Fat Stacks quote from Breaking Bad by Jesse Pinkman

How to investigate traffic drops and lower ad RPMs (for the layperson)?

Man falling off building

These days cars come with all kinds of fancy warnings that seem to come on out of the blue.  You get the engine warning, tire inflation warning and other warnings.

When I see them, I’m clueless.  I have no idea what to look for.

The only thing I can do is whip out my phone and schedule a service.

Fortunately, this is not the case with my websites.

While I don’t have sufficient SEO skills to operate an agency, I do have a few tricks up my sleeve to investigate problems.

In the website publishing world, the engine warning signal is a drop in traffic and/or ad RPM.  When I see either of those to a significant degree, I slap on my layperson website mechanics hat and start investigating.

Sometimes I figure out the problem and fix it.  Sometimes I don’t.

Like the layperson mechanic who can fix minor car problems, my ability to do some preliminary investigations has helped me on more than one occasion.

This article explores investigating traffic and ad RPM drops.

High traffic vs. high RPMs?

The website business is in some ways different than other businesses.

The way I see it, website traffic to ad-supported sites are our customers.  RPM (revenue per 1,000 page views) is our profit per customer (less expenses).  So far, so good.

However, consider the traffic and RPM conundrum.  More traffic is more customers (usually a good thing for all businesses).  Higher RPM is a higher profit per customer (also usually a good thing).

Many businesses would prefer higher profit margins over more customers.

In fact, many businesses choose business strategies that reduce the number of customers but increase profit margins because in some cases this results in more overall profit and a leaner business.

However, when it comes to my sites, if forced to choose, I prefer growing traffic and lower RPMs.  In fact, I’d opt for more traffic and lower RPMs even if it means a lower overall profit.

Why is traffic more important than RPMs?

Because traffic is an indicator of site health in search.  Good site health means there’s growth potential.

Since most of my traffic is from Google search, I want more than anything for my sites to be considered high quality in the Google algo.

Put another way, if traffic keeps dropping, business suffers.  On the flip side, as long as traffic is growing or maintaining, business is decent.  Without traffic there’s no business.

What’s the point of this?

The point is when I check revenue in my various ad accounts and I notice that it’s down, my first investigation is to determine if it’s less traffic or lower RPMs.  If it’s lower RPMs, I’m not concerned.  If traffic drops, that’s a concern.

How does this apply to you?

It applies the same way it applies to me.

I get a lot of emails asking about ad revenue fluctuations.  My first question is “did your traffic drop or your RPM drop?”  If it’s an RPM drop, I explain that’s normal throughout the year.  RPMs fluctuate considerably.

That said, if it’s a huge RPM drop, you do need to investigate.  I set out some ways to do this below.

If it’s a traffic drop, that COULD be something far more serious requiring further investigation.  However, I also suggest to look at traffic patterns during the previous year.  If during the previous years the drops coincided with a recent drop, that’s nothing to worry about.

For example, many sites experience a traffic downturn during the Summer.  At any given time, a large percentage of the population is on vacation during which internet use is lower.  Of course, summer niches are the exception – but then they suffer downturns during other times of the year.

Over the years I’ve noticed traffic patterns throughout the year.  I enjoy high traffic months and lower traffic months.  It’s normal.  I’ll set it out below.  Hence if my traffic is down but it coincides with previous years, that doesn’t worry me.

It’s a concern when it’s a substantial drop that doesn’t coincide with previous years.  That is an issue that requires further investigation.

What’s my typical RPM cycle throughout the year?

Please note that my RPM volatility may not be the same as yours.  Each niche and site is different.  I set this out for illustrative purposes only – to show you that volatility is normal.

Quarterly RPM fluctuations

  • Beginning weeks of any new Quarter = lower RPM
  • Last 2 to 3 weeks of any Quarter = higher RPM

Annual RPM fluctuations

  • Q1 and Q2: Very good RPMs.
  • Q3: RPM drops.
  • Q4: RPM goes up.  Highest annual RPMs around Black Friday and December.

