Episode 150 | Ten Ideas for Things You Can A/B Test

Show Notes


[00:00] Mike: This is Startups for the Rest of Us, Episode 150.

[00:03] Music

[00:10] Welcome to Startups for the Rest of Us, the podcast that helps developers, designers and entrepreneurs be awesome at launching software products, whether you’ve built your first product or you’re just thinking about it. I’m Mike.

[00:19] Rob: And I’m Rob.

[00:20] Mike: We’re here to share our experiences to help you avoid the same mistakes we’ve made. What’s the word this week, Rob?

[00:25] Rob: Well a day or reckoning is here for Drip. Remember the prelaunch email I sent out to 300 something folks a few weeks ago?

[00:31] Mike: Yeah.

[00:32] Rob: Well, we got a bunch of them into trials. They got on boarded and late tonight, sometime after midnight a bunch of them are either going to convert or not convert to paying customers. It’s been exactly three weeks and we have a 21 day trial. So rest assured you can imagine me this evening sitting with very large glass of bourbon and waiting for the news and watching for the billing email to come in. Either way, it’s either going to be a great news and I’m going to celebrate or it’s going to be very bad news and I’m going to drown my sorrows in that glass.

[01:03] Mike: I see. I mean do you have it setup so that people can cancel before it gets to that point? Are they notified or…

[01:08] Rob: Absolutely. Yes. So three days before typically with like a 21 or a 30 day trial, emailing goes out three days prior and it says look, you’re going to be billed or if we don’t have their credit card on file because there’s a few folks that got a no credit card trial early on then we ask them hey, if you want to continue you need to enter your credit card. So everybody gets notification. That’s the idea.

[01:29] And as always, with all of my apps, if they do get billed tonight and they come in tomorrow and they’re like I meant to cancel, we refund them and we cancel their accounts. So it’s not about tricking people into doing it. It’s about letting them know and keeping them happy as customers.

[01:44] Mike: Right. But do you have any sense of how well it’s going?

[01:47] Rob: I do. I’m the cautious optimist. I am excited about it. I’m very pleased with the numbers that I have but I hate talking about things in public like that because if they don’t turn out I feel like I made a big mistake or I misjudge things and then be able in retrospect to talk about it in and the lessons I’ve learned.

[02:04] Mike: Cool.

[02:06] Rob: How about you? What’s the update with Audit Shark? After last week’s episode we’ve already gotten several comments and a couple emails. So I’m interested to hear what’s new?

[02:13] Mike: Well I’m labeling the new UI overhaul of Audit Shark as code complete enough. Essentially that means things are well enough that I can basically push it over the top of the existing version. I haven’t had time to do that. I just haven’t had time to try and push that out to the cloud yet. But there’s remediation steps are being added in to the database and should be done in the next week or two. I talked to the policy developer about that. There’s a few things that need to go into the dashboard but those can be added in over time so I’m not real worried about that.

[02:43] And then there’s other little things here and there like in app walk through, some documentation need to be added. But I think those are minor enough that I can push it out over the top of the existing version and just kind of be done with it and call it code complete enough.

[02:57] Rob: Cool. I re-listened to last week’s episode after the editing and there are a couple of things that I think I’d ask that wound up on the cutting room floor so I wanted to bring them up. One was regarding remediation and you know that customers want that but from what I’ve seen, you haven’t asked the customers who’ve used it if they will pay you at that point if it’s worth it or if – I talked about this is your biggest risk right now. Right? Is not knowing if they’ll pay you.

[03:21] So I guess in the back of my mind it’s like that being your biggest risk, it seems like if you asked them before you implement it then at least you have an idea of whether or not you’re going to be able to charge them for it once remediation is in. Have you thought about that?

[03:33] Mike: I didn’t because I looked at it from a perspective of some of the early feedback I got was okay I’ve done this stuff, now what do I do? It’s like if you used the application and you get to the point where it gives you the results and you just don’t know what to do then that’s a failure on my part. It essentially means that I need to present the information about how to go about fixing that stuff.

[03:55] And when somebody sent me a screencast it actually had some of that information in there. And they just said I’m trying to match these up between what the rules are and what the results are and it’s hard to do and it doesn’t tell me what I’m supposed to do to fix this. I don’t quite understand. I knew that was kind of a gap but I wasn’t sure if I’d b able to get away with it. But it seems pretty clear just from the – you know how you can hear things in a screencast or a video that are a little bit – you get the intonations and things like that you definitely don’t…

[04:22] Rob: Nuance.

