⌚ Timecodes: 00:00 Introduction 00:26 The evolution of Ahrefs’ homepage 01:44 Our concerns with the old homepage 04:00 Three concepts of a new homepage 06:45 Market perception vs company’s vision 09:06 Copy vs design 11:04 A/B testing 12:23 Was it worth it and should you do the same?
Tip: Start with the timecode 0:00 and put the list of timecodes in the description, so you can leverage RUclips's new feature: twitter.com/RUclips/status/1266051626562990082
I don't think you'd need so much information to get a statistically significant A/B test. I'd measure trial signups, and also measure how many of those trial signups turn into paid customers over time (although there are many more factors in this regarding followup). If I were A/B testing a homepage, I'd be most interested in measuring bounce rate, exit rate, what next pages they go to, and trial signups.
for each metric you want to A/B test you need to come up with a logical hypothesis on what value you aim for and why. High bounce rate is not necessarily bad. If a person realises fast that the product is not for them and leaves swiftly - that is cool. Less strain on support team, less refunds, etc. "What next pages they go to" is also something you're in control of. So before A/B testing this, you need to formulate a hypothethis on which pages we WANT them to go to and why. And we need to formulate which CTR we will consider success and which would be failure. And the more you think about this, the more complex your research will get. So our strategy was to release the page the way we want it and make sure it won't plummet our growth rate. So far I can say it worked :)
PPS: I'm not advising that everyone else should make changes on their website without testing it first. I'm just sharing how we do things at Ahrefs. Some more details will be shared in the very next episode, two weeks from now ;)
@@ahrefspodcast I appreciate the detailed response. It does seem a bit like overthinking though, and if you make the issue so complex, I can understand how it's tempting to just not do it at all. "High bounce rate is not necessarily bad. If a person realises fast that the product is not for them and leaves swiftly - that is cool. Less strain on support team, less refunds, etc." - You could say the same thing in inverse. A high bounce rate is not necessarily good if it's also resulting in missed signups. Someone lands on a page out of interest for a product or service, and it's up to that page to help them make a decision. When I think "What if I converted more traffic?" the first thing to come to mind certainly isn't "It's because they aren't my ideal customer, and they'll want refunds." I'm certainly making some big guesses on how much traffic your homepage gets in regards to A/B testing viability. I figured you'd be able to get 100k+ visitors per test pretty quickly, and that will give you valuable insights. With such drastically different pages, the differences in these metrics are likely to be significant. You'll also get very interesting insights depending on the traffic source for each page variation. If one page has a 25% higher bounce/exit rate than the other, I think it's significant enough to be considered. Then while considering this, I would compare trial signups, and the rate of trial -> paid conversions. If you test, a winner shall be revealed. This is all based on another principle for webpage design that comes from my PPC background (This is what happens when an ad guy gets into SEO) - the page is never totally complete. The internet moves so quickly now, and with the amount of data that we have available, it makes sense to update pages as it makes sense. SEO's typically get this when it comes to updating old content for "freshness" or to adhere to new ranking factors, but the same concept can (and likely should) apply to any other important pages online. After all, what would the impact of a 1% lift in conversion rate (trial signups -> paid) have on ahrefs' profits?
