SEO Plateau review request

[Jdamise]

Hi Guys,

I started a website 8 or so months ago now. Things started pretty quickly, the site hit huge speed issues in month 6 which messed a few keywords up and declined the site traffic by around 30% or so. Since, it has rebounded and despite the impressions being roughly the same, the CTR is a lot higher. So we have moved to 2000+ monthly uniques consistently now, but the impressions seem to have stayed the same for months, despite new content.

The site was agreed to sell a month or so ago, which actually fell through on the buyers end. I am in two minds now whether to try and resell it or continue the journey with the SEO. I am a little disheartened as the impressions remain the same but we do well under two specific pages of the site that drive around 60% of the traffic. I have manually written around 65 articles, all the copy, and have since added 2 Ai pages to watch the performance.

Does anyone have any ideas of how I may move to the next level? For me, it is a little bit of a crossroads and I am undecided.

Thanks guys

Jamal

https://africabettips.com

[chaumi]

If you want to pursue it and see if you can grow:

The first thing I'd do is look at rankings for existing pages (I can see the two that seem to be current drivers, others are way too far down in SERPs - though not hopelessly down - to get anything more than the odd visitor)...

And work on improving/enhancing those pages with an eye on the terms they're ranking/attracting traffic for. In order. ie highest ranked first, and make a list of the others in descending order.

Make all (or most) of them better. As you do it, think about the overall site structure, whether any of them don't fit, and maybe would be better off binning/renaming/repurposing.

The good news...you are getting some traffic and some highish (but not high enough) rankings. So, there is a little juice in the site, and that, at least, is mildly encouraging.

****************

If you work on the current best-ranked pages first, they'll then have a bit of time to see if they can go higher. While waiting for that, and instead of working on all the others, you could take the opportunity to fully review the site structure. That way, you're probably hitting it the best way overall.

And picking up a couple of good value links may help.


**********

PS Use your GSC to validate figures!

[PaulEchere]

Whatever Chaumi said about existing ranking pages will definitely keep you busy for some time. Still, your way forward would largely depend on what your aim was so far.

As you said, you have created over 65 new pages / articles, but before writing the content have you done the necessary research?
I am referring to the KWs you would potentially want to target and what were the impression numbers that you would have expected based on that?

If you were expecting a certain number of impressions and you aren't getting close to those numbers - you might want to start with reviewing your site's structure (as already recommended above). Linking between internal pages is what I mean, potentially creating new supplimental pages and interlinking your to form a structure that makes sense and is relevant.

And then of course usual external link building with relevant KW anchors will not hurt.

[Jdamise]

As ever, you offer sound advice Chaumi. Thank you!

I think you are right. When I first started the site, as you know, I was very much a newbie in undertaking my own project. The content side of things have gone ok, but the IA of the site has been somewhat unstructured. I have not focused on keyword dense anchor links in any of the core pages. I think I knew what I was going to hear when I posted this thread!

Thank you for the advice, it is confirmed that if this site is going to move forward it needs some clean up work - ill check in with you again no doubt at this stage!

Jam

[Jdamise]

Thanks for the feedback Paul.

I have used Ahrefs to research each keyword and article placed on the site. The pages are also optimised, though perhaps not optimally. The sites structure is probably the most important thing at this stage for sure.

The site has taken over 238k impressions with a 3.6% CTR, but I wish I had not gone down the predictions route on this project. Still, it has been valuable learns and is my first website. I am interested to hear more about "suplimental' pages - could you elaborate on this please?

Again, thanks for taking the time to feedback on the thread.

Jam

[PaulEchere]

Have you tested your Ahrefs KW research against Google's KW planner? Asking this because I can see you are targeting Kenya (at least partially) and, for African countries (apart from ZA), Ahrefs data is quite often not really accurate, so might recommend things that in reality have close to 0 search volume.

Regrading supplimental pages, I would recommend getting on a VPN connection and searching the KWs that your pages are targeting and going over to the "Related searches" section. Those are pretty much the KWs for your supplimental pages.

This is a bit nuanced though, as in some cases you might want to simply take those KWs and add a section covering that topic to your existing pages. In other cases you will create a separate page and link to it from the "main" one.

Getting this bit wrong may potentially backfire, so before you do one or the other - I'd recommend looking at your close (successful) competitors to see how they have it.

Thanks for the feedback Paul.

I have used Ahrefs to research each keyword and article placed on the site. The pages are also optimised, though perhaps not optimally. The sites structure is probably the most important thing at this stage for sure.

The site has taken over 238k impressions with a 3.6% CTR, but I wish I had not gone down the predictions route on this project. Still, it has been valuable learns and is my first website. I am interested to hear more about "suplimental' pages - could you elaborate on this please?

Again, thanks for taking the time to feedback on the thread.

Jam

[Former Member 14]

2000+ monthly uniques consistently now

Figures mean very little, when most, if not all sites, are being bombarded by bots and other nefarious crap, on a daily basis. I've seen sites boasting 5x that figure BUT when one crunches the numbers, the actual real-human-traffic can be as low as a few 100 per month.