[04:23] Mike: Yes. The subtle nuances about how somebody says something and it gives you the impression that’s really, really important. So I have not out right asked but my impression was yes, this is extremely important. It’s got to be there.

[04:38] Rob: Right. And I don’t doubt that, that remediation is something you have to have in advice, at least information on how people can fix their stuff. The thing that I’m wondering is is it that enough? Is that the point at which Audit Shark provides enough value that someone will pay you? Or in say two weeks to get remediation done, is it okay? Here’s my early access people. Now we have remediation. Now are you willing to pay for it? And they’re like well, not really because you still have to build out these other few things. And that’s where the launches just go on and on and on or you kind of never get to launch.

[05:07] So that’s what I’m curious about. I’d love to see you do either some emails or just some interviews, quick Skype calls with some of your early access people and ask them if Audit Shark had this in it, is that the point at which you’re willing to pay my price for it?

[05:22] Mike: That’s a great question. I definitely have to do that. That’s kind of next up on the docket. I mean I have to start reaching out to them and finding out more of that information. One of the things that has come up, a lot of the stuff in the news lately about the NSA spying on people and gathering information from them and there’s questions about sending data from their servers back to mine which are stored on Microsoft’s azure platform. So it’s like well if the NSA has access to those then do they have access to my data and it’s just kind of an open ended question that in some ways has to be addressed in a technical level but it also has to be addressed in a marketing level too.

[05:56] But it did give me an idea about if I can address it at the technical level, would that be a leverage point that I could use to get into additional customers. Hey do you want to make sure all of your stuff is protected from the eyes of the NSA? Not that it’s a guarantee but I can certainly help lock down those machines so that people can’t get in inadvertently…

[06:17] Rob: Very good.

[06:18] Mike: But I totally get where you’re coming from where like there’s certain questions that I have to start asking and that’s definitely one of them.

[06:23] Rob: So in other news, I’ve started doing a little bit of angel investing. It’s interesting. I looked back about 8 months at some goals that I have for 2013 and I really wanted to get more involved with just a handful of startups. I’m only going to do maybe four or five a year and they’re very small amounts like $5,000 to $10,000 certainly until I get the hang of it. But I don’t know, I just find it intriguing to be kind of part of other startups that I’m not running and to have some kind of stake in them even if it’s a small one, I really enjoyed being part of the WP engine investors just hearing the news and watching the growth.

[07:01] It’s that interesting thing of taking those steps up the ladder where you’re an employee and you’re self employed then you’re a business owner and then you’re an investor. Those are kind of – I think of Robert Kiyosaki. He did lay that out. And I liked that idea. So to me I see investing and being able to make money from that is really kind of a high tier of doing things. And I have some cash sitting around from all of my businesses and since I’m not acquiring my next one I feel like it’s just not working for me. Right? It’s not actually doing work. It’s not earning anything sitting in a bank account.

[07:35] So I’ve invested in one here locally that’s pretty cool, kind of a little tech hub software cluster thing that’s going on. And then I think about 3 or 4 others in the last couple of months.

[07:43] Mike: That’s really cool. Yeah, you have to let us know how that goes. I mean I understand there’s only so much that you can say about them or how it’s structured and everything but it’d be interesting to see how that progression goes because I totally get the idea of being a full time employee and moving up to starting your own company and building your own stuff and going on to invest in other peoples.

[08:02] Rob: You know I’m a bootstrap founder right? I certainly believe in raising funding and I think there are cases when funding is a good thing. I think at some point I’ll probably – once I get the hang of it and have any type of insight into it, I’ll probably talk about it on the podcast. But at this point I’m just such a newbie at it and it’s not even worth really hearing my thoughts yet.

[08:21] Mike: The only other things I have going on as I talked about before I was doing some paid advertising campaigns through Facebook and they still need some refinement. Last week I was getting a lot of noise because some of the links that I had pushed out there to some of the landing pages I wasn’t using the right query strings so now at this point I’m starting to get good data but one of the other things that I’ve tweaked was the twitter strategy that I’m using.

[08:45] In just the past week, the twitter followers for the Audit Shark account have kind of stalled out about 180 followers or so. And over the past week I’ve added another 75 followers. The traffic is – I can see it in the analytics its actually translated back into website visits for Audit Shark and a landing page sign ups and that’s nice to see.

