How To Build a Keyword Matrix [and Why You Need One]

by Nick · 56 comments

in SEO

How To Build a Keyword Matrix [and Why You Need One]
  • Buffer
  • Buffer

If built and used properly a keyword matrix can increase your rankings and your keyword spread exponentially.

Keyword matrices have traditionally been used for paid search, but also have far reaching implications for organic search; essentially using other factors of influence to help qualify and quantify your opportunities.

So, What is a Keyword Matrix?

Unlike the virtual world built by the Wachowski brothers, the idea is relatively simple, a matrix is a 2-dimensional table that enables you to cross reference two aspects, and analyze opportunities – with the potential to slice and layer additional information on top for deeper analysis.

When applied to keywords, it gives you 2 axises from which to gauge different qualitative heuristics.

Michael Martinez discusses passive versus active keyword matrices, putting the emphasis on purpose; a passive matrix records keyword data that other websites are using, where as an active matrix is used as a design template for your website and content.

I am going to be taking a different approach to defining and building keyword matrices for organic search.

I want to look at search intent versus search volume and search volume versus SERP competitiveness. These 2 scenarios are very interesting because the first is an inverse relationship, and the latter is linear.

Here’s a look at the relationship between search volume vs. user intent: search volume vs intent

Query complexity tends to also have an inverse relationship with search volume; as complexity decreases (from long-tail to head), search volume tends to increase.

In the graph above you can see how search volume steadily decreases as intent increases (thus the inverse relationship), and again, it is usually the case that the query length (and complexity) increase in relation to intent.

Now look at the relationship between search volume and SERP competitiveness:

Not surprisingly the higher the search volume, the more competitive the SERP. Now let’s overlay intent segments over the search volume versus search intent correlation graph:

Looking at this further – this starts to lay out what your matrix will look like.

With the implications between hummingbird and moving entirely to secure search, a focus on intent at the content level is critical.

We are definitely seeing a move away from the big volumes offered traditionally by “head terms,” into an era where the long-tail is even more important, as Google serves up more personalized results. This means that a simple result for ‘insurance’ will no longer be shown, and instead the results will be more relevant to something such as ‘cheap car insurance quotes for a 33 year old male’.

 

- Simon Penson, Founder of Zazzle Media

Many SEO’s are already gleaning intent from their keyword research:

Keyword research is excellent for understanding user intent, which is one of the main pillars of my content strategies. I focus on crafting content that will meet customer expectations in hopes of creating a better overall experience.

 

- Gisele Navarro, Head of Outreach at NeoMam

One major issue is that getting your head wrapped around intent can be difficult at first

I think initially, understanding keyword intent can be difficult for some people. Once you get past that, I think the challenging part becomes getting the right level of granularity for the topic of a page, and how diverse your set of keywords are to include on a page can be tricky for people.

 

- Geoff Kenyon, SEO Consultant at Distilled

And this is precisely how using keyword matrices can be beneficial; they help categorize your target terms by providing a visual cue to brainstorm and prioritize.

The hardest thing I deal with is how to balance both keyword strategy & UX when you’re trying to cut up your products & categorize them in different ways. i.e. some issues that will come up that are purely judgement calls:

 

  • Trying to label subcategories as highly searched keywords, but those subs aren’t necessarily the best way to cut them up from a user perspective (they wouldn’t mine down to the products they want that way; i.e. too specific or non-inclusive of what they’re looking for)
  • Trying to create a category for Product Characteristic A, and one for Product Characteristic B (because they’re both big keywords), but a product could possibly have both characteristics, which could confuse the user at the point of entry (even if you have some products in both categories, which would be one possible solution)
  • This is more research oriented, but not knowing enough about your product’s terminology that could give insight into what keywords people are searching to find your products, as well as variations off of that (keyword planner doesn’t give the best related keyword suggestions IMO).

So really what keyword strategy (and research) comes down to for ecommerce sites is A) knowing some general SEO rules of thumb (i.e. an optimal URL structure, what to do when you come across situation X, etc.) B) knowing your products inside & out (this is essential and spells doom for anyone who tries to find shortcuts!).

 

- Jon Cooper, Owner of Point Blank SEO

Why You Need To Be Using Keyword Matrices

To Jon’s point above, you need to take into consideration how keywords are going to shape your content, and how your content needs to be formed to support your user’s intent, both in getting searchers to your site and to help them navigate while they’re there.