@@MorganCrozier "my PPC background" - paid traffic is usually "cold," so it makes sense to test any variable that will help you convert it. We don't drive any paid traffic to Ahrefs homepage, so people who open this page are (for the most part) well educated about what Ahrefs is (mainly thanks to our blog traffic and youtube views).. So we're not THAT cautious with making changes to our homepage. As I saw someone saying in a thread discussing our new homepage - we could make the whole thing blank and only leave the trial button, and people still would sign up. I think that person had a point :)
Great video! I've come full circle in my career from managing website redesigns from the design, copy, and implementation side - first at agencies, then in-house, and now as a copywriter. Understanding this perspective as a copywriter / consultant is really helpful. I'm always sure to partner with the design team as my very first step in the process. I'm going to use your video to help educate clients! (even if it means they bring in 2 other copywriters/agencies :)
Very refreshing to hear a more value & perception driven reasoning for the changes. I agree that A/B testing is a very valuable tool in every online marketers tool box, but don't make it too easy for yourself by thinking it will fix all your problems. Well done. #TIM30
I was really curious about this. When I saw the change in the homepage, the first thought I had was....I don't like it. But the next thought I had was...They probably have data that proves it works better. Thanks for the behind-the-scenes look at how you made that decision. I too am currently ramping up the redesign, and copy writing for my book marketing software's homepage. #TIM30
In all honesty, when I first saw the new design I didn't like it too :) I thought maybe our designer is trolling me. But then I realised that it's just my cognitive bias. I got used to the old homepage so much, that my brain couldn't appreciate the novelty and it's awesomeness. But now, after some time, I LOVE our new homepage. Remember the launch of iPhone X and how everyone made fun of that stupid "eyebrow?" Well, the sales beat the record and other brands started copying that eyebrow on their phones ¯\_(ツ)_/¯
Great idea to revamp your homepage! I know not everyone will be able to spend that much $$ but awesome to see how much thought goes into it and hopefully will cause businesses to think deeper before changing their homepage. Thank you! #TIM30
I liked how clearly your explained the challenges with A/B testing and factors you would need to consider (font, color, sizing, copy, metrics to measure etc.) This makes it very complicated to A/B test. Also really interesting to see you and the team take the best aspects of every copywriter and do what you thought was best for Ahrefs (not just accepting what was delivered). And finally setting expectations early on in the process definitely makes the whole process go lot smoother. I hope your CEO is happy with you 😀 #TIM30
I should say that you are a brave man, Tim! Launching a new homepage during pandemic was a bold move - seems proper assessment of the results may be a headache (some marketers would sign up if not covid impact on their business), at least we have this problem in our business now. #TIM30
true.. but links is what makes the web the web. It is also at the core of Google's PageRank algorithm, that has changed the web. So not the worst brand name for an SEO tool at the end of the day
#Tim30 It is refreshing to hear there is no "one" way to build a homepage, but that is also quite scary. We did enjoy your suggestion to hire an outside party as a consultant to get a different perspective on the project.
This is super helpful Tim, especially as we are in the process of redefining our homepage ourselves. Love the confidence in backing your collective wisdom and going for it. Your input on where we're at would be awesome #TIM30
Your homepage and that sleek typeface look great, Tim. I can't wait to understand the thinking behind each of the elements on the page. I do have a question though: in most contexts, especially in large companies, there's a certain amount of friction against having one page look a certain way and the rest of the pages look another way. How did you reconcile that, or did you even have that issue to begin with? This leads to a second question: do you think it's substantially more constructive to redesign or realign a website one page at a time, or do the whole thing at once as a "big bang" launch? #TIM30
great question! As you can tell, we released new homepage, while leaving the rest of the pages intact. Yes, this might lead to a slightly confusing experience to some visitors, but we don't see this hurting us in a substantial way. Within about a month we plan to adjust the design of the most popular pages, and in a few more months our entire website would be aligned in terms of design. Bottomline: we decided it is more convenient for us to gradually update our pages, rather than roll out the new design everywhere at once.
@@ahrefspodcast Yeah, that seems like a good strategy, considering there's no real pressure to push the new design super fast across the site, it would probably also allow you to spend more time on details and measurements to fine-tune each page if required, instead of panicking and having to audit and edit dozens of pages at once.
Very interesting approach and a bit bold to do such a drastic change without testing :-) But you bring up an interesting problem for most SaaS type businesses; how to measure conversions... Maybe you could do a video focused on your takes on KPI's and the problems that they all have? #tim30 :-)
our KPI is very simple - it's our monthly growth and our ARR. Every time we start discussing some other "intermediary" metric that we could possibly measure we come up with all sorts of arguments on why that measurement might not be objective and why knowing the exact value of that metric won't really affect our business/product/marketing decisions.
Hi Tim, thanks for these videos. In my opinion, the design and copy looks great but the TYPEFACE, the Typeface could have been different. The blog reading was better with the previous typeface. #TIM30
I'm interested if you can share some details regarding the process of hiring the 3 agencies. Although, it's probably a topic of a whole new episode. I guess you had different sources to collect ideas and start your initial research. Such as recommendations from fellow marketers, hiring websites, forums, WOM etc. As far as I remember, there was also a post in the Ahrefs insider FB group to gather additional ideas. Also, was it tough to pick exactly those 3 agencies? For example, probably there was a moment when you narrowed your list down to 5-6 options. From there, how did you select the chosen 3 agencies to work with? #tim30 PS: I really hate when I talk to people in our industry and when I mention them about how great of a tool Ahrefs is, they say smth like this: - Ahrefs?! Yeah, I know them- they are great at finding links!" and their only association is with links. So sad.. but yeah, hopefully over time more and more people will get familiar with the true power of the all-in-one toolset. I guess we can't blame them all for their ignorance. But growing your blog & youtube audience definitely helps to achieve this mission.
glad to see you here, Jason! :) Yeah, the way I found those three agencies was purely through recommendations of my friends/peers. Luckily I've made quite a few friends in the online marketing / SaaS space, so asking them for their advice on copywriters who they have worked with was all I had to do to come up with an initial list. From there I only did a little bit of research on each agency and went with my gut feeling on which 3 should be shortlisted.
