Welcome to WebmasterWorld Guest from 54.90.204.233

Forum Moderators: Robert Charlton & goodroi

[How to] Calculate as Accurately Possible Keyword Volume?

     
7:14 am on Aug 2, 2018 (gmt 0)

New User from NL 

joined:June 28, 2018
posts:13
votes: 1


Hi Folkes,

Right now I am trying to calculate as accurately possible keyword volume for specific keywords.

Can I combine and average Keyword Planner data with Adwords data?

Because if I compare 10 single words, I get all different data. See

https://content.screencast.com/users/kasper87/folders/Jing/media/58b3fb64-0c73-437e-8cdd-e46f1a6e3397/2018-08-02_0912.png

Let me know what your approach is to have as accurately data as possible!
1:53 pm on Aug 7, 2018 (gmt 0)

Administrator from US 

WebmasterWorld Administrator goodroi is a WebmasterWorld Top Contributor of All Time 10+ Year Member Top Contributors Of The Month

joined:June 21, 2004
posts:3439
votes: 323


Multiple data sources plus common sense. Each keyword data source is flawed in at least one way. To minimize these flaws I look for consistent high traffic signals across the different data sources. Even if there are consistent signals, you still want to manually review the data.

It isn't really possible to precisely predict future traffic volumes. Traffic volumes are impacted by many factors that just can't be precisely predicted. For example if your product/service is in a news story or a viral social moment, it can spike traffic. If the competition sends out an unexpected big sale, it can drop the traffic volume for your product/service because your customers are buying at your competition. Even the weather can impact traffic volume (good weather=less internet surfing, weather disaster=no internet access).
3:23 pm on Aug 8, 2018 (gmt 0)

Senior Member from US 

WebmasterWorld Senior Member tangor is a WebmasterWorld Top Contributor of All Time 10+ Year Member Top Contributors Of The Month

joined:Nov 29, 2005
posts:9092
votes: 756


Different methods for different folks for different reasons will usually give different results. Unless the underlying data is the same and all you are looking at is different interpretations, stick with the data, do your own work, and use the other stuff to make sure you are not going seriously wrong.

Number crunching is whatever you want it to be. Just bear in mind that "averages" tend to hide spikes and lows which, over time, can indicate trends or areas of action in response to either.
7:58 am on Aug 9, 2018 (gmt 0)

New User from NL 

joined:June 28, 2018
posts:13
votes: 1


Thanks for these helpfull answers!
I would like to hide spikes as much as possible and I know that these have influance. But what I dont understand is the ENORMOUS HUGE HUGE difference in volume/clicks/impressions.

Look at product 1 the lowest is 14.800 and the highest is 733.500 that is just insane. How can I calculate an accurate number? (I know its impressions vs Volume but still)

Product 1: Volume 14.800 [Exact] 29.000
Product 3: Volume 1.300 [Exact] 4.400

Is it smart to just add them up and divide them by 2?
6:42 am on Aug 10, 2018 (gmt 0)

Senior Member from US 

WebmasterWorld Senior Member tangor is a WebmasterWorld Top Contributor of All Time 10+ Year Member Top Contributors Of The Month

joined:Nov 29, 2005
posts:9092
votes: 756


What number are you chasing? Average? Average Median? Prior v Current as an Average/Median?

The data is what it is. Curious on why one might want to suppress spikes/lows ... these I would include in a weighted (over time) analysis.
2:20 pm on Aug 10, 2018 (gmt 0)

Senior Member

WebmasterWorld Senior Member Top Contributors Of The Month

joined:Apr 1, 2016
posts:2366
votes: 629


I would use an index, that is take one product, say product 1 and calculate the value relative to that product's value.

----------------- value ---- calculation --------- index
product 1 == 14800 --- 14800 / 14800 == 1
product 2 == 4400 --- 4400 / 14800 == 0.297
product 3 == 1300 --- 1300 / 14800 == 0.087

Repeat for each category.

This will standardize the values across categories and eliminate the effects of the size of the different categories. Broad impression is 50x of Keyword Planner. So comparing them is difficult. For example product 10 is 50 for kw and 40 bi, pretty close right? Wrong! For kw product 10 is 50/14,800 vs for bi 40/733,583 so there is a big difference and that difference is only really visible when you account for size.

This will not smooth out outliers but it will make value comparable across categories.
7:54 am on Aug 15, 2018 (gmt 0)

New User from NL 

joined:June 28, 2018
posts: 13
votes: 1


@ Tangor:

Well my goal is to give an as accurate as possible estimate about SEO traffic to my website with a certain keyword.

So I have 2 keywords: CHAIR, TABLE, COUCH and want to know (as accurate as possible) how much traffic I will get if I rank #N1 position in Google.

But what data shall I use:
- Keyword planner
- Adwords [exact]
- Adwords Broad
- Adwords "Phrase"
- Impressions Search Console

Do I need to pick 1, or can I combine all these data and use the average number of that?

NickMNS, use an index, that sounds interesting, can you please tell me more about that?
 

Join The Conversation

Moderators and Top Contributors

Hot Threads This Week

Featured Threads

Free SEO Tools

Hire Expert Members