In 2014 I wrote that Domain Authority doesn’t work for link acquisition. A couple of weeks ago I read this tweet from Mike King, agreed, and wanted to expand on it.
None of our metrics are “real.” They are all estimations or extrapolations from what we think Google is doing. Use em or don’t. The end.
— MyCool King (@iPullRank) December 22, 2016
Metrics vs. KPIs
DA doesn’t work as a KPI for link acquisition. As a metric to assess the worth of a site it’s fine. Flawed – but fine. And no disrespect intended to Moz here: as Mike points out, all metrics that are trying to do what DA does are flawed, so I’m using “Domain Authority” as a proxy for link metrics generally.
Domain Authority works when it’s used to assess the relative value of websites. Here are the reasons why, for me, DA can’t be used as a KPI:
- Domain Authority can be gamed so easily (read the previous post) which means it must be combined with a strict QA (quality assurance) process.
- No QA process with any substance will leave us an abundance of high DA websites to go after – there are so many sites selling links, for example, that we don’t want linking to us – which makes reaching a KPI consisting of a number of links meeting specific DA criteria hard to reach.
- The QA process, therefore, is non-negotiable. Potential linking sites either meet our criteria, or they don’t.
- DA on the other hand is a variable applied to the sites that have passed an agreed QA process.
Choosing a metric
So, for me, here’s where you should spend your agency hours (again a catch-all, as I’m using in-house link building teams if you, the SEO stakeholder, are responsible for their KPIs):
- The link builder compiles a list of sites and individuals that are relevant to the brand. We’re looking at audience; subject matter; sites competitors have links from and we don’t…whatever the strategy is.
- Each site is subjected to QA.
- The remaining websites are all are prospective link targets and if we’re all honest with each other you want links from every one of them. But because resource is limited we need a method of prioritisation: metrics! DA – fine. TrustFlow – fine. Make up a metric – no problem. We prefer to prioritise based on our best relationships because there’s a better chance of bigger links. Generally we’d also prioritise based on audience – many blogs or sites with the exact right target audience (potential buyers) don’t have the right DA or PR etc. right now, but we want to build a relationship with them for the future.
Our relationship with each of our clients is different. Relationships are often better when we’ve discussed:
How we’re paying for the work. Most (but not all) of Branded3’s link building is done as part of a retained agency/client relationship which means that a) these links aren’t the only links we’ll build and b) link building isn’t the only thing we’re going to do. We agree KPIs (often traffic forecasts) at the beginning of a relationship and keep them updated which means the KPIs we agree for our link building efforts will only ever be secondary KPIs. The main aim of any campaign is to meet our SEO/sales targets – not to deliver a specific number of links.
Managing expectations. We need to be realistic about the links that can be acquired (mainly based on how much time we’re spending) and we need to be realistic about what effect those links are going to have. Link building is hard and if there are more important things for us to be doing (on-site) and we can’t – or if we’re going up against a link building behemoth like Amazon – the difference between 10 and 20 links is pretty negligible in the scheme of things.
How we’re going to prioritise. Link building might be the right thing to do now, but priorities can and should change as the market and your competitors do. It isn’t anybody’s job to get more and better links – we want more and better results and we need to constantly tweak our SEO strategy and tactics accordingly.
Forecasting. Decision makers want to know why we’re building the links we’re building: quantities and choice of sites. Because we supply and update monthly traffic and revenue forecasts for almost all of our clients we make it absolutely clear that success looks like the numbers we’ve already provided – we forecasted based on being able to do what we need to do. We need the conversation to be: if we do this activity we’ll get this much traffic. This is our primary KPI for the campaign.
Setting KPIs for the number of pieces of coverage we get, the number of links we drive or, say, the amount of social shares we get or specific rankings isn’t always helpful. The coverage, links and rankings in themselves are not the results we’re looking for. What we want is to ultimately increase brand awareness, traffic to the website, conversion and sales. These are the things that actually make a difference to the bottom line and business objectives. You could get all the coverage or links in the world, but if nobody sees it or takes action as a result of it, then it’s not going to make an impact. Likewise, obsessing over ranking for a particular keyword is pointless if it doesn’t drive any traffic or it drives traffic that doesn’t convert.
This is actually covered in the Barcelona Principles from a PR point of view, which are:
- Importance of Goal Setting and Measurement
- Measuring the Effect on Outcomes is Preferred to Measuring Outputs
- The Effect on Business Results Can and Should Be Measured Where Possible
- Media Measurement Requires Quantity and Quality
- AVEs are not the Value of Public Relations
- Social Media Can and Should be Measured
- Transparency and Replicability are Paramount to Sound Measurement
This covers two of my key points – measure outcomes, not outputs and look at the effect on business results, not tactics. That’s been out there since 2010 and yet, five years on, it’s still being ignored. You can read more about the principles on the AMEC site.
How to KPI link acquisition
To build links we need to use budget – and obviously if we’re using budget we want to know what success looks like before we start.
- Choose an objective. Why are we going to try and acquire some links? Typically our objectives are visibility in search engines > traffic > conversions > $$$$. Based on our assessment of the competitive landscape we’re confident that more (better) links are what’s required to achieve this.
- Scope the work. What are we going to do to meet that objective? What value are we going to add that we think will deserve some links?
- Then choose our KPIs. How will we measure whether the work has been a success?
Remember our objective is visibility, traffic, conversions, $$$$. So why aren’t these always our KPIs?