[Advice] Dealing with an immense URl structure full of canonicals with Budget & Time constraint
-
Good day to you Mozers,
I have a website that sells a certain product online and, once bought, is specifically delivered to a point of sale where the client's car gets serviced.
This website has a shop, products and informational pages that are duplicated by the number of physical PoS. The organizational decision was that every PoS were supposed to have their own little site that could be managed and modified.
Examples are:
- Every PoS could have a different price on their product
- Some of them have services available and some may have fewer, but the content on these service page doesn't change.
I get over a million URls that are, supposedly, all treated with canonical tags to their respective main page. The reason I use "supposedly" is because verifying the logic they used behind canonicals is proving to be a headache, but I know and I've seen a lot of these pages using the tag.
i.e:
- https:mysite.com/shop/ <-- https:mysite.com/pointofsale-b/shop
- https:mysite.com/shop/productA <-- https:mysite.com/pointofsale-b/shop/productA
The problem is that I have over a million URl that are crawled, when really I may have less than a tenth of them that have organic trafic potential.
Question is:
For products, I know I should tell them to put the URl as close to the root as possible and dynamically change the price according to the PoS the end-user chooses. Or even redirect all shops to the main one and only use that one.I need a short term solution to test/show if it is worth investing in development and correct all these useless duplicate pages. Should I use Robots.txt and block off parts of the site I do not want Google to waste his time on?
I am worried about: Indexation, Accessibility and crawl budget being wasted.
Thank you in advance,
-
Hey Chris!
Thanks a lot for your time. I did send you a PM the day after your original post, I will send you another :).
Thanks a lot for your additionnal advice. You're right about managing client's expectations and its crucial. You're pointing out some valid points and I will have to ponder about how I approach this whole situation.
Charles,
-
Hey Charles,
No problem, I've been out of the office most of the past week so I'm trying to catch up on a few of these now, sorry! I don't recall seeing any PMs either.
I feel weird to recommend shaving 3/4 of their site on which they put a lot of money in.
That's perfectly normal and I'd have the same reservations. If you do decide to go ahead with it though (and I'm absolutely not looking to push you into a decision either way, just providing the info) you can highlight the fact that paying a lot of money for a website doesn't make it inherently good. If those extra pages are providing no unique value then they're just a hindrance to their long-term goal of earning a return from that site via organic traffic.
It's a conversation we have semi-regularly with new clients. They think that because they just spent $20k on a new site, making changes to it is silly and a waste of the money they invested in the first place. "Sure it's broken but it was expensive"... I don't think search engines or users really care how much it cost
in the eyes of the client, it may come off as bold.
It certainly is bold and don't be fooled, there is a reasonable chance their rankings will get worse before they get better. In some cases when we perform a cleanup like this we'll see a brief drop before a steady improvement.
This doesn't happen all the time by any means, in fact we did a smaller scale version of this last week for two new clients and both have already started moving ahead over the weekend without a drop in rankings prior. It's really just about managing expectations and pitching the long term benefit over the short term fear.
Just be very careful in the way you project-manage it - be meticulous with updating internal links and 301 any pages that have external links pointing to them as well. You want to end up with a clean, efficient and crawlable website that retains as much value as possible.
You understand many sets of eyes are directed at them and a lot is to gain.
Also a very valid concern!
I'm probably not telling you anything you don't already know anyhow so don't think I'm trying to lecture you on how to do your job, just sharing my knowledge and anecdotal evidence on similar things.
-
Hey Chris!
Thanks for that lenghty response. It is very much appreciated and so is your offer for help. Let me check with some people to see if I can share the company's name.
[EDIT] Sent you a private msgOne of the reason I want to test the waters is, to be real honest, I feel weird to recommend shaving 3/4 of their site on which they put a lot of money in. I guess it comes down to reassuring them that these changes will be positive, but in the eyes of the client, it may come off as bold.
Another thing is, it is an international business that have different teams for different country. For more than 20 countries, they are the only one to try and sell their product online. You understand many sets of eyes are directed at them and a lot is to gain.