Think of this scenario:

Your manage an eCommerce website that sells high-end camera’s and camera equipment.

Two visitors come into your site using the search term “SLR camera accessories,” they both searched Google using the same body term with commercial intent.

They are both looking for the exact same product, a Noktor 50mm lens, but here’s where it gets interesting, and where alignment between keywords and your content becomes critical to conversion – one visitors click’s into the category “SLR Accessories,” while the other clicks into “SLR Lens.”

Do they both find the Noktor 50mm lens they’re looking for?

It depends on if you have listed the product in all of the intent-based locations where it’s relevant.

The separation is between product versus function, one user is looking for a product specific attribute, a lens, where the other is looking for a functional attribute, an accessory.

Keyword research is one of the first places to find consumer pain points – by nature, they are looking for answers to their questions. So, it’s a great place to do research. But finding the keyword is not enough – you must also do a competitive analysis to determine whether or not your solution can be the best, or the cost-benefit may not be worth it.

 

For example, we built a marketing checklist and did comparative research: are there any other checklists out there? Do they compare to our proposed solution? What competitive advantages do they have? After determining we could create by far the best solution (not a requirement), we moved forward in building it.

 

- Ross Hudgens, Founder of Siege Media

In the scenario above it’s not as much about the pain points as it is how different brains approach solving the same problem, and how content needs to be structured as a path toward a solution, regardless of which path is taken.

How SEO’s Currently Approach Keywords

The idea of using a keyword matrix is far from new at this point, but it really isn’t talked about much on the interwebs.

I wanted to get a sense of how important keywords were to some SEO’s when it came to producing content – as a litmus test for how much impact a strategic tool like a keyword matrix might have on their efforts.

Ross puts it beautifully:

Simply because a keyword exists and it could send you relevant traffic, does not mean you should build that page. Your competitors may simply be better positioned to rank for those keywords, both today and in the future.   The second part is semantic touch / IA / UX combination.

 

It’s hard to mix the usability / information architecture / SEO needs in a neat and non-obtrusive way, and that’s something that takes a lot of experience to get right / blend delicately.

How important is aligning keyword research with your content production efforts?

Understanding the possible intent of the searcher for each keyword we’re targeting allows us to have a better approach and purpose for the overall content.

 

- Jason Acidre, Owner of Kaiser The Sage

In a recent article by Peep Laja, he said:

SEO is going to get increasingly harder. If you’re still doing SEO by optimizing for keywords, you should know: it’s not 2008 anymore. Things have changed! Google understands context, natural language, intent.

And he is completely right.

Expanding on head terms and focusing on traffic is not how you are going to win the SERP’s as we move into the world of 100% not provided and complete geo-personalized results.

But getting intent right is hard, and making sure you are developing the right content to support the right intent, at the right phase in the conversion funnel is even harder. But using research, analysis, and a dash of creativity gets the ball rolling…

Coming from an SEO background I’ve always felt that keyword research is one of the best sources for content development. Before Google stole our keyword data, leveraging your existing analytics could show popular topics that were performing well, but were under serviced.

 

For example, a marketing company may find articles around “Facebook Advertising” were generating 25% of their leads, but only made up 3% of their content. It’s a good indicator of topics you could possibly flesh out and create more content around.   Of course the other great use of keyword data is using tools like UberSuggest, Google Keyword Planner, Google Trends, Google Suggest, etc to build out a content plan around topics that are relevant to the personas you are targeting.

 

I’ve ran several workshops for teams of editors in big media companies and showing them how simple it is to get content ideas using tools like this really changed the way they approached their editorial meetings.   I’ve always been a big fan of owning a topic. If we are talking about keyword lists, then in my mind that’s a owning something like “best red widgets” as the core keyword, but it will have a bunch of secondary keywords hanging off it “best red widgets 2013″.

 

In these cases you are trying to create a relevant content campaign that has multiple pieces in different formats all focused around that central topic. If you split your campaign like this, interlink those pieces and have a solid plan to distribute links across the different pieces, it gives Google a better reason to rank your content. You can obviously just go for one stand out piece if targeting a big keyword.

 

- Kieran Flanagan, Marketing Director at Hubspot

How to Build a Keyword Matrix

Initial keyword research is the first step toward building an informative keyword matrix.

