Goooood Morning. Very interesting issue this week - seems like everyone has been publishing knockout content recently. Before we move onto the links, I want to discuss/debate/rant about the subject of one them 'Forecasting SEO Impacts on Organic Traffic'. First point is that the post is one of the most pragmatic approaches to forecasting for SEO I've read so check it out! Second point is that people rarely follow this approach, which means most forecasts are absolute garbage.

There are a couple of reasons forecasting in the industry is poor. I think they're often forced at pitch stage with objectives set as part of the brief which can often force SEOs to string together a set of loose projections (not an excuse - don't do it). In truth though, we simply aren't very good at them. Forecasting is tough, there are so many variables, however we should at least be able to make a reasonable baseline forecast using historical data - forecasting on top of this should be justified by data, but caveated to the hilt.

The main problem I see, is that there's not enough due diligence carried out on historical data to make a reasonable prediction. If you're calculating a base trend, you need to ensure that firstly there's a large enough dataset to work with. You then need some context from the client on historical marketing activities and whether they will activated in the forecasted time period. Getting this part right will set you up for success. I'd love to hear your thoughts on forecasting, as I may follow up with a blog post. Let me know on Twitter!

Andrew Charlton

Post of the Week