc). The web analytics metrics to consider are: • Unique Visitors • Pageviews • Conversions (for paid subscriptions, email subscriptions, etc) • #pageviews/visit or Page/Visits - Happy visitors should consume more media, therefore this number needs to go up. • Time on Site - This metric is a little tricky, slow performance is included in this measurement - however if media consumption increases - you will see this number go up. • Time on Page -Important for embedded video content, as the more time spent on the screen the more ads can be displayed. In traditional web content this does not matter as the number of ads delivered does not increase as viewers spends more time on page. Of course for video content you would need to look at other metrics in addition to this one. • Bounce Rate • Exit Rate - Most people look at Bounces, but that is half the story. Visitors will be more likely to exit a site on a slow page - this will not count as a bounce, unless it was the first pageview. You can use this metric if you implement changes to only webpage (ideally the one that is the slowest and has the highest exit rate). • New Visits by Google - Search engines are a major contributor to website traffic. Since Google takes in account web performance in their ranking, you should see benefits (in theory) in your ranking. You should also consider looking at any SEO and or Adwords spend numbers to see if there was an impact. The cost calculations will be a little more tricky. As mentioned earlier an optimization activity could lower or increase the cost depending on what is being implemented. There will be fixed onetime costs, and variable costs that will be impacted. The major ones to look at: • Lower bandwidth utilization( keep in mind that as media consumptions increases, this increases). Moving to a CDN will lower bandwidth utilization tremendously! Implement GZIP will also improve it.  A lot of the techniques dealing with merging (sprites) or minifying will also decrease this cost. • Number of Servers. Depending on what optimizations are done there will be impact of the amount of servers required to serve content. Less requests to a server for a page, the more pages the server can server - the more users it can handle. However, more processing at the server would mean lower throughput. • Less Servers, means less datacenter space, etc. • There will be other things impacted and someone on the IT side will need to look carefully and list out these impacts and costs associated with them.   Hope this helps...