So fire up keyword snatcher (or whatever keyword mining tool you fancy), drop in your head term (just one), and send it off to the races. The Excel file is designed to handle *any* keyword data, as it uses a natural language macro to look for signaling words (that you can manage in the key) to tag searcher intent.

Export to CSV > Filter out illegal characters > cut into lists of 10,000 keywords > Save CSV.

Upload into Google Keyword Planner > Get Ideas > Export to CSV.

Use keyword combiner to join all of your individual research files.

Open in Excel > Sort descending by search volume > Add a column for intent > Add a column for “Best” > Save.

Here’s where you should be at:

keyword-matrix-setup

Tagging For Intent

Tagging for search intent is critical, this is where effective prioritization starts to really come into focus. Here is how I approach tagging for the 4 top-level buckets of user intent:

  • Informational – a specific question where the results are the information the searcher is looking for, may contain phrases such as:  info, more information, details, features, benefits, etc.
  • Navigational – contains the name of the brand, product, service, or a person at the organization.
  • Commercial Investigation – easiest to identify and least ambiguous, these queries contain specific parameters for researching a purchase, such as sizes, colors, versus, best, price, pricing, etc.
  • Transactional -indicative of shopping behavior further down the conversion funnel, at this point the searcher knows what they want – so look for terms like buy, purchase, sale, coupon, discount, or locations.

To make this easier on you, I’ve built an Excel macro that, with a bit of keyword specific tweaking, automates much of this process for you. You can watch a quick video of it in action here ›

And in case you want to use the macro, here is the VB code you will need:

Don’t Feel Like Learning Visual Basic?

Well, you probably should if you’re doing much enterprise SEO, buuut, I’ll help you out.

If you don’t feel like going through the time and pain of getting that snippet working, instead of wrestling with someone at Fiverr, you can grab a copy of the actual file I used (complete with all my sample data) for just $5.

Grab a Copy of Search Intent Template for $5

Tagging For Best

Here’s the human component that is still needed to really analyze keyword opportunities for maximum impact and potential.

I call it my “best” column, and it is my process for manually reviewing and tagging the keywords that I believe are the priority targets for the campaign – and it is exactly as easy as it sounds.

Once I have gotten my Excel file laid out to my liking, I sort by type of intent and filter for keywords with a minimum monthly search volume of 1,000 or more.

Now based on the specific goals of the campaign, I tag my top priorities – the best keywords. This helps focus on a top tier of say 20 to 50 (at most) keywords.

SEO-Nick_Keyword-Matrix_Best

I then run each keyword through SEMRush to look at the competitive landscape of that SERP.

And here’s what I’m paying attention to:

The Related Keywords Report

SEMrush_related-keywords_slr-camera

This report gives me an immediate sense of the commercial interest and general competitive landscape of this SERP (as well as offers some fantastic contextual suggestions for queries closely related to mine).

Which Websites Are in Top Organic Results

This is a quick and dirty tactic to get a sense of the SEO competitiveness of the term without having to actually inspect the SERP (and also without having to worry about personalization, localization, and all the other signaling algo’s that may affect what you see).

To get this data just use the default ‘Overview’ view on SEMrush, and scroll down:

SERP_SEMrush_slr-cameras

Laying Out Your Matrix

Though I have seen people do this in Excel, I prefer the ability to quickly wipe away, move, change my mind, and finalize my concepts much like a figure artist with a sketchpad. For that reason I recommend a sketchpad or ideally a whiteboard.

I’m going to stick with my earlier example of building out keyword intelligence visuals for SLR  accessories since there are a lot of terms and a lot of potential variations for each stage in the intent funnel.

Once you’ve finished tagging all your keywords for best and intent, there’s one additional caveat that I like to add, occasionally, which I call content type. And this is exactly what it sounds like, it’s a human element that provides some context around what might be the best path for creating content around that keyword – and while there’s a lot of options, I tend to focus on blog posts, aggregate pages, and evergreen pages.

Here’s a quick look at what this might look like (click to enlarge):

SEO Nick Keyword Matrix

In the above view I have sorted the intent column by Average search volume > Search intent for commercial investigation > Best.

Next we move onto the keyword column to separate which keywords should be targets for organic versus PPC, because all keyword opportunities are not created equal. These are based on a few heuristics but really come down to average bid price versus AdWords competition on the PPC side, and search volume versus organic competition on the SEO side.