[09:04] Rob: That is nice. Instinctively man, I would totally not do twitter at this point. I think it’s too early. Just the value of email, I would say it’s ten times more valuable but I really believe that and I know you’re saying it generated some people on your email list but I think prelaunch, I have a Drip twitter account but at this point it’s really just to kind of gather some followers. We’re not pushing anything out because I just don’t know the real value that’s going to get you at this point right? So early. Unless we’re going to do content marketing or you’re going to be really pushing a lot of stuff out which then gets really time consuming. I think focusing on building that email list is the way to go.

[09:40] Mike: Well the thing is dripping out that content on the Audit Shark twitter account is outsourced at this point. I don’t actually do it. I’m having somebody else do that. The other thing that I found is of the links that go back, about 30% of them are converted into emails.

[09:52] Rob: That is nice. That might be worth a thing. Yeah.

[09:57] Mike: So yeah. I mean it is worth it but there’s a question like how many twitter followers do I actually have and how many am I reaching? Right now that number is really small so that 30% I don’t put a lot of stock in it just because the sample size I think is too small. But it may turn out that if I’m able to amp this up over time, if I am able to kind of get a sustained following week in week out and scale this up then that could be a viable strategy moving forward and it could conceivably help take it to the next level.

[10:29] But because some of that stuff is kind of automated right now, I’m not too worried about spending my time on it. I obviously got a lot of other things to do but I’m just looking at the numbers themselves and making sure that I’m capturing all the information I need to make sure I understand whether it’s worth it or not.

[10:44] Rob: Right. And by automated you mean you have someone doing it. It’s not a bot.

[10:47] Mike: Yes.

[10:48] Rob: Right. Cool. Last point of discussion before we dive into kind of the main feature point of this episode, MicroConf Europe speaker Andy Brice, you might know him from successfulsoftware.net long time blogger he’s a solo founder and he has Perfect Table Plan which is his desktop software that helps people plan their seating charts at weddings. And he’s made a full time living off this for years, offers a lot of good insight on his blog successfulsoftware.net.

[11:15] He is hosting a two day training in Witlshire England in late November. I’m only getting up in the show notes. He has tested a ton of marketing approaches over the years. He has kind of that B to C desktop product. So if you’re anywhere within driving distance of Wiltshire, England, I would definitely consider taking him up on that offer.

[11:34] Music

[11:36] Mike: Today we’re going to be talking about 10 ideas for things that you can AB test in your business. I’m in the midst of working on a bunch of different landing pages for Audit Shark. In doing some of the landing page research, I started coming across a lot of AB testing sites which offer ideas on things that you can test that were not necessarily specific to landing pages.

[11:56] So in episode 148 we talked to Clay Collins specifically about some of the different marketing trends. We talked a little bit about the types of things that do and don’t work but I also wanted to discuss some of the specific tactics that I’ve come across. Not all of them are related to landing page design but some are just simply AB testing you can use to test out your marketing funnel or to use on your pricing page and things like that.

[12:18] I thought it would be a good idea to talk about some of these different things that you can try, not that anyone who’s listening to this is going to try all 10 of them but they’re just interesting ideas that may work in your business or they may not even be applicable but I thought it’d be interesting to just kind of highlight some of the different things you could use.

[12:35] Rob: Right. So these are 10 things aside from headlines right? Because the headlines are the first thing that you should start with. That’s going to tend to be set the messaging up and everything so that’s kind of like number 0 and then starting with number 1, what do we have?

[12:49] Mike: So the first one would be to test whether or not a free trial button is going to work out and we’ll link this up from the show notes because several of these come from a blog article from Neil Patel. But one of the things he highlighted was that if you have a free trial button there, try taking it out or adding one in and letting people know that there is a free trial and seeing if those people go through your funnel and covert. I think the example that he had shown, there was something like 160% increase in signups by adding in a free trial button.

[13:18] Rob: And is that oppose to like a download now button if you’re trying to get say get someone just to get their email and to download something or versus a buy now button?

[13:28] Mike: It was versus a buy now button in that particular case.

[13:31] Rob: And I’m curious about all these. Right? I mean I’ve seen people send – let’s take paid acquisition or SEO traffic. I’ve seen them send it to a landing page and just asked for the email with a download now button offering some type of template or report or something like that. And then track those conversions over time versus just having that try now versus having the buy now.