Some very cool insights into homepage redesign process in this video, will use those in my workshops with non-marketers! Thanks, Tim! My question is, what arguments did you come up with that convinced the CEO that you NEED to look for an outside content agency? After all, one could argue, you are the marketer, you already publish - why not take on yet another "easy" content task? #TIM30
well.. Ahrefs is a pretty large company/brand at that point. So everything we do we want to do on the highest quality level. Yes, we do have many smart and talented marking people here at Ahrefs, but none of them has experience writing copy for multiple SaaS companies in multiple verticals. And besides, with all the market and customer research that went into it, we'd have to put aside some other marketing tasks that our team is busy with. So that is how we decided to hire professional help. As for hiring not one but 3 agencies, I'm sure many people have already guessed that in part that was done to be able to generate some word of mouth around our new homepage :) Like I said in the video, having 3 copywriting agencies work on your homepage is a total overkill.
At what point in your homepage revamp did you discuss 'Buyer Personas' if at all? Were they defined by you or the copywriting agencies you hired? Thanks, as always great content. #TIM30 =)
#tim30 I wonder about the A/B test opinion. Even a 6 month test linked to ARR would be worth putting place and revising. Knowing in 6 months time is better than never. Consider the 80/20 rule much of the changes you've made could be ditched. Wouldn't it be nice to know scientifically which bits are and aren't adding value? Perhaps you could even run a MVT with all the lovely traffic.. yes, linking to ARR could be hard, but that could get you closer to launching Dmitry's 90/10 search engine ;)
well.. wouldn't it be odd if for 6 months some people would see the old homepage and some people would see the new one? :) I think the new page would inevitably "leak" and many people would be visiting it just to browse around, skewing some of our metrics. In any case, after the release of the new homepage we had our highest growth month in quite a long while. I personally consider it a nice "sequential A/B test" :)
Thank you for sharing this Tim. I just wanted to know-how has the new page helped you? Business-wise. What exact metrics that you follow when it comes to the homepage? Is it the traffic or the number of clicks in the CTA or cumulative ones(the others Ahrefs keep track)? #Tim30
I personally was watching the total number of daily signups and weekly ARR growth. Since the release of the new page both numbers went up quite visibly. There might be some other contributing factors as to why we get more signups and started growing faster. But at least the new page doesn't seem to be a flop, like some people thought :)
Thank you for a such a detailed and explanation. Would be interesting if you can share more thoughts on choosing the right positioning for your product. #tim30
Thanks for the background of this story :) What is the reason for changing "just" the homepage and not the signup page or the pricing page. Now it has a different style and it deteriorates the feeling from the whole signup process. I assume that if you devoted less time on the HP itself but more time on those 3 pages (HP > pricing > free trial page) you would get better results. Thanks #TIM30
other pages are currently in the process of being updated too :) We just didn't see a big issue with updating homepage first, while leaving the rest of the website intact.
I like the redesign, it's different than anything else on the market. Your anchors are not working properly under "Ahrefs is an All-in-one SEO toolset, with free Learning materials and a passionate Community & support" #tim30
Great insights. Also, you may test this new homepage on account of gross revenue. But isn't is possible that your revenue will be impacted by other factors? Eg. your marketing tactics, or even campaigns from your competitors perform well or bad. #tim30
well, if all the marketing tactics will split all the signups evenly between new and old homepage variations, then the A/B test should not be influenced by anything. The only thing that bothers me is that people who signed up via old homepage, might discover the new one, which would influence their decision to stay and bring the results from two homepage variations closer together
@@ahrefspodcast "Ahrefs is an All-in-one SEO toolset, with free Learning materials and a passionate Community & support". You can have a passionate community, but you can't have a 'passionate support.' It's poor grammar. Also, the copy shouldn't use random capitalisation, i.e.: 'Learning' and 'Community' should be learning and community. "Junior marketer or SEO expert, you'll find Ahrefs' toolset powerful yet easy to master", doesn't make sense. It needs a question mark after the word expert to be grammatically correct. Also, the use of the possessive apostrophe is incorrect; it should be 'Ahrefs's'. It should read: "Junior marketer or SEO expert? You'll find Ahrefs's toolset powerful yet easy to master." This line needs correcting too: "Ask any questions to our customer support team who is happy to talk to you 24/5 via our support chat. Or just simply poke us on Twitter." Should read: "Ask any questions of our customer support team members who are happy to talk to you 24/5 via our support chat. Or you can poke us on Twitter."