What’s my traffic cycle throughout the year?

Again, the patterns below pertain to my sites.  Your sites may behave differently.

  • Q1: Usually a very nice traffic boost through the Spring.  I have no idea why this is the case, but it is.  It’s not a boost from December’s low traffic, but increases well above previous year’s high.
  • Q2: Traffic continues to grow but levels off toward end of Q2.
  • Q3: plateaus and/or drops a bit.
  • Q4: plateaus and then big drop in December.

How to investigate significant RPM drops?

Small RPM fluctuations are not a big deal. They’re normal. However, large RPM drops require investigation.  Here’s what I do.

Step 1: Ensure all ads are displaying properly on my site

Sometimes, for whatever reason, an ad stops displaying.  This is an easy fix.  Usually, it’s because I messed something up while making changes.

Step 2: Ensure all reporting is up-to-date with your ad networks

Some ad networks have a reporting delay.  Sometimes reporting has a glitch and therefore is not as timely as it normally is.  If this is the case, then it’s just a wait and see (and hope).

Step 3:  Check out ad impression volume

Sometimes demand in ad networks drops so you don’t get 100% fill.  This isn’t a huge problem and there’s not much you can do.  If the problem persists, you might want to switch ad networks.

Step 4: Check key website metrics like time on site and page views per visitor

Maybe one or both metrics decreased significantly which can cause an RPM drop.  If this is the case, you need to investigate what’s going on.  Check the metrics for both desktop and mobile individually.  Maybe there’s something wrong with your site on mobile. If so, get it fixed.  I know I don’t check my sites on mobile as often as I should because I work on a desktop computer.

Step 5:  Ask around

When I see dramatic changes, I ask colleagues if they’re experiencing the same.

Often I do this first because if they suffered a dramatic RPM drop, it’s not just me and so there’s likely nothing I can do.

Step 6:  Check for traffic source changes

Are you getting more traffic from lower RPM countries?  This is a simple explanation.  However, if your US traffic drops significantly, that’s a concern.

If you do all of the above and still can’t figure it out, send in support requests to your ad networks.  They may be able to investigate on their end.

How to investigate significant traffic drops?

I’m not a super skilled technical SEO.  However, there are some things I do and have helped colleagues with when they suffered big traffic drops.  Here are a few things to check.

1. Check Search Console for any Messages

Search Console has helped me twice with website problems.

a. My site was hacked and deindexed

If Google penalizes or takes a manual action against your site, they usually notify you in Search Console.  A couple years back one of my sites was hacked.  It was so bad Google deindexed it. They notified me.  I hired a security team to fix it then submitted it for review.  The site was reindexed and that was that. Amazingly, rankings returned very quickly.  The entire thing took about 1 month (horrible month though).

The point is Google let me know what happened in Search Console, so it’s a good first place to check.

b. URL indexation issue

You can check for other errors and anomalies in there.

I also once had it where the number of URLs indexed for my site went from 3,500 URLs to 120,000 in a matter of weeks.  It was nuts.  I knew that was an issue.  3,500 was the right number.  120,000 was nuts.  It turned out image URLs were being indexed.  It was a big problem.  I had to hire an SEO person to fix it.

Again, turning to Search Console helped me figure out the problem.

2. Were there any recent Google updates?

If you don’t pay attention to SEO news, you may not know about any big recent Google algo updates.  Type “recent Google algo updates” in Google and see what comes up.  Searchenginejournal.com always reports on the big updates as well.

If there was a recent update, then you need to investigate as to what the update targeted and how you can fix it.  Usually you have to wait for this info and hope for the best.  I’ve been on the nasty receiving end of such updates so I know what it’s like.

3. Did I lose rankings for any high traffic keywords?

I love ranking #1 for high search keywords (who doesn’t) but the downside is you can get dethroned which can seriously put a dent in your traffic.  If your site gets 10% of traffic from one keyword, if you get dethroned, you’ll notice a big drop in traffic.  This is one reason why I don’t intentionally go for big keywords with link building promotions – I prefer having hundreds or thousands of articles all pulling in drips and drabs of traffic targeting lower volume keywords that aren’t in the sights of bigger websites.

