Improve Your Merchandising By Analyzing Browse To Act Ratios

Keys to Unlocking Your Web Marketing Genius

Improve Your Merchandising by Analyzing Browse-to-Act Ratios By Chris Grant, Enlighten

Based on WebTrends

Take 10 Series

Improve your Merchandising by Analyzing Browse-to-Act Ratios Whether you have an e-commerce site or not, chances are you’re making merchandising decisions. What content should you feature on the homepage to get more visitors to check out your products and services? Should you include an image of a deliverable to encourage higher registration? Whatever merchandising questions

In other words, you’re able to observe actions that imply interest in a particular item, prior to a final action such as purchasing. And if you know which items seem interesting to a lot of people, you can draw conclusions about why, and then apply those insights to other items.

you’re asking, web analytics can give you the insight to answer them. It’s terrific customer intelligence.

Defining Merchandising Merchandising, of course, is a motley collection of factors that together affect that mysterious intangible called “appeal.” Some of the factors have to do with their presentation on your web site, but many do not. Web-related merchandising factors include the size and quality of an item’s picture, the amount of detail shown in the image, whether all possible product colors are displayed, whether the item is shown in use or alone, the descriptive text, the caption or heading text, the placement on the page and the location on the site to name just a few. Non-web factors might include price, warranty, shipping costs and shipping time. And don’t forget that the total mix of items that you offer is also merchandising, involving questions about whether you need more items of a certain type, or whether some should be dropped.

Understanding Appeal —the E-Commerce Context Let’s start with a sales-based merchandising question. Consider the following items, all priced identically: - Item A is a best seller. An impressive 10% of visits include a purchase of Item A. - Item B never sells very well. Only 1/10 of 1% of visits include a purchase of Item B. - Items C and D are in the middle. 1% of visits include a purchase of Item C and 1% include a purchase of D. Given this information, would you … - Raise the price of Item A? Or lower it? - Drop Item B? Or promote it more? - Feature Item C more often? Or treat Items C and D exactly alike?

The web is unmatched for consumer analysis. The wonderful thing about web site analytics is that you’re in a position to closely watch thousands of anonymous people wander through your “store.” Not only do you know what items they bring to the cash register, you also can see more subtle actions: where they went first, what items they stopped to look at, what items they picked up for a closer look, and which ones they ignored. And that’s just the beginning. You know how long, and sometimes how carefully, they examined an item, whether they asked questions about the item and how much comparison shopping they did on your site.

To make smart decisions, you need to know more than just the sales patterns described here. You need to know two other metrics: exposure and appeal. Exposure - how often the item is browsed (looked at or viewed) Appeal - if the item is looked at, how frequently it is bought An item’s success = its Exposure + its Appeal

K E Y S TO U N L O C K I N G YO U R W E B M A R K E T I N G G E N I U S

For any given item that has an associated action, what percent of visitors act on the opportunity? That percentage is an indicator of something about the item’s appeal. - Chris Grant

Exposure can be calculated by dividing the total number of visits that view an item by the total number of visits to the site. For example, 40% of the visitors to your site may view Product A. Appeal can be calculated by dividing the number of actions taken (in this case purchases) by the number of visits to the item. Notice we’re only looking at those people who have already expressed an interest in the item by viewing it. Let’s call this measure of appeal the “browse-to-buy ratio.” A high browse-to-buy ratio indicates the item has high appeal. If an item is bought every time it is viewed, its browse-to-buy ratio would be 100%.

Beyond Sales Numbers: Non-Purchase Actions that Imply Interest The browse-to-buy ratio is a commerce metric but the concept is much broader than sales. A purchase is just one type of visitor action that implies some kind of interest. Other actions that indicate interest can include drilling down for a closer view or more information, downloading a screensaver, adding an item to a wish list, signing up for notifications, putting an item in a shopping cart and requesting a quote. Every one of these actions is worth measuring, because they reflect the effectiveness of the merchandising of an item and give you actionable insight about changing the merchandising. That’s why I more broadly refer to the metric as the “Browse-to-Act Ratio.”

You’ll have to use your judgment to interpret the ratios, because the metrics are just flags that help you focus your attention on two subsets of your items, those with the highest and the lowest ratios. Look for patterns. Are the items with terrific ratios displayed in a different way from those with poor ratios? Do they have bigger or brighter photos, are they less expensive, more appropriate to the season or perhaps displayed higher in a list? Don’t forget that drilling down can sometimes indicate a problem too. For example, a visitor may be frustrated by a lack of information in his first glimpse of the item. If you see too many drilldowns happening, you might be burying some important product information too deeply. Consider moving some of it up in the navigation if you find this to be the case. PUTTING AN ITEM ON A WISH LIST If you have a wish list (or even a tell-a-friend feature) on your site, you can calculate the ratio of wish-list-adds to item views. This gives you an even stronger indicator of an item’s appeal than just looking at drill-downs. But a wish list is, after all, a list of items that aren’t going to be bought just yet, implying that the items on a wish list may be enticing, but there’s apparently something that’s preventing an immediate purchase. Think about whether this is happening.

