15 Comments
User's avatar
Philipp Götza's avatar

Finally got to re-reading :D

Is the priorization framework mostly done manual or did you find ways to automate parts of it? I mean obviously MSV can automatically be weighted according to a scale, but what about other factors?

I like the framework of a galaxy for this.

Expand full comment
Kevin Indig's avatar

i do the weighting of the signals manually, but I could see that a clever machine learning algorithm with enough data could make smart suggestions. Just needs enough data, which is often the bottleneck.

Expand full comment
Nilesh Parashar's avatar

1 / For a long time, I operated with the filter that managing a lot of people was the optimal way to respect and money. Turns out, it’s really not. - Can you please explain more?

2 / Do you categorize keywords based on funnel (TOFU, MOFU and BOFU) or the 5 step customer stages you mentioned - Problem Recognition, Research, Comparison, Decision and Post-Sale

3 / What are your thoughts on when people say that top of funnel content is not relevant now because it's not helping in conversion, so they start content creation from conversion keywords. I get it where they are coming from because the stakeholders want some sort of business value from SEO & content. But my point is that why does every content has to convert? Why can't we create content for brand / category awareness? Even if we use another channel, say social media to target those keywords people will still search on Google because they have a question and most of the time social is not the right place to get your answers. I am telling this because a well-known indian beauty brand creates a lot of reels on instagram for top of funnel searches, but when I checked their console data i understood than people still Google the questions. They are not satisfied with what they see on social. Am i wrong here? or am i missing a point which people of your experience and expertise see?

Expand full comment
Kevin Indig's avatar

1/ I make a lot more money now than during my highest inhouse positions. From the outside, high titles at popular companies get a lot of respect, but in the industry, everyone knows that it's a chore.

2/ it depends on the philosophy of the customer. I think either can work.

3/ I think SEO for brand awareness and conversion can both work and be fine. If you have to prioritize or are just starting out, creating content for high-intent keywords is probably smarter because you get ROI quicker. TOFU content takes much longer to convert, which sometimes makes it hard or impossible to measure. and, depending on high up the "funnel" you go, sometimes TOFU content doesn't have any impact. but it depends on B2C vs. B2B and how complex the product is.

Expand full comment
seoaishow's avatar

Dont' mean to dumb this down too much but isn't leveraging Google keyword planner data (specifically the CPC, SV trend, Competitive (low, medium, high) enough to tell you what keywords actually convert? Otherwise why would advertisers spend money on them? That with the search volume YOY change trending up, you can narrow down exactly the high earners. In fact we do this (probably the only ones actually) over at SEO Copilot. PS I think you'll love our KD metric, far more advanced than looking at old outdated DA and backlinks. Let me know what you think.

Expand full comment
Kevin Indig's avatar

CPC is directionally okay but since Google Ads is a bidding system, you never know exactly how well those keywords convert.

How do you compose your KD metric?

Expand full comment
seoaishow's avatar

Our algorithm calculates an average keyword difficulty score (SERP Top 10) by analyzing factors like keyword presence, page authority, content relevance, backlink profiles, and traffic levels across the top search results. This score is then compared to other keywords to determine a relative difficulty rating such as "Very Easy," "Easier," "Average Difficulty," "Tough Niche," "Very Tough Niche," or even "Not Recommended" for extremely competitive keywords. This rating aims to gauge how challenging it would be to rank well for that particular keyword - Furthermore we developed a metric called Keyword Revenue Ballpark which ties in with what I was talking about above, aims to give the user a quick reference to how Lucrative a keyword is.

What other data points would you suggest we factor in?

Expand full comment
Kevin Indig's avatar

yeah, that's cool!

I think if you could let users define their average Customer Lifetime Value to refine the revenue ballpark calculation, that'd be nice. In terms of keyword difficulty, I wonder if a definition of direct competitors could also help refine the relevance of the metric.

Expand full comment
seoaishow's avatar

LTV is a good idea, we thought about that in terms of avg conversion value for example, but we're using some data a Google economic published so it's pretty reliable. That said, we don't allow users to input direct competitors but rather use the SERP to define the competitors, but why do you think adding a competitor that's not ranking in the top 10 would help build a more accurate KD score?

Expand full comment
Kevin Indig's avatar

You’d obviously filter them out if they’re not ranking in the top 10

Expand full comment
Keep The Light On's avatar

This might be a dumb question, but I'm curious to why you didn't use Ahrefs' KD metrics. Is it because it's not orthogonal to your sites, or because it's not accurate? Thanks in advance!

Expand full comment
Kevin Indig's avatar

No particular reason. Just had it handy :).

Expand full comment
seoaishow's avatar

It's an outdated, not useful metric anymore e.g. DA and backlinks - it is not how google ranks pages in 2024 and thus not a real analysis of how hard or easy a keyword is to rank

Expand full comment
Kevin Indig's avatar

I generally agree. But what would you use instead?

Expand full comment
seoaishow's avatar

An advanced AI that researches the top 10 search for you for and measures actual ranking factors of your competition, it's called Keyword Discovery by SEO Copilot.

Expand full comment