[13:54] The interesting thing is that depending on the niche, depending on how good your messaging is, there no rule of thumb that I know of the best way to do that. Right? It really depends on you price point, how well you pre-sold someone with an ad, how well your landing page copy is. I think those are maybe some variations that you can think about not just the button itself but actually the whole kind of message of the landing page.

[14:16] Mike: Yeah. That’s exactly right. And again you have to think about this list that we’ve got here are things that you can try that some may make sense for your business, some may not. In terms of the question that you pointed out, is it a free trial versus a buy now or download now or sign up for an email list. The specific example that we’ll link to as I said was a buy now, but take that into consideration about what it is that you’re exactly trying to get people to do.

[14:39] So obviously you’re not going to have a free trial for an email, when you’re trying to capture an email. Free trial versus buy now and even the text on that free trial you may have some other text other than a free trial, 30 day trial or something along those lines. You can try different things.

[14:53] So the next one we have is asking for credit cards upfront versus later on. I think my inclination is to believe that if you’re collecting the credit cards upfront it’s going to essentially disqualify or weed out the people who are not interested in your product or they’re just kind of kicking the tires a little bit. They see that there’s a free trial to sign up for it but they may or may not necessarily be engaged with the product.

[15:14] And depending on how you kind of walk them through the sign up process and on boarding process, it’s going to have a lot of influence over whether or not they ultimately become a customer of yours. That said, I think collecting those credit cards reduce this friction later on and it’s not to say that having friction upfront is a bad thing or having it later on is a bad thing. It depends a lot as you said on your product. But this is certainly something that you can try. My inclination is to believe that capturing them upfront is probably better but it’s something you should definitely test in your business to figure out whether or not it does make a difference and how much of a difference it does make.

[15:49] Rob: I know companies who make both of these approaches work. I’ve said on the podcast before many times that I always default to asking for credit card upfront but the companies who make not asking for credit card upfront do some very specific things. They’re very experienced. Just reminds me a little bit of having a free plan in the sense that free plans can work in certain circumstances if you’re an expert. I feel like credit cards are the same thing.

[16:13] If you punt and you say I’m not going to ask for credit cards then you have to do some specific things during the trial. Most of the companies I know who are making it work are actually getting telephone number instead of credit card and they are having people call them. So if you’re not going to do that or you’re not going to have a well crafted trial email sequence that provides a lot of value that gets people on boarded and then that encourages them heavily towards the end to enter their credit card then you have failed. I don’t know anyone who is not doing that who is making the no credit card approach work.

[16:44] So credit card eliminate tire kickers. They bring in more qualified leads. They bring in people who when they request a feature from you, you can say well at least this person was into this, the value prop of my app enough that I can listen to them. Let’s say you run an experiment and you didn’t ask for credit card and you get 100 people in your free trial. Then you run the same experiment a month later and you get 20 people. So it’s 20%.

[17:09] Who are you going to listen to when people ask for features? You should listen to the people that are more qualified as a rule. And so especially in the early days of your app when you really are not sure what your value prop is, keep trying to figure out what to build. You want more qualified people because those are going to be the people that are more likely to pay you and the people that are more likely to get value out of your app long term.

[17:28] Early on, you don’t need more feedback. Trust me. I’m going through this right now. I have a feature list a mile long and my biggest task these days is weeding that down to which one should we build for the very specific audience that we serve and it’s going to get value from my app. So that’s where especially early on I think asking for credit cards is critical and then testing later on is definitely worthwhile but only once you know what you’re doing and you’re able to handle it with some expertise.

[17:55] Mike: I think the important piece that I heard out of that was the part where if you’re not asking for a credit card, you have to ask for other information. I was at a business software conference a couple years ago where there’s a company and I won’t say their name but they had a software that you could download and you had a trial. As part of that, they would ask you to fill out this information and then you’d get your free trial.

[18:18] Part of that, they would ask you for your contact information and they would actually have developers call and talk to the prospective customers and find out what they liked about the software, what they didn’t, whether they were going to purchase it or not. And they find out that by making those calls, their conversion rate increased something like 60%. It was ridiculous. And their average price point for those purchases was something like 40% more than if somebody had just simply purchased off the website.