DRILLING DOWN FOR A CLOSER VIEW Do you show items in a list or thumbnail-view page and give visitors a chance to click for more information, a sample or a larger picture? Try calculating the ratio of drill-down visits to overall visits to the product-listing page. You’ll get an idea of whether the item’s title, description and thumbnail image are interesting enough to attract further investigation. Of course, the item’s intrinsic interest is also important—more people will probably click on chocolate truffles than on licorice no matter how good the sticky black stuff’s merchandising is.

I M P R O V E YO U R M E R C H A N D I S I N G B Y A N A LY Z I N G B R O W S E -TO - A C T R AT I O S

percentage is an indicator of something about the item’s appeal.

Taking it to the next level Ready for more? Once you know the browse-to-act ratio of an item, plus its exposure ratio (the number of visits to the item divided by the total visits to the site), you can categorize the item in a way that will lead to actionable insight into your web merchandising. I find the following 4-category scheme very helpful. Each category implies a different possible merchandising adjustment. The categories are Hidden Gems, Bread-and-Butter, Underperformers and Losers.

HIGH

ITEM’S BROWSE-TO-ACT RATIO

LOW

Underperformers This quadrant holds items that are very exposed but still convert poorly when they’re seen. Something’s wrong here. Maybe people are interested in the general idea of the item (they looked at it, after all) but something about it just didn’t close the sale. You may want to adjust the merchandising of these items, or perhaps abandon these but add other, more appealing items of the same type to your item line.

ITEM EXPOSURE

action, what percent of visitors act on the opportunity? That

Losers Try to move all your items out of this quadrant, which contains items that are rarely seen and, when looked at, just don’t convert to sales. Assuming these aren’t seasonal items, their very existence in your item line should be questioned.

LOW

terms of percentages. For any given item that has an associated

Winners You’d like all your items to be here. This quadrant contains items that are very exposed and also convert extremely well when they’re viewed. You may get even more return by giving these items more exposure, or you can capitalize on the demand by raising the price. You might also try to expand your item line by adding items that meet a similar need.

HIGH

ADDITIONAL RATIOS Hopefully additional ratios that relate to your web site are jumping to mind as you read this paper. Just remember to think in

Hidden gems This quadrant contains items with low exposure but which convert extremely well on those rare occasions when they’re looked at. Maybe these could be in the breadand-butter category if more people saw them. Feature them for a while, or make their description clearer so more people check them out.

LOW

PRINTING A PAGE An item’s browse-to-print ratio suggests strong appeal, perhaps indicating an intention to buy offline. For this reason, adding a “Print this page” button to your product pages, and making sure its clicks are measurable, can give you valuable information.

HIGH

BACK TO ITEMS A, B, C AND D Now we can make smarter decisions about items A, B, C and D with the new metrics. Here are their exposure statistics: -

Item Item Item Item

A is viewed in 40% of visits B is viewed in .5% of visits C is viewed in 40% of visits D is viewed in 2.5% of visits

Here are their browse-to-act ratios (% of visits that purchase divided by % of visits that view): - Item A: 25% (10/40) - Item B: 50% (.1/.5) - Item C: 2.5% (1/40) - Item D: 40% (1/2.5)

K E Y S TO U N L O C K I N G YO U R W E B M A R K E T I N G G E N I U S

ITEM’S BROWSE-TO-ACT RATIO

REQUESTING A QUOTE This usually applies to big-ticket or complex items such as services or insurance policies. It often indicates a very high level of interest, because the visitor knows that he is asking a human being, somewhere, to do some work on his behalf. The visitor is also committing to a future interaction with you.

ITEM EXPOSURE LOW

HIGH

ADDING THE ITEM TO THE SHOPPING CART The percent of item-views that turn into “cart-adds” is a significant measure of interest for any item. A cart-add action suggests the visitor wants to look at the product again, if not actually buy it. A high browse-to-cart-add ratio suggests that most characteristics of the item are satisfactory or close to it, including the detailed description, price, warranty, color, and of course, how the item is displayed. But if an item with a great browse-to-cart-add ratio also gets abandoned a lot, then you should be questioning what is preventing that final purchase step from happening.

New Way Of Evaluating

Conclusion

Old Way Of Evaluating: Sales (Action)

Item A

Best Seller 10% of all visits buy it

Poor Seller Item B

1/10 of 1% of all visits buy it

Item C

1% of all visits buy it

Item D

Medium Seller

Medium Seller 1% of all visits buy it

Exposure

Browse-toAct Ratio

High Exposure 40% of visits view it

25% of visits that view the item buy it

Extremely Low Exposure of 1% of visits view it

50% of visits that view the item buy it

High Exposure 40% of visits view it

2.5% of visits that view the item buy it

Low Exposure 2.5% of visits view it

40% of visits that view the item buy it

As you can see, browse-to-act ratios can drive smarter web merchandising decisions than simply relying on sales numbers alone. With web analytics, you can get the insight into visitor interest and behavior that will help you uncover the hidden gems and losers in your product line.