-
Hi Charles,
That's a tough one! I definitely see the motivation to test the waters here first before you go spending time on it but it will likely take less time than you think and either way, the user experience will be significantly better once you're done so I'd expect that either way, your time/dev investment would likely be viable.
I suppose you could block certain sections via Robots and wait to measure the results but I'd be more inclined to throw on the gloves and get elbow deep!
You've already mentioned the issues the current structure causes so you are aware of them which is great. With those in mind, focus on the user experience. What is it they're looking for on your site? How would they expect to find it? Can they find the solution with as few clicks as practical?
Rand did a Whiteboard Friday recently on Cleaning up the Cruft which was a great overview of the broader areas you can often trim your site back down to size. For me anyway, the aim is to have as few pages on the site as practical. If a page(s), category, tag etc doesn't need to exist then just remove it!
It's hard to say or to give specific advice here without seeing your site but chances are if you were to sit down and physically map out your website you'd find a lot of redundancy that, once fixed, would cut your million pages down to a significantly more manageable number. A recent example of this for us was a client who had a bunch of redundant blog categories and tags as well as multiple versions of some URLs due to poor internal linking. We cut their total URL volume from over 300 to just 78 and that alone was enough to significantly improve their search visibility.
I'd be happy to take a closer look at this one if you're willing to share your URL, though I understand if you're not. Either way, the best place to start here will be reviewing your site structure and seeing if it truly makes sense.
Browse Questions
Explore more categories
-
Moz Tools
Chat with the community about the Moz tools.
-
SEO Tactics
Discuss the SEO process with fellow marketers
-
Community
Discuss industry events, jobs, and news!
-
Digital Marketing
Chat about tactics outside of SEO
-
Research & Trends
Dive into research and trends in the search industry.
-
Support
Connect on product support and feature requests.
Related Questions
-
Question on AMP
I'd like to utilize AMP for faster loading for one of my clients. However, it is essential that this client have chat. My developer is having trouble incorporating chat with AMP, and he claims that it isn't possible to integrate the two. Can anyone advise me as to whether this is accurate? If it is true that AMP and chat aren't compatible, are there any solutions to this issue? I'd appreciate any leads on this. Thanks!
Intermediate & Advanced SEO | | Joseph-Green-SEO0 -
Canonical Chain
This is quite advanced so maybe Rand can give me an answer? I often have seen questions surrounding a 301 chain where only 85% of the link juice is passed on to the first target and 85% of that to the next one, up to three targets. But how about a canonical chain? What do I mean by this:? I have a client who sells lighting so I will use a real example (sans domain) I don't want 'new-product' pages appearing in SERPS. They dilute link equity for the categories they replicate and often contain identical products to the main categories and subcategories. I don't want to no index them all together I'd rather tell Google they are the same as the higher category/sub category. (discussion whether a noindex/follow tag would be better?) If I canonicalize new-products/ceiling-lights-c1/kitchen-lighting-c17/kitchen-ceiling-lights-c217 to /ceiling-lights-c1/kitchen-lighting-c17/kitchen-ceiling-lights-c217 I then subsequently discover that everything in kitchen-ceiling-lights-c217 is already in /kitchen-lighting-c17 and I decide to canonicalize those two - so I place a /kitchen-lighting-c17 canonical on /kitchen-ceiling-lights-c217. Then what happens to the new-products canonical? Is it the same rule - does it pass 85% of link equity back to the non new-product URL and 85% of that back to the category? does it just not work? or should I do noindexi/follow Now before you jump in: Let's assume these are done over a period of time because the obvious answer is: Canonicalize both back to /ceiling-lights-c1/kitchen-lighting-c17 I know that and that is not what I am asking. What if they are done in a sequence what is the real result? I don't want to patronise anyone but please read this carefully before giving an answer. Regards Nigel Carousel Projects.
