Visitor Insights

In the traditional offline media (TV, radio, print), agencies typically charge 15% of the marketing spend as their fees. So for example, if your company is spending $100,000 on print advertising, the agency that you’ve hired gets $15,000 for their services, which could include ad design and creation, media engagement, etc.

I recently came across a great column by Kevin Lee mentioning that because of all the optimization work that has to be done around SEM, the 15% fee may not be enough and clients should be willing to spend more on agency fees. Kevin is right. There’s a lot of work that needs to be done in the area of SEM optimization. Here’s a few must-dos for an effective campaign:

• Research the list of potential keywords to be targeted
• Select the keywords to be targeted through SEO, those through SEM, and those through both
• Create the right ads for keywords. You simply cannot use the same ad for all your keywords and the rule applies as you go from one outlet to next
• Determine the budget to be allocated to each
• Create specialized landing pages for each ad
• Well, you get the idea …

This becomes especially an issue with smaller clients since they won’t be able to leverage the economies of scale and scope enjoyed by larger clients who spend more money on more keywords.

At the same time, I couldn’t help but notice the disincentive between CPC and agency fees. I understand why agencies use a 15% fee on offline ads - there’s no viable measurement beyond the output. You can only reasonably measure the number of times and the money spent on ads. But in the online world, things are different. You can measure the entire visitor cycle - from impressions, to clicks to conversions.

Think about it: would you rather spend $10,000 and generate 10 leads or spend $1,000 and generate 50 leads? Of course you’d like the latter, but in the world where agencies are paid by CPC, their incentive is to spend more money. They’ll benefit from having you spend $10,000, whether the campaigns are optimized or not, whether the conversion funnels are streamlined or not.

Ideally, agency incentives should be the same as the clients: get the most you possibly can from your marketing budget.  This means that their fees should be tied to your conversions - online purchase, leads, registrations or whatever they may be. Don’t get me wrong. I’m not saying agencies should not be paid for all the optimization work that they do. They absolutely should. But perhaps it’s time to introduce more complex pricing schemes that do every body justice: a one-time fee and maintenance fee to cover the work to be done to get the campaign going, and a conversion/engagement bonus that gives the agencies the incentive to outperform.

This way, the agencies will get compensated with all the leg work necessary to get started and will be incentivized to constantly outperform, to constantly work on ways to increase their campaign conversions.

I’d also like to hear about some of the payment schemes you’ve used that have worked for you.

This is the third and final post of a three-part series on top reports for your site search. As a reminder, they are:

1. Search Usage
2. Searches per Search Visit
3. Top Search Terms
4. Top Failed Terms
5. Visitor Segmentation
6. Top Pages Driving Search
7. Search Groupings
8. Success Rate
9. Micro Step Success Rate

In this post, we’re going to cover the last three items.

7. Search Groupings

This practice refers to grouping of search terms into categories that make business sense. These groupings ought to be done by business people and not necessarily managed by the capabilities of your site search tool or content management solution.

For example, an electronics reseller may want to categorize the terms “ipod” and “zune” as the category “MP3 players” and treat “ipod case”, “ipod speaker” and “ipod armband” as the category “MP3 Accessories”. We find this practice to be manageable when done in a spreadsheet such as Microsoft Excel and using the vlookup function.

Why go through this exercise? The answer is because it reveals information about which categories or groupings are most popular among searchers. The information can be used to build a better navigation menu – one that supports your visitors’ interests. So for example, if you find the category “MP3 Accessories” to represent a large enough portion of your site search traffic, then you can add the category into your navigation menu.

8. Success Rate

Let’s face it. The whole point of implementing a site search is to enhance the user experience towards your business objectives. So if your business objectives is online sales, then site search performance should be assessed using this objective. And if your site goal is to generate leads, then your site search results should be assessed using that objective. Most web analytics tools today will account for this type of reporting. The goal is simply to understand what search terms are converting best so that you can improve your site-wide promotional offerings. Consider the table shown below.

Simply looking at the search conversion report reveals what products have the better success rate and therefore will provide a higher return when being promoted on your site. So if you’re considering promoting a gaming device on your home page, you can see from the above data that Playstation 3 provides the biggest bang for the buck. Even though its conversion rate is less than Xbox 360, because of its higher price it provides more revenue per search than other products.

The search conversion report will provide you information on what works best – in terms of conversions and dollar values. However when it comes to providing more insight about the reasons, it can come short. This is why we use micro-conversions so that we gain a deeper understanding into the sales funnel. This is covered in the next point.

9. Micro Step Success Rate

Lets face it. Your visitors don’t buy or fill out a form straight from your search pages. The goal of your site search is not to sell an item, but to help you find it. There is a sales cycle that visitors have to go through. First they have to find the product (this is what your site search is designed to do). Then, they have to add the product to cart, and they’ll have to go through the checkout steps on your site. At each stage, various things have to be accomplished to get the visitor to the next stage.

Let’s go back to the conversion table from the previous example, where the conversion rate for each search term is shown. We can see from the table that the terms “wii” and “wii sports bundle” are really underperforming compared to others in their category. Does this means that your visitors are not interested in Wii products? Not necessarily. The best way to find out is to look at the micro conversions for these same search terms. This is shown in the table below.

The micro conversions reveal a very different picture. It shows that in fact the click-throughs from search to product pages for the terms “wii” and “wii sports bundle” are higher than those of “playstation 3” and “xbox 360”. It is however at the product pages that the Wii products underperform. We therefore need to revisit these product pages, at which point we come across the following scenario, shown below.

These products enjoy a high click-through, but once visitors get to the product page, they’re presented with an “out of stock” message that stops their progress.