Tim, great video. Your Spanish homepage has some English words mixed in between, i just thought it would be good for you to fix that, hope it helps!. Pablo Suarez #tim30
Why did you feel the need for a copywriter? Don't you know your product & market better than anyone? Looking at the result I can't see why you couldn't do this in-house?
true - we know our product and our market better than anyone.. But at the same time this could be the "curse of knowledge," which can lead us astray. And besides, it's always useful to have an extra pair of eyes to look into it, especially if those are professional eyes :)
OK, first of all I don't want to discourage you: Ahrefs evidently knows his business, and I am one of his clients. But there is an objective fact: You have absolutely no idea what an A / B test is. You couldn't be more wrong, so half of the video is an aberration. You would NEVER have to build 3 versions if you really knew what an A / B test is. And as a UXD professional, I'm pretty surprised that none of those companies explained to you not only what an A / B test is, but also amazed why they didn't explain HOW to measure which version to use scientifically and error-free. It is basic in UXD, which presents me with two possible scenarios: Either you did not hire a UXD company, or maybe you did but decided to ignore their advice
@@ahrefspodcast sure, there's plenty of resources just doing a Google search dorve.com/blog/statistics/introduction-to-a-b-multivariate-testing/ vwo.com/ab-testing/#9-mistakes-to-avoid-while-a-b-testing hbr.org/2017/06/a-refresher-on-ab-testing unbounce.com/landing-page-articles/what-is-ab-testing/ www.abtasty.com/blog/learn-from-5-ab-test-case-studies/ If you read these articles, you will see that you were never even close to the possibility of A / B. A / B is only for a small variation of an individual page (a CTA, a hero, a slogan, a button color). In the best case scenario, you would be doing MVT (multivariate testing), but that would require taking a lot of licenses, since MVT compares many differences but still in the same design. Fundamentally: none of these tests exist BEFORE the (almost) final design exists. In UXD, a user case like yours would simply be a user preference test that is performed before approving a design for development. You can do it remote or local, but always with real user (not stakeholders) And it is not "the way I see it". It is the way that absolutely all UXD and HCI books, documents and papers see it. This is why I am greatly surprised that 3 UXD companies do not know this as it is learned in the first months of the career. Sounds more like they didn't say it for some reason
⌚ Timecodes:
00:00 Introduction
00:26 The evolution of Ahrefs’ homepage
01:44 Our concerns with the old homepage
04:00 Three concepts of a new homepage
06:45 Market perception vs company’s vision
09:06 Copy vs design
11:04 A/B testing
12:23 Was it worth it and should you do the same?
Tip: Start with the timecode 0:00 and put the list of timecodes in the description, so you can leverage RUclips's new feature: twitter.com/RUclips/status/1266051626562990082
@@ViorelMocanu thanks for the tip, mate! will try it out! :)
I don't think you'd need so much information to get a statistically significant A/B test. I'd measure trial signups, and also measure how many of those trial signups turn into paid customers over time (although there are many more factors in this regarding followup). If I were A/B testing a homepage, I'd be most interested in measuring bounce rate, exit rate, what next pages they go to, and trial signups.
for each metric you want to A/B test you need to come up with a logical hypothesis on what value you aim for and why. High bounce rate is not necessarily bad. If a person realises fast that the product is not for them and leaves swiftly - that is cool. Less strain on support team, less refunds, etc. "What next pages they go to" is also something you're in control of. So before A/B testing this, you need to formulate a hypothethis on which pages we WANT them to go to and why. And we need to formulate which CTR we will consider success and which would be failure. And the more you think about this, the more complex your research will get. So our strategy was to release the page the way we want it and make sure it won't plummet our growth rate. So far I can say it worked :)
PS: thanks for the comment! Always good to have an open discussion and see different opinions.