How do you check for lost rankings?

It’s easy to do if you have a small site, but what if you have a huge site with thousands of articles?

a. Check Google Analytics

The first step is to see if it’s actually Google traffic that dropped.  It could well be a social media issue or perhaps another referral source dried up.  You don’t want to necessarily assume it’s a search issue.  Here’s how I check for Google search traffic in Google Analytics:

1.Analytics google search traffic tutorial jun3-min

The second check in GA is to assess my top landing pages in Google Analytics to see if higher search traffic landing pages were dethroned.  Here’s what I look for.

Analytics traffic drop analysis jun3-min

I then compare traffic levels for top 50 articles or so compared to pre-traffic drop.

What am I looking for?

I’m looking for patterns and/or anything systemic.  If only a few articles dropped but the drop resulted in a huge loss of traffic, while that sucks, at least I know what happened.

In this case, I can investigate what happened with each article and decide on a course of action.  I may improve the article and update it. Some people would build links but I don’t.  If some A-list sites knocked me off I usually leave it at that unless there’s something on-site I can do.

If all or most of the top 50 articles dropped significantly, then there’s likely a sitewide problem, which is a bigger concern.  If this is the case, I have no choice but to hire a technical SEO because I don’t want to waste time guessing.  I want to get to the bottom of it.  This is assuming there wasn’t a recent Google search algo update.  If there was a recent update, I can chalk it up to that and then I must wait to figure out my options.

b. Ahrefs Analysis for ranking changes

Fortunately, this isn’t hard if you subscribe to Ahrefs.

The key is to notice a pattern.  Maybe it’s a type of article or a topic you’ve covered extensively.

For example, if you published 100’s of “Pros and Cons” articles, did they all suffer?  Is there a pattern?

When you can spot a pattern, that’s a good start because you can start formulating a plan to fix it.  Or you decide it’s not worth it and move on.

Here are some various Ahrefs investigations you can do

Ahrefs Movements Tool

1.Ahrefs traffic dorp investigation-min

Ahrefs Organic Keywords Rankings

2.Ahrefs traffic drop investigation jun4-min

If you use Ahrefs and don’t target too many keywords, you can track all your keywords with Ahrefs keyword tracking tool.  I don’t use this because I target way too many keywords so it’s not anywhere near sufficient for me.  However, if I only went after a few hundred keywords, I would definitely load them all in Ahrefs keyword tracking tool.

4. Advanced investigation (for the layperson)

Perrin of Ranq.io SEO agency also recommends the following two steps which can be done by the layperson for some basic investigation.

Step 1: Site audit/crawl

I don’t do this. I would hire an SEO for this because while I could run the software, I’m not sure what to look for.  Perrin recommends the following 2 pieces of software:

  • Screaming Frog for the layperson
  • Deep Crawl for huge sites (over 100K URLs).

Step 2: Historic traffic analysis

Compare current bad traffic (in GA) with the last time traffic was good. Export everything and group pages by % drop. Look at all those manually to try to find patterns

and cross-reference both 1 & 2 with the last 4-5 core algorithm updates or any other major events (like redesigns, etc.).

5. Too many ads?

Perrin also told me:

One last thing that might be worth mentioning if the discussion is also in the context of ad RPM — in the last couple of updates, one of the trends has been sites losing visibility (i.e. rankings) if they’re over monetized with ads or if the ads are deceptive. Not all sites, but certainly a trend.

For those of us that monetize with ads, this final point is worth investigating.  I know I can get carried away with ads in my continuous quest for higher RPMs.  Which brings me back to the traffic vs. RPM dilemma.  I’d immediately choose to remove/reduce ads if having too many is hurting search rankings.

What if I still can’t figure the reason(s) for a traffic drop?

I have no choice but to hire a technical SEO to dig further.  The last time I needed to hire a technical SEO, I hired Ranq.io.  They did some very cool investigations and solved the problem for one of my sites.

 

Reply