The Pursuit of Perfect Product Data

I’ve spent the majority of my adult life working with distributors large and small on their product data. Over the last 5 years or so, there has been an enormous emphasis placed on product data in support of ecommerce. I think everybody would agree that product content is extremely important to the success of an ecommerce site, particularly when your business is product distribution. In fact I would argue today’s distributor is fundamentally a data company who happens to ship products out the back door. But what is perfect product data, how does one know when they have it? Generally we find that nobody ever thinks their content is very good, but their reasoning tends to be very analog. Often we hear statements like, “we don’t have as much data as Grainger…”, or, “ Ferguson has more videos on their product pages…”, statements that may very well be true, but they are not very helpful when trying to evaluate the effectiveness of the data, or understanding where you are on the path to perfect product data. Relying on subjective characterizations like these to assess your product content is a flawed approach that quickly leads to frustration and an expensive random process of improvement. We need a more analytical thought process; after all the product content is not there for entertainment, it’s there to inform, sell, and to support. The data you need to make informed decisions is there, you just need know where to look. In this post we’ll suggest just such an approach.

Begin by determining what data really matters. One way to figure that out is to think about it in terms of the customer purchase journey. Think of the journey in terms of Discovery, Knowledge, and Conversion – that is finding the product, learning about the product, and buying the product, really no different than when your customers come into to your shop today. The difference online is that there isn’t a counter person to help them, so the data needs to fill that role. Data such as part numbers, descriptions, copy, features, videos, images, etc., need to support that customer purchase journey at each step. Thinking about Discovery, Knowledge and Conversion from the customer’s perspective is a great way to begin to identify what data you will need to be successful.


Can your customers find the products they are looking for? They can’t learn what you offer if they can’t find it. Put yourself in their shoes and ask yourself how they would go through that discovery process. Then identify the data that supports that, such accurate descriptions, customer part numbers, slang or regional terms, cross reference data, etc. A challenge here is that many people source their data through some industry data service, or 3rd party content developer. All very good starting points, but by definition they are somewhat generic, they cannot consider how your customers shop, they don’t know them. Do your customers buy soda or pop? Fitting the product data to your customer is your job.


OK, they found it, now is there enough information to educate them. Can they tell this is the product for their application? Do they understand why it’s better for them than whatever alternative they may be considering? The data here is going to vary greatly by the type of product. For example, a customer is going to need a lot more information about a hydraulic motor than they do about a pipe fitting. It’s easy to go overboard here, and developing this type content can be quite expensive, so remember you are trying to provide the information necessary to support your customer’s purchase journey nothing less, but not much more. After all, a video demonstrating the proper use of a sheet of sandpaper probably doesn’t advance that objective very much.


Can they buy it from you? Is the necessary information there like buying units, bundles, accessories, etc., to enable them to complete the sale? After all, the way you make your living is you sell stuff. Again, think from your customer perspective, what do they need to complete your purchase for the product type in question.

We’ve used the customer journey to identify what data we need, let’s keep that process going use the analytics to indicate how well we are doing. We’re going to focus on hits and conversions, something you’re probably already looking at. The figure below illustrates a model for grouping products into 4 quadrants, because we all love a good quadrant model. Conversions are along the x-axis and hits along the y-axis. By looking which products fall into which quadrants we gain very valuable, objective insight into how “good” our product data really is.

Lower Left (low hits, low conversion)

Customers have to find the product before they can buy it.  Products in this quadrant either aren’t very useful to your customers, or are simply not being found.  We need to look at the searches our customers are running.  What searches lead to no results, or result pages where the customer goes no further.  We also need to look at the data the search engine uses, such as part numbers, descriptions (short, long, marketing, etc.) keywords, features, etc. when trying to match a customer query.  Think about your customers, not the industry at large.  Put yourself in their shoes – How do they search for the products that just don’t seem to be getting hits?  For example, do they search for a “4in Electrical Box”, or do they search for a “4 Square”, i.e., Soda or Pop?  Try reworking the data to more closely align with how your customers think about the products.  The emphasis here should be to get the hits up first, then follow-up on conversions once you have the additional experience from the increased hits.

Upper Left (high hits, low conversion)

These products are being found, with decent frequency, but for some reason the customer is just not purchasing.  There may be several reasons for this related to the data.  First the descriptions may be misleading, i.e., the product isn’t what the customer thought it was.  Look a bit deeper and examine the search terms customers are using to get to the product.  If there is a disconnect between what customers are using, fix-it, most likely in one or more of the descriptions.  Another issue may be that there is simply not enough data for the customer to confirm this is the product they want.  Remember the journey – when they hit the product detail page they are at the Knowledge stage.  You need to give them sufficient data to confirm this is indeed the product that is right for them.  You may need more specifications, documentation, product FAQs, or digital assets such as images or videos.  Caution, this isn’t about volume.  Think about the product, and how your customer uses it, and make sure you are satisfying their need for the right Knowledge.  For example, a video of a person sanding a tabletop with a piece of sandpaper isn’t going to help me make a purchase decision if you haven’t told me the Grit, or how many pieces come in a package.

Lower Right (low hits, high conversion)

In this quadrant products sell – when we can get customers to the product detail page.  Similar to the first quadrant we looked at, it could be that very few customers use these products, or we’ve got an issue with descriptions, keywords, features, etc., i.e., the data being used by the search engine.  Again, we want to focus on driving the hit rate up by aligning the data to customer search tendencies.  If you’re stuck for ideas, survey your customers, or counter employees, ask what terms they use when searching for those types of products.

Upper Right (high hits, high conversion)

Nirvana, the sweet spot.  For these products we have the right data to support the entire customer purchase journey.  Many might be tempted to leave these alone and focus in the other quadrants, but if you do, you are missing an opportunity.  These products are highly desired by your customers and are getting high visibility on the site, they are the perfect candidates for aggressive product associations such as up-sells, cross-sells, accessories, etc.  Spend your efforts building those relationships to see even greater revenue opportunity.

Most everyone believes they need perfect product data, the problem is that they don’t know what it is, nor do they have a process for constant refinement of it.  In most companies the creation and management of product data is frustrating and expensive with no real way to measure effectiveness. We believe that has to be more objective and targeted for results.  Where do you start, what actually needs to be done?  We’re suggesting that perfect product data is largely defined in the context of your customer’s needs, so you need to come at the problem from that perspective, specifically:

  1. Recognize that this is not a beauty contest, don’t get caught up thinking about your data that way, it serves a business purpose;
  2. Consider your customer purchase journey, and
  3. Measure results, let the numbers guide your efforts and areas of investment on the path to perfect product content.

Related Articles