PPS: I'm not advising that everyone else should make changes on their website without testing it first. I'm just sharing how we do things at Ahrefs. Some more details will be shared in the very next episode, two weeks from now ;)
@@ahrefspodcast I appreciate the detailed response. It does seem a bit like overthinking though, and if you make the issue so complex, I can understand how it's tempting to just not do it at all.
"High bounce rate is not necessarily bad. If a person realises fast that the product is not for them and leaves swiftly - that is cool. Less strain on support team, less refunds, etc." - You could say the same thing in inverse. A high bounce rate is not necessarily good if it's also resulting in missed signups. Someone lands on a page out of interest for a product or service, and it's up to that page to help them make a decision. When I think "What if I converted more traffic?" the first thing to come to mind certainly isn't "It's because they aren't my ideal customer, and they'll want refunds."
I'm certainly making some big guesses on how much traffic your homepage gets in regards to A/B testing viability. I figured you'd be able to get 100k+ visitors per test pretty quickly, and that will give you valuable insights. With such drastically different pages, the differences in these metrics are likely to be significant. You'll also get very interesting insights depending on the traffic source for each page variation. If one page has a 25% higher bounce/exit rate than the other, I think it's significant enough to be considered. Then while considering this, I would compare trial signups, and the rate of trial -> paid conversions. If you test, a winner shall be revealed.
This is all based on another principle for webpage design that comes from my PPC background (This is what happens when an ad guy gets into SEO) - the page is never totally complete. The internet moves so quickly now, and with the amount of data that we have available, it makes sense to update pages as it makes sense. SEO's typically get this when it comes to updating old content for "freshness" or to adhere to new ranking factors, but the same concept can (and likely should) apply to any other important pages online. After all, what would the impact of a 1% lift in conversion rate (trial signups -> paid) have on ahrefs' profits?
@@MorganCrozier "my PPC background" - paid traffic is usually "cold," so it makes sense to test any variable that will help you convert it. We don't drive any paid traffic to Ahrefs homepage, so people who open this page are (for the most part) well educated about what Ahrefs is (mainly thanks to our blog traffic and youtube views).. So we're not THAT cautious with making changes to our homepage. As I saw someone saying in a thread discussing our new homepage - we could make the whole thing blank and only leave the trial button, and people still would sign up. I think that person had a point :)
Great video! I've come full circle in my career from managing website redesigns from the design, copy, and implementation side - first at agencies, then in-house, and now as a copywriter. Understanding this perspective as a copywriter / consultant is really helpful. I'm always sure to partner with the design team as my very first step in the process. I'm going to use your video to help educate clients! (even if it means they bring in 2 other copywriters/agencies :)
thanks for your comment, Kimberly! :) Glad you enjoyed this video :)
Very refreshing to hear a more value & perception driven reasoning for the changes. I agree that A/B testing is a very valuable tool in every online marketers tool box, but don't make it too easy for yourself by thinking it will fix all your problems. Well done. #TIM30
I was really curious about this. When I saw the change in the homepage, the first thought I had was....I don't like it. But the next thought I had was...They probably have data that proves it works better. Thanks for the behind-the-scenes look at how you made that decision. I too am currently ramping up the redesign, and copy writing for my book marketing software's homepage. #TIM30
In all honesty, when I first saw the new design I didn't like it too :) I thought maybe our designer is trolling me. But then I realised that it's just my cognitive bias. I got used to the old homepage so much, that my brain couldn't appreciate the novelty and it's awesomeness. But now, after some time, I LOVE our new homepage. Remember the launch of iPhone X and how everyone made fun of that stupid "eyebrow?" Well, the sales beat the record and other brands started copying that eyebrow on their phones ¯\_(ツ)_/¯
Great idea to revamp your homepage! I know not everyone will be able to spend that much $$ but awesome to see how much thought goes into it and hopefully will cause businesses to think deeper before changing their homepage. Thank you! #TIM30
Thanks, Brady! :)
I liked how clearly your explained the challenges with A/B testing and factors you would need to consider (font, color, sizing, copy, metrics to measure etc.) This makes it very complicated to A/B test. Also really interesting to see you and the team take the best aspects of every copywriter and do what you thought was best for Ahrefs (not just accepting what was delivered). And finally setting expectations early on in the process definitely makes the whole process go lot smoother. I hope your CEO is happy with you 😀 #TIM30
thanks for the comment, Kunal! Yeah, it was a great experience overall :)
I should say that you are a brave man, Tim! Launching a new homepage during pandemic was a bold move - seems proper assessment of the results may be a headache (some marketers would sign up if not covid impact on their business), at least we have this problem in our business now. #TIM30
It wasn't my personal decision to release it now. It was made collegially ;)
Congratulations, Kateryna! You've won a 30 minute Skype call :) Please email george.dolgikh@ahrefs.com for details
@SaaS Marketing Vlog WOW! That made my day! Thank you!😊
Guys, you're using the "backlink" HTML tag as a brand. You're the backlink tool by default. It's in your brand naming.
