With the end of the year upon us, I wanted to review my team’s 2012 performance to inform our approach for 2013 and set ourselves some targets. As we’ve been preaching to clients that good content is honest content, I thought why not post-mortem our performance in public and let the data serve as a vehicle to describe how we’ll be approaching SEO in 2013?
So with a little trepidation (I don’t remember an agency going public with so much data previous to this) here’s all of our 2012 ranking data: the good, the bad and the indifferent!
- Ranking for 1655 keywords across 14 clients in 3 different languages (PL, DE, UK) on day 1 of the year / engagement, vs. ranking on last week of November.
- Data covers a total of 376 weeks (not all clients have been with us since Jan 1st, and for most we migrated platforms part-way through the year)
- Monthly search volume (exact) across all keywords was 7,738,237 per month
- Across all keywords, client average rank had gone up 12.9 positions since the start of the year / engagement.
- Our client’s visibility had risen in more instances (against 33% of keywords our clients saw a positive rank change) than decreased (we only saw a negative rank change against 9% of keywords)
- Our clients appeared in the top 3 positions on SERPs more often. Across all clients, on week 1 of their engagement we ranked in the top 3 positions against 247 keywords, at latest count our clients rank in the top 3 against 352 terms, a rise of 42%.
- The number of keywords for which our clients do not rank has fallen dramatically: From 1028 keywords down to 789, a fall of 23%.
Not bad, especially when you consider that most of our clients tend to be very large corporate brands. Sometimes just getting a project underway and optimised content onto the website is a major challenge, so to see such positive figures proves to me that our practice is going in the right direction.
We are driven by the following 2 statements:
To rank #1 for a keyword, you must deserve to be above all other relevant sites; you must own the best piece of relevant content on the web.
SEO happens when you create great experiences for BOTH users AND search engines …and all of our tactics reflect these truths.
- We have pushed clients to move beyond basic optimisation and build content that adds value to people in-market for their products / services
- We have built hundreds of relationships with website owners in 2012, with the output being high quality, relevant links for our clients.
- We try to base content on what has proved to be successful for other brands and websites, rather than just having a default “build a landing page about the keyword” approach. Given these results, we will be sticking with this broad approach over the next year.
- For 9% of all our keywords (152 terms), ranking fell. The total audience for these terms was over 862,350 exact match searches per month.
- For 21 terms (1.3% of all keywords) we had dropped out of the top 100 results altogether.
Losing visibility for nearly 10% of all tracked keywords really is a major concern. This decline wasn’t concentrated on one particular client or keyword area, and none of our clients were penalised last year by Google, so I am satisfied that our efforts to convince other website owners to link to our client’s content are not at fault.
To me, this is a case of keywords slipping through the cracks because our keyword lists are too long. In 2013 we are going to focus on producing keyword lists that only contain terms we believe tie into our content plans and that will actually impact on client performance. You can see my posts on a new, fairer method for keyword research here.
Around 12% of our terms rank in the same place now as they did when the respective engagements began: the vast majority of these are brand terms or generic keywords where we have always ranked #1.To me the real story comes in the number of keywords which haven’t ranked during 2012. Of the 1655 keywords we have been tracking, our clients remain unranked for 757. That’s 46% of our total keywords. Major problem. Based on the idea that “you will rank #1 when you have the best piece of relevant content on the web” we quickly established 2 truths about the majority of keywords that had not ranked all year:
- Keywords were unrealistic for our client given their current web assets. E.g. one client asked us to track “save time” because there were some very specific timesaving tips to do with a key product on their website. In reality there is no way a specialised piece of content would deserve to rank #1 for a very general term.
- Keywords were relevant for the client but not present on the website. E.g. one client wanted to rank for “cheap cars” but refused to use the word “cheap” on their website. Another common flaw was having “review” keywords in our keyword lists but no review-based content on the actual website.
To improve performance, these are two problems our agency has to tackle in 2013. I suspect that amongst the SEO industry we are not alone.We want to retain the idea that SEO is about creating content which will earn its own distribution rather than SEO being a group of people building links to crap pages. We also want to do better at setting client expectations on the keywords it could realistically expect to rank for. As a result, we believe SEO in 2013 is going to be about the following:
- Base work on content initiatives, not landing pages. A content initiative is a project that revolves around the development and publication of a resource. The key point is that this resource adds a ton of value to people in your target market. E.g. for a financial services client you might build a free PDF eBook which covers every aspect of setting up an online business, with useful tips and recommendations for each step of the process. Using this resource as an anchor, you can develop a full distribution plan in advance of publication, then execute upon launch. When a great piece of content and a distribution plan come together, you have a content initiative.
- Base engagements on shorter, more focussed lists of keywords. In order to rank #1 we almost need an individual content strategy for each keyword, and so having 200 keywords per engagement is just unrealistic; major corporate clients are not going to cope with that many new initiatives. Fewer keywords = easier to prioritise where to begin less chance of inaction.
With regard to the second point we have already completely revamped our keyword research process, and now we are moving towards a point where we have a full process in place for planning and executing content initiatives. Whilst looking at our entire dataset for ranking was a bit traumatic, and there were undoubtedly some harsh truths to take on board, I’m glad we did it because it has reinforced our core principles, and helped us identify the most urgent areas for attention in 2013. As an SEO team we can accept that you aren’t always going to improve ranking for every single term, but we’re determined that we’ll see more action and better results next year. It'd be great to hear what you guys are doing to update and adapt next year – see you in the comments!
Twitter Canada MD Kirstine Stewart to Keynote Toronto
ClickZ Live Toronto (May 14-16) is a new event addressing the rapidly changing landscape that digital marketers face. The agenda focuses on customer engagement and attaining maximum ROI through online marketing efforts across paid, owned & earned media. Register now and save!