[18:45] Rob: Right. So if you have the expertise and the man power to do that, if you think of some of the big venture funded startups like at Kiss Metrics, they don’t ask for credit card upfront but let’s be honest, they have a lot of money. They have a lot of people and they have a lot of folks making phone calls. They’re really kind of working it exactly the right way. And yeah, take another example like Hub Spot. It doesn’t look like they ask for credit card before free trial but go to their free trial page. They ask for first name, last name, email, phone, company name, website URL, number of employees on and there’s like four more questions literally and they’re all required.

[19:17] So yeah you’re right. They’re not asking for credit card but they’re asking for that all information. They are pre-qualifying people because they don’t want to waste their time chasing after tire kickers.

[19:25] Mike: And I’d be willing to bet that they have probably AB tested the heck out of that page just because I know how Dharmesh is in giving his predilection for going after stats and statistics and measuring everything.

[19:37] Rob: Absolutely.

[19:38] Mike: So the third idea for something that you can AB test is to try using trust symbols such as hacker skin and icons or SSL certificate icons, maybe even an extended validation certificate so you get the little green bar in the website browser. And I think these are more applicable on a landing page or where your – to actually taking order information, you’re trying to get them to follow through with an action and you’re trying to inspire confidence and trust.

[20:03] Sometimes you can do that with text but obviously there’s all these different trust icons that you can use from different vendors that will help inspire trust and confidence in your company and in your website.

[20:13] Rob: This is something I’ve never tested. I have added trust symbols to my pages but I’ve never tested conversion rates with trust symbols and without – I heard they work and improve and all this trusted app but I’d be interested if someone has data. I was having dinner with one of the founders of Foxy Cart who sponsored MicroConf last year. They have hosted shopping carts software.

[20:35] And one of the things he said was payment preferences like offering PayPal versus Amazon versus just entering your credit card number and the major differences that they see and one of the things he said was having a buy with Amazon button when someone’s purchasing a product on an e-commerce definitely raises conversions. So I think that’s another thing that you should consider is offering – it’s that easy low friction ability to pay by clicking that one button since you already have your credit card on file with Amazon.

[21:01] Mike: So the fourth idea is to use a live chat widget. This is actually something that I’m testing right now. I’m working with a company called Chatter Lime and essentially what they have is they have a live chat widget and I haven’t integrated into the site yet but the time this episodes goes live, it will be out there.

[21:18] But essentially the idea behind having a chat widget is if somebody has questions about your product, they come to your website, they’re not really quite sure what to do nobody’s going to go to the FAQ page and actually read it. But if there’s a help widget there that either pops up dynamically and asks if they have questions or if it’s just off to the side that they can interact with if they decide that they want to, then they can go ahead and do it.

[21:42] And the difference with something like Chatter Lime is that they actually have people who are manning that behind the scene so I don’t have to do it essentially off loading the capabilities to them and essentially providing them with script. But they can provide me with transcripts to let me know what the conversation was like, what sorts of questions were asked and then essentially enhance whatever the database of answers is that they’re going off of.

[22:05] I feel like this is a huge win in terms of being able to interact with your customers not just for me in Audit Shark but I think in general, most people would find that to be very, very helpful. Obviously I’m going to test it. They’ve see 30% increase in conversion rates in some cases but it seems like it’s definitely worth testing.

[22:23] Rob: I agree. An interesting part is there are a lot of these widgets coming out that you can stick in your site. We were talking before we started recording that you could have a Drip widget in the main area of your site and then once you get into kind of a checkout flow you could add the chat widget. So right when people are at the decision point they have questions about your pricing or your registration page, that’s when you can add a chat window and bravo to Chatter Lime because they’re following this trend that I brought up a couple episodes ago about the DFY versus DYI. It’s like done for you versus do it yourself.

[22:54] And it’s going up market. It’s basically charging more but then they don’t just provide a widget like a bunch of other places do. They’re actually adding that concierge element. So as a founder of Audit Shark, you’re willing to pay them more so that you don’t have to man it yourself or have one of your people doing it. So I definitely think we’re going to be seeing more and more of that kind of business market where you’re going up market and doing the software and service as a single offering.

[23:19] Mike: So number five is to evaluate the number of form fields. As you mentioned before with Hub Spot on the free trial sign up, they’re actually asking for a lot more information and the general rule of thumb is to ask for less. I think that with Hub Spot, what they’re doing is they’re asking for a lot more and in some cases especially if you’re selling any sort of products that deals with sensitive information, it almost seems like asking for more information is going to inspire more trust from the person who’s making the purchase because they’re essentially giving more information.