what he said, i suggest something very inclusive like document.querySelectorAll('*').com
@@garethjax LOL
true.. but links is what makes the web the web. It is also at the core of Google's PageRank algorithm, that has changed the web. So not the worst brand name for an SEO tool at the end of the day
#Tim30 It is refreshing to hear there is no "one" way to build a homepage, but that is also quite scary. We did enjoy your suggestion to hire an outside party as a consultant to get a different perspective on the project.
Glad you enjoyed it!
This is super helpful Tim, especially as we are in the process of redefining our homepage ourselves. Love the confidence in backing your collective wisdom and going for it. Your input on where we're at would be awesome #TIM30
thanks! :)
Your homepage and that sleek typeface look great, Tim. I can't wait to understand the thinking behind each of the elements on the page. I do have a question though: in most contexts, especially in large companies, there's a certain amount of friction against having one page look a certain way and the rest of the pages look another way. How did you reconcile that, or did you even have that issue to begin with? This leads to a second question: do you think it's substantially more constructive to redesign or realign a website one page at a time, or do the whole thing at once as a "big bang" launch? #TIM30
great question! As you can tell, we released new homepage, while leaving the rest of the pages intact. Yes, this might lead to a slightly confusing experience to some visitors, but we don't see this hurting us in a substantial way. Within about a month we plan to adjust the design of the most popular pages, and in a few more months our entire website would be aligned in terms of design. Bottomline: we decided it is more convenient for us to gradually update our pages, rather than roll out the new design everywhere at once.
@@ahrefspodcast Yeah, that seems like a good strategy, considering there's no real pressure to push the new design super fast across the site, it would probably also allow you to spend more time on details and measurements to fine-tune each page if required, instead of panicking and having to audit and edit dozens of pages at once.
Great new Homepage, Tim! How did you tackle the translation part of that as your site seems to be available in 13 languages.
We have some outsourced translators, who keep our website and our platform up-to date with their native language :)
Very interesting approach and a bit bold to do such a drastic change without testing :-)
But you bring up an interesting problem for most SaaS type businesses; how to measure conversions...
Maybe you could do a video focused on your takes on KPI's and the problems that they all have?
#tim30 :-)
our KPI is very simple - it's our monthly growth and our ARR. Every time we start discussing some other "intermediary" metric that we could possibly measure we come up with all sorts of arguments on why that measurement might not be objective and why knowing the exact value of that metric won't really affect our business/product/marketing decisions.
Hi Tim, thanks for these videos. In my opinion, the design and copy looks great but the TYPEFACE, the Typeface could have been different. The blog reading was better with the previous typeface. #TIM30
can't please everyone, right? I personally am in love with our new typeface :)
@@ahrefspodcast Hahaha, yes sir it is subjective.
I'm interested if you can share some details regarding the process of hiring the 3 agencies. Although, it's probably a topic of a whole new episode.
I guess you had different sources to collect ideas and start your initial research. Such as recommendations from fellow marketers, hiring websites, forums, WOM etc. As far as I remember, there was also a post in the Ahrefs insider FB group to gather additional ideas. Also, was it tough to pick exactly those 3 agencies? For example, probably there was a moment when you narrowed your list down to 5-6 options. From there, how did you select the chosen 3 agencies to work with?
#tim30
PS: I really hate when I talk to people in our industry and when I mention them about how great of a tool Ahrefs is, they say smth like this:
- Ahrefs?! Yeah, I know them- they are great at finding links!"
and their only association is with links. So sad.. but yeah, hopefully over time more and more people will get familiar with the true power of the all-in-one toolset. I guess we can't blame them all for their ignorance. But growing your blog & youtube audience definitely helps to achieve this mission.
glad to see you here, Jason! :) Yeah, the way I found those three agencies was purely through recommendations of my friends/peers. Luckily I've made quite a few friends in the online marketing / SaaS space, so asking them for their advice on copywriters who they have worked with was all I had to do to come up with an initial list. From there I only did a little bit of research on each agency and went with my gut feeling on which 3 should be shortlisted.
sounds like a solid approach, thanks! :)
Some very cool insights into homepage redesign process in this video, will use those in my workshops with non-marketers! Thanks, Tim!