Here’s a peak at this list tagged for targets:

SEO-Nick_Keyword-Matrix_Target-tagging

Making Cents of The Data

Now you have a enough data to start making informed decisions regarding your keyword targeting and timing. Using a keyword matrix gives you the ability to quickly refer back to a set of data (as long as you keep it up to date!) to drive your editorial calendar, your on-page content, when and which pieces warrant video, and any other opportunities for keyword optimization.

Happy hunting.

About Nick
Nick is the VP of Digital Strategy at W.L. Snook & Associates, Co-Founder of I'm From The Future an ecommerce consultancy, and the author of this SEO Blog. Follow Nick on Google+.

Follow me on Twitter · Visit my website →

{ 46 comments… read them below or add one }

die blitzcraig March 18, 2014 at 2:23 pm

Bro… This is awesome! No other words necessary! Can’t wait for the excel file and any other seonick awesomeness!

Reply

jimmyp March 18, 2014 at 2:35 pm

I’ve been using a similar approach for a while. I like to include adwords cpc data and compare that to my competition metric. It’s interesting to see what people are bidding on that has low traffic, and I would also consider to have low intent versus high traffic keywords with similar intent that have low bid cpc.

I also sort the target keywords by what I call “hotness” which is a ration of traffic versus competition compared to cpc bid price and intent. This is a good way to select all level of keywords, from category to long tail.

Reply

Nick March 18, 2014 at 4:08 pm

Hey Jimmy – Yup, very similar to my approach for paid search.

Hotness is a cool idea, I segment by seasonality but that’s a very interesting consideration to take trend velocity into account; great idea!

Reply

Ryan McLaughlin March 18, 2014 at 2:59 pm

Hey bud – Ever since you tweeted out that Jing showing off your macro for assigning intent, I was hoping you might lay out the VB for us… Thanks so much!

Reply

Nick March 18, 2014 at 4:09 pm

You always could have asked :) But really glad you dig it man. Thanks.

Reply

Joe March 18, 2014 at 3:39 pm

Badass dude! This encompasses so many good keyword research articles encapsulated into one.

I know you’ve touched on keyword and search traffic forecasting before and about using 3 tiers – conservative, best case, and middle. Would you say this is the best source to build traffic and keyword forecasting off of? I know that keyword forecasting is dangerous territory but some clients love the idea of it

Reply

Nick March 18, 2014 at 4:11 pm

Thanks Joe.

I took a stab at creating a model to assist with forecasting called my keyword opportunity model.

Is that something like you were looking for?

Reply

Joe March 18, 2014 at 5:54 pm

That one’s good too – but there was a specific post I had in mind that you wrote where you said what you learned from you Finance past is that you want to offer 3 projections to a client – a conservative one, a middle one, and best case scenario. I’ve been looking through your posts but can’t find it…

Joseph Robison June 27, 2014 at 10:01 pm

So if anyone else is looking, I actually did finally find what I was looking for after all this time. On your Hitreach post – http://www.hitreach.co.uk/blog/nick-eubanks-interview/ – you talk about the Budget, Plan, and Goal models. That’s exactly what I was thinking of.

victorpan March 18, 2014 at 4:14 pm

Here’s a keyword matrix for thought:

Y-> Predicted Keyword Category Growth/Decline
X-> Avg Serp Pos

Segment into 4 quardrants – A horizontal line on 0% growth (growing and shrinking keyword spaces) and a vertical one at 11 (difference between page 1 and page 2 in SERPS). Congrats, you now have a BCG growth matrix, except for SEO.

Now let’s add a layer. We’re talking about matrixes after all. How about keyword volume? Sounds like we should prioritize keywords with high volume, high category growth, but lower SERP rankings on pages 2+.

Another layer. How about keyword competitiveness? Why not work on the lowest hanging fruit with the highest traffic benefit in a growing keyword space?

Reply

Nick March 18, 2014 at 4:16 pm

Love the idea.

For Y are you looking at trends and velocity or using another predictive data tool? (and what is it).