[23:50] Their vendor is prequalifying you. Maybe they’re looking into seeing who they’re actually working with. They’re doing a little bit more to protect the information. And whether that’s true or not is kind of immaterial. The ideas that if you’re giving them more information, then chances are you’re checking it out. And I’ll give you a very specific example of that. How would you feel about going and buying let’s say an eBook where all they’re asking for is your credit card number and the expiration date?

[24:16] Rob: Well maybe I’d want them to get my email so they can send me the book.

[24:19] Mike: Right. But if they’re not asking for like the CCID number or your name or address or zip code or any of that stuff, wouldn’t that strike you as a little bit odd?

[24:30] Rob: It wouldn’t to me but only because I know that all you need is a credit and an expiration to charge most cards and then you need CVV maybe as some added validation. So I don’t actually ask for address often times like if you go to the HitTail registration from, I don’t ask for address. I don’t think we ask for zip because stripe indicated at a zip is not – if I’m not doing AVS it’s not needed to do just the basic charging and it’s not going to help me get more charges through.

[24:54] Mike: I can definitely see that from programmer’s perspective but I think that from the average user’s perspective they don’t necessarily know all that stuff.

[25:01] Rob: Right. And I’m the man behind the curtain in that scenario.

[25:03] Mike: Right. And that’s kind of my point. But until you actually try this you don’t necessarily know. I think it depends a lot on your product and the type of service that you’re offering and price point and all that other stuff. But again it’s something else that you can test.

[25:15] One thing that I found interesting and this is number six that I came across was adding a sign up form directly to your home page. Have you seen people do that before? Instead of having a call to action where they click through and go to a pricing page or something like that, they actually have their sign up for embedded right into the home page.

[25:33] Rob: I have seen that. You can see an example at outright.com which is an accounting package. If you go to basecamp.com at least the version I’m seeing, I know they split test quite a bit. That one has start your free trial form right on the home page.

[25:46] Mike: Yeah. That’s not something I actually thought about a lot but I did come across something that people are testing to find out whether or not that works and converts or not.

[25:55] Rob: I think it’s an interesting question. I would tend to not do that. I think it’s worth testing but to me I want to optimize the site for first time visitors or for maybe returning visitors who are thinking about potentially signing up. So for first time visitors it’s very unlikely they’re going to land at that page and know that they want to sign up. You have to give them some information about what you are, what your pricing is, what the app does, what the benefits are.

[26:20] I’m not necessarily convinced that having that form right in front is really going to be that useful now. If Basecamp is testing and it’s worked for them and maybe they have kind of a different market or if they’re sending certain types of traffic to it, I think this comes back to something Clay Collins said a couple episodes ago where he talked about the two step opt in process and he talked about like a giving page versus a taking page.

[26:42] And if you arrive at a page and it’s instantly asking you for information, it may put some people off. So actually not having the form there but having that sign up for free trial then you’ve made them click a button to get to that next page in order to – they actually initiated some type of action in order to do it.

[26:58] Mike: I almost feel like there might be other stuff that’s probably lower hanging fruit that you could probably go after and push this off to later.

[27:04] Rob: Yup.

[27:05] Mike: So number seven is to test your testimonials with or without photos and names and websites and company names. This one struck me as a little bit odd. I wouldn’t have thought to start removing some of those things. I would’ve thought that the most compelling types of testimonials would be ones that have pretty much everything. It’s got the name. I think put in a face and a name and a company or a title or something along those lines, giving prospective buyers the confidence that other real people are actually using your product would be the way to go.

[27:36] Rob: I would agree. In my experience, video and photos with testimonials have increased conversions. Adding all the other stuff, adding a bunch of text, name, website, company name, all that stuff, maybe. I mean if people will recognize the company name that it has a meaning but really just having kind of a first name in a website URL or giving them an indicator that they’re legit is one thing but then adding a bunch of text to, I don’t know how much that helps. I think you’re going to have much more impact when dealing with a visual element. Like having that nice headshot there versus not having one or even having a very short small video testimonial.

[28:13] Mike: Number eight is to leverage different types of directional cues and directional cues can be anything from giant arrows that you’ve actually put on the screen or people looking or pointing on different stuff on the page. You can find these in a variety of different places. And I’ve actually done this on the Altiris training website where’s there’s a picture in the lower right corner of somebody who’s actually looking at the view pricing button.