Just putting it out there

[Jdamise]

Figures mean very little, when most, if not all sites, are being bombarded by bots and other nefarious crap, on a daily basis. I've seen sites boasting 5x that figure BUT when one crunches the numbers, the actual real-human-traffic can be as low as a few 100 per month.

Just putting it out there

How would you crunch the numbers and discern the difference?

[Former Member 14]

How would you crunch the numbers and discern the difference?

Raw log files are a good start...

But surely you'd have that data on hand, given, as you stated, the site's sale fell through. Any serious buyer should/would be asking for your stats data.

[chaumi]

I'll make a wild guess you don't know how/where to get raw log files, Jam (I assume these come from hosting)...and even if you had them, how to interpret them effectively (never seen any myself, but I'll make an equal guess I wouldn't know how to get them/what to do with them either).

But there are ways around it (although I'd be prepared to make a bet that what AD suggested is the 'easiest/best' way).

So AD is almost certainly right, the data you have is almost certainly flawed. Depends to an extent on where you're collecting it from.

But how much flawed?

If you have anything that tracks traffic in real time and shows device/IP/location/how many pages navigated etc etc, then that will be giving you some clues about how much garbage traffic you're getting. You can normally see by the general profile(s) whether it's 'real' and how much is probably (or definitely) not real.

When you combine that with tools like GSC, Bing webmaster, ahrefs etc and what they are telling you, you can build a reasonable picture.

Examples

.....if you can see Google, Bing, Yandex etc visits (and they look to be from search), then check your rankings for that search (or what you rank for relevant to that page) . If it's one or two visits, you might be getting seen for a long tail. If dozens/hundreds, then most likely you have to be ranking for a term that gives that volume. If you can't find yourself (at least on page 1) then it's a fair bet it's false traffic (though remember you have to think about originating countries, and search for yourself appropriately. And remember that some SERPs results can vary from search to search (different datacentres, Google testing, etc etc).

...if you have bursts of what looks to be direct traffic, then almost certainly this is garbage. Unless you think someone is telling all their mates ' go look at this site, it's the dogs b****cks (an English term for 'very good' for anyone that doesn't know it), don't search, go straight to it' (and that ain't gonna be happening, for ABT or for the sites/pages of 99.99% of the rest of us).

None of this is precise (which is where the raw server logs I guess may have the upper hand), but it can give a picture (probably clearer over time and wrt consistency) that helps you understand what's happening.

[Jdamise]

I'll make a wild guess you don't know how/where to get raw log files, Jam (I assume these come from hosting)...and even if you had them, how to interpret them effectively (never seen any myself, but I'll make an equal guess I wouldn't know how to get them/what to do with them either).

But there are ways around it (although I'd be prepared to make a bet that what AD suggested is the 'easiest/best' way).

So AD is almost certainly right, the data you have is almost certainly flawed. Depends to an extent on where you're collecting it from.

But how much flawed?

If you have anything that tracks traffic in real time and shows device/IP/location/how many pages navigated etc etc, then that will be giving you some clues about how much garbage traffic you're getting. You can normally see by the general profile(s) whether it's 'real' and how much is probably (or definitely) not real.

When you combine that with tools like GSC, Bing webmaster, ahrefs etc and what they are telling you, you can build a reasonable picture.

Examples

.....if you can see Google, Bing, Yandex etc visits (and they look to be from search), then check your rankings for that search (or what you rank for relevant to that page) . If it's one or two visits, you might be getting seen for a long tail. If dozens/hundreds, then most likely you have to be ranking for a term that gives that volume. If you can't find yourself (at least on page 1) then it's a fair bet it's false traffic (though remember you have to think about originating countries, and search for yourself appropriately. And remember that some SERPs results can vary from search to search (different datacentres, Google testing, etc etc).

...if you have bursts of what looks to be direct traffic, then almost certainly this is garbage. Unless you think someone is telling all their mates ' go look at this site, it's the dogs b****cks (an English term for 'very good' for anyone that doesn't know it), don't search, go straight to it' (and that ain't gonna be happening, for ABT or for the sites/pages of 99.99% of the rest of us).

None of this is precise (which is where the raw server logs I guess may have the upper hand), but it can give a picture (probably clearer over time and wrt consistency) that helps you understand what's happening.

You guys have both made some real valuable contributions thank you. You are completely right though - Raw Log and all these words - I need to go back to my SEO podcasts and search for these terms haha! You know me too well!!!

To be honest. I can see in GSC the fake traffic to an extent. We rank highly, top 3 positions in multiple African countries for around 20 search terms that generate a good amount of traffic. The other bits, a few pages on page 2 and 3 which I can see from impressions to clicks are doing "ok" - and then, theres the rest. But I am not going to make a killing on these two pages. Therein lies the stalemate for ABT

[universal4]

There are a ton of log parsers that can do the trick, both commercial and free.

Awstats is one that is one of the most common, but you will need to run something like that on the server itself. Many hosts supply this and if not it is opensource and is maintained pretty well.

IMO one of the best commercial programs is smarterstats if you want to run it locally on a windows machine where you ftp the logs for the reports. Smarterstats is not a cheap solution, especially for a lot of sites.

Rick
Universal4

[statsdrone]

What about country pages for all African countries?