My question is, what arguments did you come up with that convinced the CEO that you NEED to look for an outside content agency? After all, one could argue, you are the marketer, you already publish - why not take on yet another "easy" content task? #TIM30
well.. Ahrefs is a pretty large company/brand at that point. So everything we do we want to do on the highest quality level. Yes, we do have many smart and talented marking people here at Ahrefs, but none of them has experience writing copy for multiple SaaS companies in multiple verticals. And besides, with all the market and customer research that went into it, we'd have to put aside some other marketing tasks that our team is busy with. So that is how we decided to hire professional help. As for hiring not one but 3 agencies, I'm sure many people have already guessed that in part that was done to be able to generate some word of mouth around our new homepage :) Like I said in the video, having 3 copywriting agencies work on your homepage is a total overkill.
At what point in your homepage revamp did you discuss 'Buyer Personas' if at all? Were they defined by you or the copywriting agencies you hired? Thanks, as always great content. #TIM30 =)
#tim30 I wonder about the A/B test opinion. Even a 6 month test linked to ARR would be worth putting place and revising. Knowing in 6 months time is better than never.
Consider the 80/20 rule much of the changes you've made could be ditched. Wouldn't it be nice to know scientifically which bits are and aren't adding value? Perhaps you could even run a MVT with all the lovely traffic.. yes, linking to ARR could be hard, but that could get you closer to launching Dmitry's 90/10 search engine ;)
well.. wouldn't it be odd if for 6 months some people would see the old homepage and some people would see the new one? :) I think the new page would inevitably "leak" and many people would be visiting it just to browse around, skewing some of our metrics. In any case, after the release of the new homepage we had our highest growth month in quite a long while. I personally consider it a nice "sequential A/B test" :)
Thank you for sharing this Tim. I just wanted to know-how has the new page helped you? Business-wise. What exact metrics that you follow when it comes to the homepage? Is it the traffic or the number of clicks in the CTA or cumulative ones(the others Ahrefs keep track)? #Tim30
I personally was watching the total number of daily signups and weekly ARR growth. Since the release of the new page both numbers went up quite visibly. There might be some other contributing factors as to why we get more signups and started growing faster. But at least the new page doesn't seem to be a flop, like some people thought :)
@@ahrefspodcast I guess the gamble has paid work. Kudos to you and your team. Looking to see more videos from your end. #Tim30
I just hate the new design. hashtag #honest feedback
it's perfectly ok not to like it. Some people don't like Thirty Seconds to Mars, but it doesn't mean their music is bad, right? :)
Thank you for a such a detailed and explanation. Would be interesting if you can share more thoughts on choosing the right positioning for your product.
#tim30
that's a tough question :) I'll see if I can do an episode with Ahrefs' thoughts on positioning :)
Did you paid those agency ? Even those you grab couple ideas on each one them #tim30
We paid all three agencies their full fee. They did their work and that work had to be paid for, regardless of how we used it afterwards.
Thanks for the background of this story :) What is the reason for changing "just" the homepage and not the signup page or the pricing page. Now it has a different style and it deteriorates the feeling from the whole signup process. I assume that if you devoted less time on the HP itself but more time on those 3 pages (HP > pricing > free trial page) you would get better results. Thanks #TIM30
other pages are currently in the process of being updated too :) We just didn't see a big issue with updating homepage first, while leaving the rest of the website intact.
I'm certainly interested in how you eventually decided that those 3 agencies would be the best combination, if you are willing to share #TIM30
I just asked for some recommendations from people who I respect and shortlisted the professionals who had the most "votes" :)
How did you use ahrefs research tools to help you with your copy? #TIM30
there was no need to use Ahrefs tools/date for our homepage copy, because we didn't target any SEO keywords.
As a long term user , the new UI and copy is poo. It almost feels like some generic guru BS
Sorry to hear that. Which part of the copy exactly reminds you of "generic guru BS?"
great experiment, #team30. I'd love to get on a call with you to learn more.
I like the redesign, it's different than anything else on the market.