Reply

victorpan March 18, 2014 at 4:37 pm

Y I’m looking at trends. Google trends will give you the data for that ;)

We’re looking long term. Invest in keywords the same way you would invest in stocks. Look for high average long-term returns. This way if your ranks stay the same your traffic still increases. You still get credit for that as well ;)

#BuffetSEOStrategy

Thomas Kane March 19, 2014 at 11:31 am

Jason, this is awesome, as always. Thanks for all your great post I learned a lot from them.

Reply

Nick March 20, 2014 at 9:47 am

Thanks Thomas? I think you were complimenting me.. but my name is Nick :)

Reply

Gael March 19, 2014 at 12:00 pm

I like the idea of automating keyword intent detection. I would totally pay for a SAAS that works the way you built your excel sheet ;). You know what would rock too? A video where you run through the process and explain your thought process.

Keep it up.

Gael

Reply

Nick March 20, 2014 at 9:47 am

Thanks a lot Gael… that’s a *very* good idea..

Reply

Hashim Warren March 19, 2014 at 3:29 pm

Does user intent = keyword complexity in your graph? How do you quantify intent?

Reply

Nick March 19, 2014 at 3:41 pm

It does not. In my opinion high intent does not always imply complexity, for example: 28 inch traffic cone is a *very* commercial term, not that complex (i.e. 4 word query).

I’m not sure you can quantify intent unless you design a scoring system based on natural language..

Reply

Tommy March 24, 2014 at 7:51 am

Nice post Nick!
Love to read something what is more indept then many of the latest posts on other SEO related sites. I will definitely use it.
Next step has to be something about the realism that a keyword can get a top position in the SERP? Maybe a extra import of link popularity data for the competitors?

Reply

Nick March 24, 2014 at 8:46 am

Hey Tommy – Thanks for reading and I’m glad you found it useful.

I completely agree that the “rank potential” of a keyword is an important consideration, and I have actually written 2 posts on this topic:

1) How To Do an SEO Competitive Analysis, and
2) Do You Deserve a First Page Ranking

Please take a look at those when you have time and let mw know if you still have any questions.

Cheers!

Reply

Matt March 25, 2014 at 7:16 pm

Nick,

Great post!

I was trying to get your kw combiner file, but gumroad said the page is a 404….help.

Thanks!
Matt

Reply

Nick March 25, 2014 at 7:24 pm

Hey Matt,

Thank you. Yup, I had changed the URL and forgot to update the post :P Fixed now.

Thanks!

Reply

Matt Tyler March 26, 2014 at 6:31 am

Wow what a great article, this is all new to me so I’m looking for ways to help with my site. il certainly be using this approach. il let you know how i get on.

Reply

Joe March 26, 2014 at 1:19 pm

How come you’re focused one commercial investigation queries and not so much transactional ones? Is it because it’s harder to create content on the transactional queries or because they’re low volume? Do you look at keywords at the keyword level still or look at them in aggregate for a page.

Reply

Nick March 27, 2014 at 9:47 am

Hey Joe – Sorry if the screenshots look misleading, it’s not that I’m entirely focused on the commercial investigation (though IMHO they make better organic results where as transactional usually gets the lion’s share of the AdWords dollars), this is just how I sorted that column for the shots in this post.

I actually use pivot tables to roll-up keywords at the page/content level to build target lists for each URL. Thanks for reading man.

Reply

Joe March 27, 2014 at 12:00 pm

Gotcha – I downloaded the excel and noticed there wasn’t data on the transactional keywords like there was for commercial intent, but I guess it was just an example and not complete. I like your pivot table practice – I’ve done that before and generally think it’s the way to go. Thanks!

Abdulla March 31, 2014 at 11:10 am

Wow what a great article, this is all new to me so I’m looking for ways to help with my site. il certainly be using this approach. il let you know how i get on.

Reply

alan March 31, 2014 at 3:25 pm

Hi Nick, great post and VBA function. Very slick.

Can I ask though about the reasoning for chunking the keywords into groups of 10,000?

The adwords keyword tool reports back on keyword upload lists of 800 maximum I believe or am I missing something?

