Forecasting Seasonality for SEO Campaigns

by Mike Tekula April 7th, 2010 

If you're selling SEO services, or any marketing service for that matter, you're selling the results, not the service.

And if those results are negative, you'd better believe it'll reflect negatively on you.

So what happens when something outside your control takes traffic down when your job was to bring it up? You can try to explain yourself at that point and look for those external factors, but this is backpedalling (and it never looks good even when you're right).

Take a look at the graph below. It's taken from one of my clients' Google Analytics profile, and it shows the inbound non-paid search traffic. The time period here is one in which absolutely nothing was done on the site. No new pages, no edits at all.

seasonal-ebbs

All over the place. Valleys and peaks.

Shifts in the marketplace, in the behaviors of the users who search, happen constantly. To call search traffic a moving target is a hell of an understatement. It's in perpetual motion.

But much of the time these trends follow seasonality. We're not talking the obvious here – like the fact that searches for "flowers" spike before Valentine's Day and Mother's Day. Many other verticals follow seasonal patterns, and if you're about to enter one, whether with a new website of our own or by way of taking on a new client, it's important to get the lay of the land and avoid surprises wherever you can.

Tools like Google Trends and Google's Keywords Tool give you ideas for single keywords, but when you're talking about all of the search traffic to a given website, including all of those keyword branches, seasonality rarely follows the pattern of a single keyword.

A look at the vertical overall is much better. Here's a good way to use Google's Keywords Tool to pull seasonal data and end up with a clearer big picture of seasonal trends in a niche.

1) Get a solid list of keywords together for the niche

I like to use somewhere in the range of 100 keywords – both the short head and long tail terms.

Obviously we're still not getting the absolute complete picture because those ultra-long tail keywords that see searches in the low single digits don't show up in the results of this tool. But keyword research and SEO forecasting have never been, and never will be, precise sciences. This is more like weather forecasting: you can get close, but there are always surprises.

2) Drop them into the Google Keywords Tool

Paste your list into Google's Keywords Tool and hit the "Get Keyword Ideas" button.

This tool unfortunately doesn't maintain your exact list or its order. It pulls in synonyms, drops keywords for which it has no data and mixes things up. But for this exercise we aren't looking for precision. We're interested in an overall picture of the search traffic for this niche.

By default this tool won't show seasonal data, but the drop-down menu at the top-right will allow you to enable the column for "Search Volume Trends:"

show-seasonal

For this exercise I usually also hide the "Advertiser Competition" and one of the Search Volume columns – it keeps things simple.

Here's what you end up with:

mountain-bikes-seasonal

So you're seeing some seasonal data already, but it's not terribly useful in this format if we want to look at the seasonality across the niche. We need to pull this out to Excel to mash it up ourselves.

3) Pull the CSV for Excel

Luckily Google's Keywords Tool allows you to do this with a single click. Just scroll to the bottom of the first keyword list and click the ".csv (for Excel)" link.

export to csv

Once you open the list in Excel you'll probably want to do a little formatting to make the data easier to look at.

What's helpful here is that the graphical trends column you see above, in the Keywords Tool display, is exported numerically. Each cell gets a decimal value from .00 to 1.00, which reflects the relative search traffic for a given keyword during that month.

4) Generate averages for each month column

The easiest way to proceed is to simply highlight all of the monthly columns for all keywords, including the last blank row beneath the last filled row:

highlight-rows-seasonal

Then use the Average AutoSum option to create average monthly levels for all keywords:

create-avgs

Note: I'm aware this isn't going to give us a high level of statistical accuracy. But again, this isn't an exact science – we're looking for trends.

5) Graph it

One of the nicer aspects of working with Excel the ease with which you can create graphs from data.

Here's the graph I created in a few seconds for the "mountain bike" niche:

graph-it

As we might anticipate, this niche shows increased activity in the warmer months of the year. But February is an odd high point between the relative lows of January and March.

Let's say, for example, that you've taken on a client in the "mountain bike" niche. If you're launching an update to the site in September (just after the clients busy season) but you haven't looked at seasonality for the niche overall, you may end up looking at negative traffic numbers and scratching your head.

And worse, if you haven't educated your client on the traffic patterns, they'll be just as surprised as you are. After all, they brought you on to get them more targeted traffic – why should their traffic drop?

You don't have to predict the future with precision, but there's little excuse for failing to anticipate seasonal trends.

Selling SEO services is as much about demonstrating results and informing clients as it is about doing the actual work. You may have done a great job and improved the long-term picture, but if your client sees traffic dropping and has gotten no heads-up from you as to why expect to be questioned. And if your explanation is a reaction to your client's questioning, expect it to look like backpedaling.

Mike Tekula

Mike Tekula helps small businesses market themselves better online and builds WordPress websites. He also makes a mean peanut butter and jelly sandwich.

unstuckdigital.com

You May Also Like

2 Responses to “Forecasting Seasonality for SEO Campaigns”

  1. [...] Forecasting Seasonality for SEO Campaigns Get a solid list of keywords together for the niche I like to use somewhere in the range of 100 keywords – both the short head and long tail terms. [...]

  2. [...] Forecasting Seasonality for SEO Campaigns – this one I also found interesting as I’ve been research universal search, which is somewhere we see a LOT of movement from seasonal flux. A worthy addition to this week’s editon. [...]