Micro conversions let you see the progress at each step and therefore let you diagnose your site at each step of the way. They make it easier to understand where in your site you should concentrate in order to streamline the site conversion.

Conclusion

Of course, based on your site objectives and site search deployment, you will find some reports more valuable than others, but you should be able to select from the list provided in this series to better optimize your site search. Also, if you need help getting the right reporting dashboard in place, we’re here to help.

This is the second post of a three-post series on top reports for optimizing your site search. In our last post, we covered the first three of the list below. Again, the top reports are:

1. Search Usage
2. Searches per Search Visit
3. Top Search Terms
4. Top Failed Terms
5. Visitor Segmentation
6. Top Pages Driving Search
7. Search Groupings
8. Success Rate
9. Micro Step Success Rate

In this post, we’re going to cover the remaining items.

4. Top Failed Terms

Failed terms are defined as search terms that provide no results to the user. Why is this important? Because it reveals gaps in your site search deployment. You can use this information in a number of ways:

• If a product that you carry is showing up as a failed search, then you clearly need to reindex your site to make sure that those products are displayed in your search results.
• Products you should carry? You may find a series of products that you currently don’t carry but should be in your inventory. The failed terms report will let you know how much money you’re leaving on the table by not carrying these items.
• Misspellings – Let’s face it. Visitors don’t always correctly spell every word. You should use this report to detect such misspellings and optimize your site search tool for such occasions. For an example of good implementation, go to Overstock.com and do a search for “bedding” and “beding” (screen shot below). You’ll see the same results in both cases.

5. Visitor Segmentation

Not all visitors are created equal. Visitors of different segments will behave differently and therefore will search differently. Two segments of interest are “new visitors” and “returning visitors”. By understanding the difference in search behavior between these two segments, you’ll be able to see your site navigation’s effectiveness between these two groups. Typically, you want to look for abnormalities in ratios between new and returning visitors.

Consider the example in the above screen shot. What you see is the top search terms and the breakdown of search between first time and returning visitors. Overall on the site, returning visitors constitute 19.80% of site visitors and we can see that this ratio also holds for most of the search terms. However, you can see that the terms “vaio battery” and “macbook battery” both get a greater share of search from returning visitors. Why is it that a larger share of returning visitors are searching for battery products compared to other products? Perhaps it’s because your battery inventory is not well promoted on the site and as a result a smaller share of first time visitors is aware of their existence on your site. Whatever the reason, this high discrepancy in ratios should alert you about an investigation that ought to be made between your navigation and the battery products.

6. Top Pages Driving Search

This report provides a listing of pages that the visitors were on prior to conducting a search. In most web analytics applications, you can simply look at the navigation reports or previous pages to get this data. It reveals what pages did your visitors get stuck before resorting to search. For many people, this may simply be the home page or the landing page, which is not a cause for concern since some people simply resort to site search as their first option. However, if the search originated on a product page, then it could be cause for concern since the product that the visitor was browsing may not have been what they were searching for. In this case, you’re looking for those oddities that could give you insight onto the navigation flow and visitor interest.

In our next and final post, we’ll cover the remaining reports that you should be looking at for site search analysis.

What is the most valuable report that you can get from your Web analytics package? Arguably, it’s the site search report – also referred to as internal search.

Why? Because most of what web analytics gives you is quantitative data – what your visitors do on your site - and it leaves it to you to figure out from the volumes of data what your visitors are looking for. Site search on the other hand is user-entered data – your visitors are actually telling you what they’re after.

In this three-part series, we’re going to provide information on the top reports or data points to look for when analyzing site search. The top reports are:

1. Search Usage
2. Searches per Search Visit
3. Top Search Terms
4. Top Failed Terms
5. Visitor Segmentation
6. Top Pages Driving Search
7. Search Groupings
8. Success Rate
9. Micro Step Success Rate

1. Search Usage

This metric is measured as the ratio of visits (or sessions) conducting site search to overall site visits. For example, if your site had 1,000 sessions and 65 of them conducted a site search (hit the search results page), then your search usage (as a visitor ratio) is 6.5%.

Why is this important? Because the search usage could be an indicator of navigation shortcomings. Site search is a navigation feature and is meant to complement your site navigation design. In cases where your navigation fails, then your visitors resort to site search. This isn’t always the case, as we’ll discuss in ensuing points, but if you have a high search usage, then it is an indicator that your visitors are bypassing your navigation and therefore you should be revisiting your navigation design.

Like all KPIs, search usage should be trended so that you can see the impact of site changes to the ratio. In our studies of top e-commerce sites, the site search usage is between 5% to 10%.

2. Searches per Search Visit

This KPI is measured as the ratio of the total number of searches conducted divided by the number search visits (or sessions). Going back to the previous example, say the 65 sessions or users of site search conducted a total of 260 searches. This means that the average search user conducts 4 searches.

Obviously you want your site search results to be accurate and to the point. This means that the higher this number, the more cause for concern you have that your site search is not providing the necessary information to users, forcing them to refine their searches. Realistically, you want the average user to search for no more than two searches per session. Obviously, based on the complexity of the product or service, this number may vary.

3. Top Search Terms

This is the core of site search reporting. What are your visitors searching for? What are they telling you? For example, if your top search terms are “Wii”, “Xbox” and “PlayStation”, then it will be a good idea to have a gaming related promotion or banner on your home page and other main landing pages, since a large number of your visitors are interested in this category. Additionally, because site search complements your navigation, you can use the information to conduct gap analysis on your navigation menus. This means that you can look at your navigation menu and if the category “Gaming” is not being presented in there, then you should consider adding it to the menu. The end goal is to create a navigation menu that reflects your visitors’ intent.

In our next posts, we’ll cover the other reports and KPIs that you should be looking at for site search analysis.