New ways of measuring SEO performance

[danwhitehouse]

Recently I've been thinking of another way of measuring performance for an SEO campaign other than keyword rankings. We all know that the volume and rank of keyword equates to changes in volumes of traffic, but I don't want to use this model anymore. Here's why..

1. Keyword rankings are incredibly unstable

2. With the implementation of Search plus your world and other social search metrics, including G+ logged in, keywords are positioned depending on social value, rather than link value

3. Devalue of anchor text in links

4. Google shutting down ranking scrapers

So that leads me on to.. If we are in a world where we can't rely on keywords as a performance metric for SEO any more, what can we use? I know it might not be right now that keywords are becoming redundant, but there's nothing like being prepared.

For me, it's all about content, so maybe looking at the value of pages as the main KPI? But this again would lead back to the number of inbound keywords that page has..

Any thoughts?

[goodroi]

I like to simplify things so instead of constantly monitoring different metrics I normally only monitor one metric - Conversions generated by search referrals.

I'm collecting data on several different areas but I normally only look at that data when there is a problem and I need to diagnose what broke down.

[topr8]

totally agree with goodroi

... i've never been that interested in specific keyword rankings, i only care about one thing ... the sales we make, that is the only metric that matters.
aka:Conversions generated by search referrals. (in my case conversions are sales, in other cases i can see they could be something else like signups, whatever)

but i just work on my own ecommerce sites and always have ... i can understand how an seo company would want to use things like ranking as a measure of success though

[jo1ene]

I concur. I had a client that was all about getting the most "hits" and I could not convince her that it was all about people picking up the phone, sending the email, buying the book or whatever. It's better to get traffic from a few meaningful, targeted phrases than anything else - those are the ones that want what you're selling.

[superclown2]

Last year I realised that there was no way I was going to win the war against the big brands which are hoovering up every key phrase with ugly, plain vanilla doorway pages (I though G didn't like those?) in my sector so I set to work brightening up my sites with more interesting graphics and making them even easier to navigate. Result; a great increase in conversions and hopefully this will have a positive SEO effect eventually as G discovers that the sites are more sticky.

[Buts]

We realized this like 4-5 months ago. Now I'm checking my ranks like ones in a week. And when I'm checking I'm taking it as an average position. Anyway we (at my company) measuring our self just by phone calls.

This was just one way to deal with all the "black market SEO". Google has few more rabbits in their pocket.

[treeline]

If someone is hiring you as an SEO consultant you'll need a method to demonstrate your accomplishment. While actual sales are a beautiful metric, they aren't reliable for an SEO working with a business that has any other presence in the world. They'll work great for a website that is an island without other visibility.

Good press or an appearance on the front page of WidgetWorld Daily (unrelated to the SEO) could make you look great. Falling sales because of poor customer service, racist/sexist interviews by the site owner, or lack of inventory will make the SEO look incompetent.

An SEO Consultant needs to distinguish what their impact has been.

[Gmorgan]

We haven't been talking to clients about rankings for a while (though many refuse to give up on being obsessed about these).

The kind of things we prefer to measure are:

- Amount of traffic from organic results
- Number of different keywords bringing in traffic
- Quality of traffic (time on site, bounce rate, etc)
- Number of conversions (with a cost attached, i.e. our fee divided by the number of leads/sales)

[khaty]

The kind of things we prefer to measure are:

- Amount of traffic from organic results
- Number of different keywords bringing in traffic
- Quality of traffic (time on site, bounce rate, etc)
- Number of conversions (with a cost attached, i.e. our fee divided by the number of leads/sales)

I agree with this list, I can't just focus on the conversions if I know that I can still do more on our website like targeting different keywords and do some test by checking all these metrics.

[danwhitehouse]

I've always looked at number of keywords as a metric to define success, but it all spirals back to 'keywords' when this is what we're probably try to move away from. Maybe an option would be to define a trendline, which would be the visits and other factors such as conversions, which can be campaign successes. Either way, it's more than likely going to be something new and a combination of metrics.

[fom2001uk]

"Number of different keywords bringing in traffic" - this used to be one of my favourite metrics, but Google's Not Provided (increasing every month) has killed this.

[sandyallainseo]

There are several ways to judge the success of an SEO campaign. Most people consider keyword ranking to be the most important. While it's an important metric, conversion and engagement are even more important. However, by themselves no data point tells the whole story, so it's best to look at several factors that paint a clearer picture

[nandrews]

Some great points being made here - I have my SEO team focusing on engagement metrics for SEO performance these days which had been the case for us for around a year now. That is focusing on organic search conversions as an overall % and then as far as possible with not provided data report on non brand conversions for non brand SEO campaigns.

Looking ahead we are evolving this further towards measuring conversions by referral source and through attribution modelling. Rankings are certainly very much out of the picture now though as the earlier post mentioned, it's too unstable and ranking positions are a poor indicator of performance. A rounded SEO strategy that focuses onimproved content, better design and usability and highly targeted useful backlinks that might generate a lead is certainly the point of our focus and reporting.

[mcskoufis]

Use number of keywords & non-branded number of keywords growth, but then again the "not provided" is killing this metric.

What is also important for me is GWT data, like how many organic impressions, click-throughs, etc. I am taking a more "advertising" approach, trying to make the figures close to what they see for their display advertising reports.

For most of my clients what is crucial is organic exposure - visits, engagement, popular content and how specific other marketing activities have contributed to organic traffic (like TV commercials, Press Releases, etc.)

Agree that conversions are a key metric. After all SEO is (among other things) about getting the right message to the right audience, the ones looking to buy.

Have to admit I had a hard time convincing my clients to steer away from just keyword rankings as a metric the past few years :)