Your anchors are not working properly under "Ahrefs is an All-in-one SEO toolset, with free Learning materials and a passionate Community & support"
#tim30
uh-oh.. they worked before.. something went wrong as we were making small changes to the page. Will ping the guys to fix it
Great insights. Also, you may test this new homepage on account of gross revenue. But isn't is possible that your revenue will be impacted by other factors? Eg. your marketing tactics, or even campaigns from your competitors perform well or bad. #tim30
well, if all the marketing tactics will split all the signups evenly between new and old homepage variations, then the A/B test should not be influenced by anything. The only thing that bothers me is that people who signed up via old homepage, might discover the new one, which would influence their decision to stay and bring the results from two homepage variations closer together
Were these copyrighting agencies native English speakers? There are some minor grammatical errors on your front page.
Can you point them out?
yeah, can you please point out those errors please?
@@ahrefspodcast "Ahrefs is an All-in-one SEO toolset, with free Learning materials and a passionate Community & support".
You can have a passionate community, but you can't have a 'passionate support.' It's poor grammar. Also, the copy shouldn't use random capitalisation, i.e.: 'Learning' and 'Community' should be learning and community.
"Junior marketer or SEO expert, you'll find Ahrefs' toolset powerful yet easy to master", doesn't make sense. It needs a question mark after the word expert to be grammatically correct. Also, the use of the possessive apostrophe is incorrect; it should be 'Ahrefs's'.
It should read: "Junior marketer or SEO expert? You'll find Ahrefs's toolset powerful yet easy to master."
This line needs correcting too:
"Ask any questions to our customer support team who is happy to talk to you 24/5 via our support chat. Or just simply poke us on Twitter."
Should read:
"Ask any questions of our customer support team members who are happy to talk to you 24/5 via our support chat. Or you can poke us on Twitter."
Tim, great video. Your Spanish homepage has some English words mixed in between, i just thought it would be good for you to fix that, hope it helps!. Pablo Suarez #tim30
oops.. thanks for letting me know, we'll dig into it
Great Home Page
🔥🔥🔥🔥
#TIM30
#TIM30
When can we connect for our 30 minutes call?
Thank you for the valuable insights! #tim30
Thanks, Sachin!
Why did you feel the need for a copywriter? Don't you know your product & market better than anyone? Looking at the result I can't see why you couldn't do this in-house?
true - we know our product and our market better than anyone.. But at the same time this could be the "curse of knowledge," which can lead us astray. And besides, it's always useful to have an extra pair of eyes to look into it, especially if those are professional eyes :)
OK, first of all I don't want to discourage you: Ahrefs evidently knows his business, and I am one of his clients.
But there is an objective fact: You have absolutely no idea what an A / B test is. You couldn't be more wrong, so half of the video is an aberration. You would NEVER have to build 3 versions if you really knew what an A / B test is.
And as a UXD professional, I'm pretty surprised that none of those companies explained to you not only what an A / B test is, but also amazed why they didn't explain HOW to measure which version to use scientifically and error-free. It is basic in UXD, which presents me with two possible scenarios: Either you did not hire a UXD company, or maybe you did but decided to ignore their advice
can you please post the links to any articles/resources that would explain what an A/B test is, the way you see it?
@@ahrefspodcast sure, there's plenty of resources just doing a Google search
dorve.com/blog/statistics/introduction-to-a-b-multivariate-testing/
vwo.com/ab-testing/#9-mistakes-to-avoid-while-a-b-testing
hbr.org/2017/06/a-refresher-on-ab-testing
unbounce.com/landing-page-articles/what-is-ab-testing/
www.abtasty.com/blog/learn-from-5-ab-test-case-studies/
If you read these articles, you will see that you were never even close to the possibility of A / B. A / B is only for a small variation of an individual page (a CTA, a hero, a slogan, a button color). In the best case scenario, you would be doing MVT (multivariate testing), but that would require taking a lot of licenses, since MVT compares many differences but still in the same design. Fundamentally: none of these tests exist BEFORE the (almost) final design exists.
In UXD, a user case like yours would simply be a user preference test that is performed before approving a design for development. You can do it remote or local, but always with real user (not stakeholders)
And it is not "the way I see it". It is the way that absolutely all UXD and HCI books, documents and papers see it. This is why I am greatly surprised that 3 UXD companies do not know this as it is learned in the first months of the career. Sounds more like they didn't say it for some reason
and still its looking too old and boring.
are you talking about our homepage or about me? LOL
@@ahrefspodcast Home page of ahrefs
Not mobile friendly after login...Menus are mixing to each other..
A lot of money for nothing really