Reply

Nick March 31, 2014 at 3:27 pm

Thanks Alan – Nope, great question – and actually a good reminder that I need to go update that part. Back in November 2013 when I started writing this post users could still exploit Keyword Planner for up to 10,000 rows of keyword data simply by accepting the bug it generated and downloading the CSV file.. Since then this loophole has been fixed :(

Reply

Dinesh Kumar April 3, 2014 at 2:41 am

hey good one i am new to seo and and your article helps me to improe my skills thanks for your informative article

Reply

Nick April 16, 2014 at 3:14 pm

Thanks Dinesh

Reply

Kaily Daida April 13, 2014 at 11:09 pm

Excellent article. I love that you modified the keyword matrix to analyze long-tail keywords/user intent: it addresses the buzzy trend of “content marketing”.

Reply

Nick April 16, 2014 at 3:17 pm

Thanks so much Kaily.

Glad you picked up on the focus on the matrix, which is to align keywords with the specific types of content that can be nest utilized to rank for them :)

Reply

Phill George April 29, 2014 at 10:40 pm

Hey Nick,

Great spreadsheet by the way.

I plugged in 50k keywords and clicked the Get intent button however the LSV and Bid info didnt update (even though macros was enabled). Is this a known bug or am i doing something wrong ?

Cheers.

Reply

Joe Robison May 2, 2014 at 1:21 pm

Hey Nick I tried to update the Navigational search intent keywords and replace “Brand” with my actual brand name, but I hit get intent after doing so and nothing happened. Is it meant to work that way? Is there something I have to edit in the macro?

Thanks,

Joe

Reply

Nick May 6, 2014 at 4:10 pm

Hey Joe,

Ack! yeah that makes sense, it’s using basic natural-language processing and specifically (in the macro code) looking for the brand tag to complete the intent filtering.

Take a swing at finding and replacing the word ‘brand’ in the macro and please let me know if you’re having trouble and I’ll try to help.

Cheers man!

Reply

Jennifer May 3, 2014 at 9:40 pm

This is awesome, Nick! I can’t wait to try it.

Reply

Patric May 4, 2014 at 8:16 am

Very actionable article. I love this stuff. I’m looking forward to present the results to my boss and getting the credit for it! ;)

Reply

Albert May 6, 2014 at 9:43 am

Oh man, amazing! I have to practice it. Right now.

Reply

webanalysis May 8, 2014 at 2:50 am

This encompasses so many good keyword research articles encapsulated into one.your article helps me to improve my skills thanks for your informative article I need to find my website on google using special keywords.

Reply

webanalysis May 12, 2014 at 6:19 am

Nice post nick Thanks for all your great post I learned a lot from them this is awsome
more ditail click here http://www.webanalysistechnology.com

Reply

Marian May 23, 2014 at 1:08 am

Great post Nick, definitely worth trying and sharing to others. thanks a lot..

Reply

Affordable website nottingham May 28, 2014 at 3:38 am

Hi this is very good and it is nice this is share good information because u know this blog is unique and share keyword matrix.

Reply

Maxx Heth June 3, 2014 at 7:53 pm

Hey Nick, I appreciate your fresh perspective on objective keyword analysis. Definitely beats the “throw-everything-on-the-wall-and-see-what-sticks” mentality, but I’m having a slight issue with your VB code.

I tried copying it and running it through OpenOffice, but I keep getting an error message saying “Basic Syntax Error. Unknown data type Range.” I understand OpenOffice uses its own form of Basic, but do you know (or anyone else here who uses OpenOffice) what data type I would have to change it to to make it work? Thanks!

Reply

Nick June 4, 2014 at 9:56 am

Hey Maxx –

Thanks for reading and actually taking a stab with the macro code (I feel like a lot of people don’t actually try to execute).

I don’t know too much about OpenOffice versus Excel Macro differences, but they look like they are mostly pertaining to the syntax for accessing the data in the Document/Worksheet/Cell.

There are a few significant syntax differences in the selection of ranges of Cells that would make modifying the Macro trickier than just changing one function call.

The documentation for OpenOffice macros was not very clear, but these pages seem like where the pages I found regarding dealing with Cell Ranges in OpenOffice:

Cells and Ranges in OpenOffice

Creating Formulas: Ranger Operator

Reply

Maxx Heth June 4, 2014 at 11:28 am

Hey Nick, thanks for the references! =) If I manage to get it working after that, I’ll send you a copy of the revised code so you can update the article for everyone else who uses OpenOffice.

I’m sure there are a few of us out there who are at least willing to give it a shot, especially given the quality of the information made available following a successful run. =)

Leave a Comment

{ 10 websites link to this post }

Previous post:

Next post:

my script