Intermediate & Advanced SEO | | Nigel_Carr0 -
Pages with URL Too Long
I manage a number of Shopify stores for ecommerce clients. MOZ keeps kindly telling me the URLs are too long. However, this is largely due to the structure of Shopify, which has to include 'collections' and 'products'. For example: https://domain.com.au/collections/collection-name/products/colour-plus-six-to-seven-word-product-name MOZ recommends no more than 75 characters. This means we have 25-30 characters for both the collection name and product name. VERY challenging! Questions: Anyone know how big an issue URLs are as a ranking factor? I thought pretty low. If it's not an issue, how can we turn off this alert from MOZ? If it is an issue, anyone got any ideas how to fix it on Shopify sites?
Intermediate & Advanced SEO | | muzzmoz0 -
Changing URLS: from a short well optimised URL to a longer one – What's the traffic risk
I'm working with a client who has a website that is relatively well optimised, thought it has a pretty flat structure and a lot of top level pages. They've invested in their content over the years and managed to rank well for key search terms. They're currently in the process of changing CMS and as a result of new folder structuring in the CMS the URLs for some pages look to have significantly changed. E.g Existing URL is: website.com/grampians-luxury-accommodation which ranked quite well for luxury accommodation grampians New URL when site is launched on new CMS would be website.com/destinations/victoria/grampians My feeling is that the client is going to lose out on a bit of traffic as a result of this. I'm looking for information or ways or case studies to demonstrate the degree of risk, and to help make a recommendation to mitigate risk.
Intermediate & Advanced SEO | | moge0 -
What's the best URL structure?
I'm setting up pages for my client's website and I'm trying to figure out the best way to do this. Which of the following would be best (let's say the keywords being used are "sell xgadget" "sell xgadget v1" "sell xgadget v2" "sell xgadget v3" etc.). Domain name: sellgadget.com Potential URL structures: 1. sellxgadget.com/v1
Intermediate & Advanced SEO | | Zing-Marketing
2. sellxgadget.com/xgadget-v1
3. sellxgadget.com/sell-xgadget-v1 Which would be the best URL structure? Which has the least risk of being too keyword spammy for an EMD? Any references for this?0 -
URL Structure Question
Am starting to work with a new site that has a domain name contrived to help it with a certain kind of long tail search. Just for fictional example sake, let's call it WhatAreTheBestRestaurantsIn.com. The idea is that people might do searches for "what are the best restaurants in seattle" and over time they would make some organic search progress. Again, fictional top level domain example, but the real thing is just like that and designed to be cities in all states. Here's the question, if you were targeting searches like the above and had that domain to work with, would you go with... whatarethebestrestaurantsin.com/seattle-washington whatarethebestrestaurantsin.com/washington/seattle whatarethebestrestaurantsin.com/wa/seattle whatarethebestrestaurantsin.com/what-are-the-best-restaurants-in-seattle-wa ... or what and why? Separate question (still need the above answered), would you rather go with a super short (4 letter), but meaningless domain name, and stick the longtail part after that? I doubt I can win the argument the new domain name, so still need the first question answered. The good news is it's pretty good content. Thanks... Darcy
Intermediate & Advanced SEO | | 945010 -
E commerce canonical links: include category structure?
I have a client on shopify. All categories have correct canonical links. however, the links from all menus, category pages, etc. follow this structure: /collections/COLLECTION_NAME/products/PRODUCT_NAME but the canonical link on the above product url is: /products/PRODUCT_NAME I have a feeling this is hurting our product detail page's seo. Our collection pages are ranking fine, but for some reason the detail pages aren't. It could be that they are deeper, but I am trying to make sure nothing big is causing it first before I get into the smaller factors. Any best practices on this?
Intermediate & Advanced SEO | | no6thgear0 -
How important is it to canonicalize mobile URLs to desktop URLs?
I know many SEO's prefer a stylesheet and single URL, but if you use m.domain.com, do you canonicalize to your desktop URLS?
Intermediate & Advanced SEO | | nicole.healthline0