Managing a Large E-commerce Website

When it comes to e-tailing (commodity based e-commerce), the first name that comes to mind is Amazon. Amazon maybe the largest e-commerce website in the world today, but there are websites that have more than a million products distributed across multiple catalogs who forecast themselves to grow like Amazon Inc. in their respective geographical areas. These websites can be classified as horizontal e-tailing websites.

It is a challenge to-

1. Handle so many products (product pages, facet pages, listing pages, landing pages)
2. Handle multiple niches (fashion, gadgets, books, movies, gifts, jewelry, etc)
3. Coordinate with different departments and business folks

SEO takes a completely different dimension when you consider such a website. Normally, these large websites have a decent amount of buzz going around so, authority is not a big problem. SEO is no longer about "gaining links." SEO is transformed into a real inbound marketing team.

Challenge #1: Getting a buy-in for SEO.

Most companies are built on the shop/outlet ideology. People start off by thinking that like any physical outlet, they just need to set shop and that’s all. That leads to a business biased hierarchy. After the euphoria and the initial press mentions die out, they find that the traffic is tabling out and/or they are spending too much on paid traffic. That’s when the need for a SEO professional is felt. Even if the need for a SEO is acknowledged, most people in the hierarchy are unaware about how SEO works and what resources are required. If an organization does not have buy-in for SEO right at the top, it's very hard to get anything done.
The Solution: Education. You heard that right. Education is the most effective way to get your voice heard. I have found that ROI analysis works particularly well when you are asking for resources or pitching for radical changes to the website.

Challenge #2: Horrible information architecture.

This is mostly due to the points mentioned in #1. Business folks do not fully understand user perception and how users behave when they are on an e-commerce website. (Frankly, they have much more important things to look into.) The result is total chaos when it comes to information architecture.
The solution: Get your hands dirty. It may take a bit of time to fully study the extent of the change that you may need to do, but ultimately, only a SEO would know how to do a keyword research (apart from the SEM, of course). Google keyword research tools like Google Trends and Google Traffic Estimator become your friends, and it's on you what you propose. The categorizations are pretty obvious. Keywords with highest volume can be assigned as categories and sub-categories.
Pro Tip: If you have a usability team in the organization (or among your friends), ask for their help. A second pair of eyes help a lot.

Challenge #3: Getting the technical implementation done.

Almost every organization has a tech team, which is too aloof from the rest of the team or is just too busy working. Getting a buy-in ensures that you have some bandwidth assigned to you. Implementation takes quite a bit of time considering the constraints on the application as well as the infrastructure level.
The Solution: Every organization has its unique set of problems. Although a weighed trade-off may be required, the idea is to evolve and adjust to the condition. Infrastructure problems are painful, and they tend to persist for a long time. You may find yourself in a situation where you would ideally want some changes, but the infrastructure is not tuned to handle these changes. There are two approaches you can take:

1. The no B.S (forgive the adjective) path. You may end up having some burned bridges, but in the end it gets the work done.
2. The personal approach. You are not the subject matter expert (SME) here, so it’s best to understand the constraint and genuinely try to work around the problem with the SME.

I favor the second approach since most of your changes depend heavily on infrastructure.

Challenge #4: You cannot apply the same formula everywhere.

Gadgets have a very different set of touch points as compared to fashion. People may not want to see a larger image for a MicroSD, but they definitely want to do that for a pendant or a shirt. Therefore, it's important to think at two levels: micro and macro. The success of your planning would depend on how well you marry these two concepts.

Challenge #5: Classical Enterprise SEO challenges.

Namely:

1. Attention to all product pages.
2. Huge amount of duplicate pages.
3. Unique content on all product pages.
4. Dealing with product life cycle.

#1. Attention to all product pages: When you are talking about millions of products, it becomes virtually impossible to focus on all of them. The idea is to choose a manageable set at a time. This set can be derived from market demand, business focus, and page quality. Based on these 3 factors, you can zero in on 20,000-30,000 products. The idea is to interlink intelligently.
#2. Huge amount of duplicate pages: It may be a good thing to have millions of pages indexed, but you must ensure that all possible duplicate pages are blocked. If you can identify patterns, it is best to use the robots.txt to block these. Identify duplicate pages and point them to your focus page using a “rel=canonical” tag.
#3. Unique content on all product pages: This is where it gets really tricky. Either you need a battery of content folks who can keep churning truckloads of content every day or lots of freelancers doing that for you. The problem would be that you would have to ensure that the content is user-friendly and adheres to your quality standards. This is sadly a manual process, and there are no short cuts to it. The best way I have found is to recognize focus areas and then prioritize them according to that. Insights from business, product, and management make this task a lot easier.
#4. Dealing with product life cycle: In e-tailing, there will be cases where products come and go. A classic example would be a fashion catalog where the catalog changes with season and you cannot find the same or even similar products in the future. Same is the case with gadgets where better versions come up and older models are discontinued. Example Apple iPhone 3G -- this phone was a craze when it launched, but it died out with the release of newer versions of the phone. Apple stopped selling this product altogether.
There are two theories that are in practice to deal with product life cycle. The first theory is to persist with the page with an upfront message that the product is permanently discontinued, and there are links to the current offerings in the same category. This theory offers a chance to up-sell and helps in retaining the SEO authority of the page. In my opinion, this approach has a downside as well. With time, the query volume of that product (and hence the page) diminishes, so you get less traffic. So, you are left with doing the same stuff over and over again for newer products. The second theory is to redirect the page to its most appropriate category with a clear message to the user that they have been redirected and the reason for the redirect. I have seen that this works better as we transfer some of the link juice to a current page rather than doing repetitive work.
Okay, now let us come to the interesting part. Based on my experience, I have put down some high level actions (like guest posting, social, SEO, etc.) and how they fare in terms of a large e-commerce website.

Great I was hired by a large e-commerce company. What now?

Well, this is just the beginning. Every organization is different and has a different style of working. You must start by “careful recon.” (That’s how I like to call it.) No one in any organization will kill you for taking a couple of weeks to settle in. Be personable and meet up with folks who would be working with you (apart from your team). Understand what has been done, how much buy-in do you have, if any, from the management. Study the website(s). Understand the industry and competition. Ask around for help and examples.
Being a SEO in a large organization is just not SEO, it's product development, project management, and usability research with SEO. So as Jonathan Colman said in his blog post: Evolve.
Present a plan and execute it well. The numbers will follow. Last, but not the least, make the website user-friendly and let the bots do their own work.
For those aspiring folks who wish to work at a large e-commerce company, learn. Build skills that make you more than just an average SEO. Read stuff about SEO on a daily basis. Gain new skills that will help you make the cut. Focus on your strengths and fight your weaknesses. Be patient, one day, you will be good enough to be picked up by one such organization.
That’s it! I look forward to hear from you all in the comments. Cheers!


Reference :- http://www.seomoz.org/ugc/managing-a-large-ecommerce-website

EBriks Infotech :- SEO Company India