With the new information, we see that: - Item A is an Underperformer. This is ironic considering its stellar position on the sales list. - Item B is a Hidden Gem, not a Loser as one might surmise from the poor sales pattern. - Item C is also an Underperformer. - Item D is a Hidden Gem rather than a mediocre performer as implied by its sales numbers.

Taking Action These examples, while simplified numerically, were taken from a real-life gifting site.

ABOUT THE AUTHOR Chris Grant Enlighten

Here’s what the site owner did with the information. Items B and D needed more exposure to visitors. Item B, buried deep in an obscure item category, was subsequently featured on the homepage at an increased price—and its revenue increased dramatically. Item D, on closer examination, was felt to have a confusing name and a poor thumbnail picture. Improving both nearly doubled the viewing rate, resulting in nearly doubled sales. Items A and C needed better merchandising, not more exposure. Item A was changed to include an offer of free shipping. Item C’s descriptive copy was adjusted to emphasize quality and warranty. Both items’ conversion rates subsequently increased. Since the view rates were so high, the increase represented a substantial amount of revenue.

Chris Grant is senior data strategist at Enlighten, a web design and development firm in Ann Arbor, Michigan. She works on site analytics projects ranging from the routine to the decidedly unique and cuttingedge. Her deep background is in market analysis, human factors, customer satisfaction research and item development research, including a long stint in the product development department of a Fortune 1000 company. Chris has always swum in business waters despite five years in academia, and feels that web site analytics is the perfect convergence for her varied analytics career. She has a Ph.D. from The University of Michigan and is a member of the WebTrends Customer Advisory Board.

I M P R O V E YO U R M E R C H A N D I S I N G B Y A N A LY Z I N G B R O W S E -TO - A C T R AT I O S

Making Web Merchandising Decisions with WebTrends 7 WebTrends 7 provides very detailed merchandising reports making it easy to perform this kind of analysis. Overall the objective is to obtain visitor-based numbers reflecting views and interest actions for a list of your individual items. By opening the Product Drilldown report, a merchandising manager can drill into his complete product hierarchy. This allows a manager to view the performance of a product family, product category, product line and individual product SKU. The Product Drilldown report features visitor-based measures such as unique visitors, visits, and product views, along with interest metrics such as orders, cart additions and completed checkouts. All of this information is available in one report, so merchandising managers can easily compare metrics for individual products and subsequent roll-ups to see how metrics are trending across their entire merchandising mix. Once the Product Drilldown report is generated, a WebTrends SmartReport for Microsoft® Excel® can then be easily created with a few simple clicks, importing your entire Product Drilldown report into a fully formatted, presentation-ready Excel file. This is not a basic .csv file that many other web analytics vendors provide. WebTrends SmartReports for Excel is a unique capability that WebTrends offers, allowing WebTrends information to be further augmented with other business critical data or, in this case, to take advantage of Excel math functions to create calculations or additional metrics such as a browse-to-act ratio or a cart-add-to-act ratio. As an example, the browse-to-act ratio would be calculated by dividing orders by product views of a specific product. Then you can simply copy this calculation to all of the products and product categories for which you would like to use this metric. It's also important to note that once in SmartReports, you have a WebTrends toolbar embedded in the Excel environment. This allows you to easily change a date range or report settings, automatically updating the SmartReport without re-importing new data from WebTrends. Now with a few clicks you can instantly refresh your most important SmartReports, making it easy to build web analytics into your daily workflow. These metrics are powerful performance indicators of your merchandising activities. WebTrends 7 makes this process easy and provides many additional capabilities such as Visual Scenario Analysis and Visitor Segmentation to improve performance. For additional information, please view the WebTrends 7 Quick Tour on eCommerce and find out more about what WebTrends 7 can provide.

For More Information To learn more about improving your web site with web analytics, visit the WebTrends Resource Center: www.netiq.com/webtrends/resourcecenters.asp

Contacts WebTrends 851 SW 6th Avenue Suite 700 Portland, OR 97204 1.503.294.7025 1.503.294.7130 fax WebTrends Sales 1.888.932.8736 [email protected] For offices worldwide, please visit www.webtrends.com WebTrends is a business unit of NetIQ Corporation www.netiq.com WebTrends, the WebTrends logo, NetIQ and the NetIQ logo are trademarks or registered trademarks of NetIQ Corporation or its subsidiaries in the United States and other jurisdictions. All other company and product names may be trademarks or registered trademarks of their respective companies. © 2004 NetIQ Corporation, all rights reserved.

Based on WebTrends Take Ten Series © 2003 NetIQ Corporation, all rights reserved. WP10679RET MH0703

CS10551HOLM

MH 0403