[28:38] I haven’t done a lot of testing to see how much that has really affected it but I have kind of learned over the years that if you use those directional cues, they can help. And because the site doesn’t get enough traffic it’s just not worth AB testing it. But if you do have enough traffic then that’s something that you can look at.

[28:55] Rob: The directional arrow is just kind of an old internet marketing tactic people have used for years. I did hear that Brecht Palombo over on distressedpro.com he said he’s sending some – I think it’s paid traffic to a landing page and it has a head shot of a woman and she’s looking at his button like the call to action button and that’s something that I’ve heard people do. I’ve never used the arrows or people looking at or pointing that stuff but that’s not to say that it doesn’t work.

[29:21] Mike: Yeah. I think for something like this you really need to have software installed that actually takes a look where people’s mouse pointer is going like CrazyEgg or Inspectet or something along those lines so that you can get a much better sense of whether or not that’s working. Because I think that just judging the numbers may be a little difficult but if you can pin point on what people are actually moving their mouse over then that would help give you a better indicator whether or not it’s working.

[29:45] Number nine is to display honest reviews of your product or filtered reviews with your product. And there’s obviously a difference between filter reviews essentially tell everybody the good things about your product. But if you notice on Amazon.com there’s a lot of reviews there that tell you how awful a product is and all the different things that are wrong with it. If you go back, I think it’s still on the MicroConf website but there’s a talk that was done by Jason Cohen of wpengine.com where he talked about honesty.

[30:17] There’s some pretty compelling evidence into here that suggested that if you’re providing all of the negative reviews along with the positive reviews then you’re going to increase sales. I think one of the example he used was Kodak. Another one was Amazon, I mean there’s a lot of anecdotal evidence that suggest that using those is going to increase the conversions. I don’t know whether that translates directly to just using filter reviews. I would think that in the beginning you want to portray your product as well as you can. But it almost seems like if you’re selling something, it’s much more of a commodity you would want to show the negatives as well as the positives.

[30:52] Rob: Yeah. I haven’t seen software vendors do much with reviews. I guess if you’re a B to C vendor, well certainly if you’re mobile then there are going to be reviews in the app stores. But if you’re a B to B vendor, unless you’re on a market place like Capterra or some larger market place, on your own marketing websites, if you don’t have enough volume of actual reviews, you do get emails from people who cancel and they may say your product needs this, it didn’t do this.

[31:19] But that’s not really a review it’s more of maybe this product wasn’t a good fit or it’s just a complaint or something like that. So it’s a little harder to translate I think in the kind of B to B Saas space that I think about.

[31:28] Mike: And the tenth thing that you can AB test is using videos versus text and images. Virtually on any website you can take the content that’s there and translate it into a video of some kind. I don’t see a lot of people doing this to be perfectly honest. Don’t get me wrong. There’s tons and tons of video out there but it’s pretty rare for me to come across information where people are actively testing video versus text. I’ve seen it done a couple of times but there’s not a lot of yes I’ll say case studies that I’ve seen out there doing this.

[32:00] Rob: Yeah. I’ve done this once. I did it what DotNetInvoice with a landing kind of a squeeze page with a video versus a standard Saas landing page which had more text and images. And the video site or the video page just got crushed. But there could be a number of reasons for that. It just happened to do that. I ran it twice and at the time even the number of people who watched the video was less than half the people who hit the page.

[32:25] So that just killed my ability to get people interested in a trial and that’s what I was surprised by. I thought that with a short video that it would be easier to get people to watch it but my rule of thumb is to do text and images, build a standard page, start with that and from there, iterate. Because it’s like you said. It’s pretty easy to go from text and images and create a video out of that like an explainer video, a quick screen cast, something like that without even having to pay a lot of money for it. Those two things are not difficult to test back and forth once you have that video created.

[32:56] I would be interested in seeing some data on this. And I am in talks right now with my growth tech intern started and we’re trying to figure out the best way to test this with Drip because I am interested to see how that test would react to this audience.

[33:08] Mike: Yeah. I’ve definitely heard of cases where it goes the other way. I think that one that I’ve heard talked about in the past is Fog Creek where they’re selling fog bugs and they’ve tested it with Joel giving a talk and a presentation of fog bugs versus one that did not have him doing that. And the one with the video, the engagement was really high and it just absolutely crushed the text only version. I don’t have the numbers for it. I don’t think they published those but I’ve heard anecdotal evidences suggested in that particular case, one of the thoughts behind it was well people just loved to hear Joel talk which is entirely possible. I think it does depend a lot on the video that you’re offering and the audience.

