Enter a search term such as “mobile analytics” or browse our content using the filters above.
Check your spelling or try broadening your search.
Sorry about this, there is a problem with our search at the moment.
Please try again later.
Last week one of our enterprise subscribers asked about how best to estimate future traffic levels, and I thought I’d answer the question by way of a blog post.
There are eight areas that I think you can focus on, to try to figure out whether traffic levels are likely to rise, fall, or stay the same.
I am thinking out loud here, so if you have any better methodologies then please leave a comment below!
1. Benchmark your traffic over the past three years
Have a look at your web analytics data. Here is what our traffic growth looks like…
It’s easy enough to extend the line on that chart to project forwards and figure out a likely end point in 12 months, but I’m not sure how accurate that will be. Can we do better than that?
I am reminded of my old economics teacher’s use of the Latin phrase ‘ceteris parabus’, which roughly translates as ‘all things being equal’. In digital, everything is subject to change, but some things can be controlled better than others.
2. Understand where your traffic comes from
The next step is to see how people visit your website. Our top 10 referral sources account for about 75% of our total visits. Let’s take a look at them…
You can see that search remains the daddy (and by search, I mean Google – Bing is in 13th place, and Yahoo in 19th). In second place there is the mysterious ‘direct’ channel, which represents all kinds of different sources. In third place is email, which is a consistent performer for us. And so on.
As our own Matt Owen explains, Google Analytics doesn't always measure social traffic accurately, and some of the 'direct' and 't.co' traffic will come from social sites.
Are any of these channels likely to change, for good or ill?
3. Evaluate your top referral channels
For us, and very possibly for you, Google is the big one. A force majeure following a Google update could be catastrophic for our predictions, but I am an SEO optimist and think we can do so much better.
Ask yourself some questions:
- Are you number one in Google for all of the top keyword queries in your sector? (congratulations if you are, though beware dragons!)
- Are you likely to claim a bunch more keywords to extend your share of search? (have you undertaken any gap analysis?)
- Are there any threats to your Google rankings? (dodgy link profiles, poor technical SEO, a lack of quality content, etc)
- How has Google performed YoY? (are visits up, down, or the same?)
Let’s look at our stats to answer the last question. I think it’s a bit of a mixed bag…
You can see that we’re attracting more than 40% more visitors from Google compared to the previous year, and that in the past few months we have accelerated somewhat. All of that is great, but this growth is lower than we experienced in the previous year…
There are all manner of reasons why this might be the case, but right now we’re only really concerned with the future.
We are making some technical SEO changes that should improve our rankings, and I’m being more strategic with regards to our content and keyword planning. Plus, we’re going to produce a lot more content, and as such I expect Google referrals to increase beyond the 40% seen in the past year.
Another big referral channel for us is email, and this is an area where we have a lot more control. Overall the picture is good, with email referrals up more than 50% compared with the previous year.
However growth seems to be slowing down for our email channel, and there are a few reasons for that. Partly it relates to how we attract new members, and partly it is about the growth in mobile usage... two areas where we should see improvements in the months ahead.
Again, there are a few questions to ask:
- How much has the email database grown in the past year?
- Is that growth likely to be consistent?
- Have you fully optimised your email strategy? (sending times, mobile / responsive, etc)
- Did you launch new email products in the last year?
- Are you going to launch any new email products next year?
We should analyse and ask similar questions of the other primary referral sources, not least social, which is bigger than email when all of the platforms are combined, and which delivers a good proportion of new visitors.
4. Think about resourcing
This really is important, with regards to next year’s forecasts and targets. If you are likely to reduce headcount in certain areas then it stands to reason that you may experience a decline in traffic. The opposite is also true!
- Are you planning on ramping up headcount in key traffic driving teams? (e.g. content / marketing teams)
- Are you likely to invest considerably more budget in advertising, or in other areas that will increase traffic?
- Are you likely to reduce headcount, or budgets?
- Are there areas where you can get more bang for your buck? (e.g. hiring permanent in-house staff rather than paying expensive day rates)
As I mentioned, a lot can change in digital, and not always for the better.
5. Beware outliers
See those traffic spikes? What are they all about, and can they be replicated? Most of our spikes come from the content we have created, but some spikes are better than others.
It’s great when a blog post goes viral within our target audience, but we once wrote a Dancing On Ice post that pulled in 100k+ visitors and had a 99% bounce rate. That was a definite outlier!
6. Consider macro events
A couple of years ago we watched in horror as the EU pushed through a law that required digital folk to slap awful messages about cookies on their websites. We saw this coming and created a lot of content about it.
There was a surge in interest, and a good chunk of Econsultancy visitors were looking for information about this nonsensical ‘cookie law’. And then, after the inevitable storm, there was silence.
You too should polish up your crystal ball…
- Are there major events occurring in your sector next year? (The Olympics, The World Cup, etc)
- Are any laws likely to be passed that will affect your business?
- What other newsworthy events can you see on the horizon?
All of these things can play a part in driving traffic. If they are in the future then you can plan for them. If they are in the past, and are not seasonal, then you may see a relative decline in traffic.
7. Take note of macro trends
The above macro factors are external, and event-related, in one way or another. But there are wider trends to think about too. For example: the increasing numbers of ‘silver’ surfers, or improving broadband speeds. These things can play a part in driving up your traffic, depending on your sector.
One big trend is the growth in mobile internet usage, which is happening across the board. Take a look at your own mobile traffic, and see how it has grown. Is your website mobile-optimised?
8. UX factors
Finally, you should have a critical look at your website, and see where the problem areas are. There are all manner of things that we’re going to change, which will greatly improve the user experience.
Things like responsive web design, and page load speeds, and better sharing tools. These things really matter, and I think they’ll make a big difference.
1. You need to be realistic when forecasting. It might be sensible to come up with a range, rather than a single number.
2. Not all of the above eight points are equal. You'll have more control over some than others, and where you have little control there might be a lot of risk.
3. Your business may be very different from ours. For example, if 70% of your traffic is referred by email, and if you can double the size of your email database, then you’re going to seriously move the needle. This is something you may be able to plan for. Second guessing Google’s next move is somewhat more difficult.
4. Traffic isn’t always the most important indicator to focus on. There is good traffic, and bad traffic. Bad traffic is actually a cost to most businesses. Focus on increasing the good traffic.
How do you go about predicting next year’s traffic? Is past performance an indicator of future performance? What are the threats and opportunities that may affect your traffic levels? Please leave a comment below!