[33:46] So just to recap, the first idea is to test free trial buttons and see if including them or excluding them is going to make a difference in your conversions, asking for credit cards upfront versus later on, using trust symbols. Number four is to use a chat widget of some kind. Number five is to evaluate the number of form fields that you have. Number six is add sign up fields directly to your homepage or landing page. Number seven is to test testimonials with or without some of the different information on them. Number eight is to use directional cues. Number nine is display honest versus filtered reviews. And Number ten is to test videos versus text and images.

[34:23] Music

[34:26] Rob: If you have a question for us call our voice mail number at 1-888-801-9690 or email us at questions@startupsfortherestofus.com. Our theme music is an excerpt from “We’re Outta Control” by MoOt used under Creative Commons. Subscribe to us in iTunes by searching for startups or via RSS at startupsfortherestofus.com where you’ll also find a full transcript of each episode. Thanks for listening. We’ll see you next time.

Twitter Digg Delicious Stumbleupon Technorati Facebook Email

9 Responses to “Episode 150 | Ten Ideas for Things You Can A/B Test”

  1. Hey Rob – when you said “[12:35] Rob: Right. So these are 10 things aside from headlines right? Because the headlines are the first thing that you should start with.” What do you mean by headlines? Do you mean your initial landing page where you ask for an email?

  2. These are 10 A/B tests that you can run aside from testing the H1 tags on your pages. For landing pages, you’ll typically have a very prominent headline that you want to also have in where the page is being linked from. But aside from A/B testing that headline, what are some other things that can be tested.

  3. Great podcast guys! And we’re excited to see what ChatterLime does for AuditShark. Aside from chat, we had a meeting this morning and plan on A/B testing the other things you two mentioned. Awesome ideas. Thanks!

  4. “So I definitely think we’re going to be seeing more and more of that kind of business market where you’re going up market and doing the software and service as a single offering.”

    This is the direction I’m going with my latest business. Instead of “software as a service” I think of it as “service enabling software”.

  5. @mike – I’d be cautious about going down the path of marketing using fear of the NSA.

    https://www.schneier.com/blog/archives/2013/09/how_to_remain_s.html is a good read. Bruce is in the know here, and his comment about “… if the NSA wants in to your computer, it’s in. Period.” They have access to zero day exploits and an understanding of weaknesses in current software. I’m sure Audit Shark does a great job of ensuring the computer is locked down but the NSA is a well funded adversary.

    The other half of the equation is that you’re a US citizen. If the NSA decides they want to put a tap on your network or have you send your customer’s data to it, you’re going to do it and not tell a soul or you’ll end up in prison. Or since you’re in Azure from what I remember, they’ll just get it from Microsoft and you won’t even know.

  6. The download MP3 appears to be corrupted 🙁

  7. @Sean:
    Yes, I already know all of that. Either I misspoke or didn’t explain it clearly. If someone wants into your computer bad enough, there’s ABSOLUTELY nothing you can do short of taking a hammer to the hard drive, and even that’s not always a sure thing. Sometimes it’s too late.

    But there’s a marketing angle here for educating people about security. It’s not just the NSA you should worry about. A headline like “NSA Got you down?” would draw attention, and then the specifics could be discussed.

    As an example, most people think that their email is secure because via Gmail (for example) it uses SSL to access the webpage. But it isn’t always transmitted using SSL over the rest of the internet. It depends on what the mail provider uses. Most of them “should” be encrypting things by this point, but some of them aren’t. So it’s not nearly as secure as people think it is.

    Small businesses don’t have the resources to defeat the NSA. But that doesn’t mean you should sit back and do nothing because they’re not the only ones after your data. So are various hacker groups. Do you really want to be the low hanging fruit that they pluck off the internet?

    In some ways, that’s what AuditShark is really about. It’s about raising your security profile so you’re no longer the low hanging fruit for hackers.

    We’ve double checked it from two different computers, including one that hasn’t touched that file and it seems fine. Can you double check it from your end?

  8. Just tried it again and it came down quickly. I tried a few times before. Maybe the NSA was blocking it? ;(

  9. That’s a distinct possibility. 🙂