{"id":6499,"date":"2018-05-30T10:08:17","date_gmt":"2018-05-30T04:38:17","guid":{"rendered":"http:\/\/www.karooya.com\/blog\/?p=6499"},"modified":"2018-05-30T11:15:40","modified_gmt":"2018-05-30T05:45:40","slug":"ppcchat-conversion-rate-optimization-may-29-2018","status":"publish","type":"post","link":"https:\/\/www.karooya.com\/blog\/ppcchat-conversion-rate-optimization-may-29-2018\/","title":{"rendered":"PPCChat | Conversion Rate Optimization | May 29, 2018"},"content":{"rendered":"

This week\u2019s PPCChat session was hosted by Kyle Shurtz<\/a> and the\u00a0 discussion was about conversion rate optimization tactics and strategies, A\/B testing of landing pages, PPC experts sharing about their favourite tools for creating landing pages and analyzing them.<\/p>\n

<\/p>\n

Here is the screencap of the discussion that took place.<\/p>\n\"ppcchat\n

<\/h4>\n

Q1: What’s your preferred landing page tool to create new pages and why?<\/strong><\/h4>\n

 <\/p>\n

I\u2019ve only used @Unbounce<\/a>, so they win by default in my book. Too busy to investigate others at this point when we already have a working solution. –\u00a0@PPCKirk<\/a><\/span><\/p>\n

Whatever the client can get launched soonest with correct tracking. For a workaround, I’ve had good success with Unbounce too. –\u00a0@robert_brady<\/span><\/a><\/p>\n

Never used a LP tool, direct changes then monitored metric –\u00a0@gravymatt<\/span><\/a><\/p>\n

I don’t work directly with landing page creation rn, but our CRO teams love Unbounce for LP creation and testing –\u00a0@akaEmmaLouise<\/span><\/a><\/p>\n

We love #<\/s>hubspot<\/a> for creating landing pages. You can use smart content so the pages change depending on context, no need to make 10 PPC landing page variations. It also allows you to see what other traffic channels move the needle.\u00a0The most important thing is we need to be able to create landing pages with all the distractions removed, dedicated PPC landing pages are essential.\u00a0 –\u00a0@jimbanks<\/a><\/span><\/p>\n

Unbounce is nice and easy to use. On a non-cro side, I’m not sure about the effect on quality score sometimes when using dynamic text. It is so easy to use and allows you to make TONS of pages. And of course, it’s nice not to do math yourself.\u00a0 –\u00a0@markpgus<\/span><\/a><\/p>\n

I’m no LP creation expert by any means, but I’ve historically used Optimizely for LP variation creation and have started using Google Optimize since it’s a nice alternative (but sometimes doesn’t play nice with tools like Marketo). –\u00a0@nataliebarreda<\/span><\/a><\/p>\n

I’m into Unbounce and SquareSpace for all the way new builds. For testing tools, I’m into http:\/\/<\/span>Convert.com<\/span>\u00a0<\/span><\/span><\/a> or it’s highly pricey cousin, Optimizely. –\u00a0@JuliaVyse<\/a><\/span><\/p>\n

We’ve been using Google Optimize lately and actually really like it. It gives us more control and allows us to rely less on our client’s IT team (or getting the right access) to run our tests.\u00a0I have really enjoyed using <\/s>@Unbounce<\/a> – As someone who has built web sites, it really streamlines what you want\/need to do for landing pages, which are different animals than “regular” web site pages.\u00a0–\u00a0@NeptuneMoon<\/a><\/span><\/p>\n

 <\/p>\n

Q2: Let’s jump into the analysis. What’s your go to tool for insights on behavior with the page? Are you team <\/s>@hotjar<\/a>? Or maybe team <\/s>@CrazyEgg<\/a> ?<\/strong><\/h4>\n

 <\/p>\n

my weakness is still to default back to <\/s>@googleanalytics<\/a>\u00a0–\u00a0@JonKagan<\/span><\/a><\/p>\n

Do you manually have to pull your finding from Hotjar recording, do you look for a specific action from your users?\u00a0–\u00a0@gravymatt<\/span><\/a><\/p>\n

If we made the test with Google Optimize, we get stats from within that tool\/GA. For User behavior, we use <\/s>@hotjar<\/a>\u00a0–\u00a0@heyglenns<\/span><\/a><\/p>\n

I think GA is no doubt still king but being a Hotjar user I love the page recordings section. I could get lost there for hours watching how the user interacts with the page. –\u00a0@KyleShurtz.<\/b><\/span><\/a><\/p>\n

We are MASSIVE fans of <\/s>@luckyorange<\/a> but we take it to the next level with tags, custom events. We’ve made some major improvements analysing PPC traffic and would be lost without it.\u00a0This sort of stuff so we can analyses traffic type (GDN, search partner) position, checkout funnel –\u00a0@jimbanks<\/span><\/a><\/p>\n

We still mostly use <\/s>@googleanalytics<\/a> unless using Google Optimize – also have used different heat map tools, depending on what client’s have integrated with http:\/\/<\/span>convert.com<\/span>\u00a0<\/span><\/span><\/a> or <\/s>@Optimizely<\/a> –\u00a0@heidinksmith<\/a><\/span><\/p>\n

 <\/p>\n

 <\/p>\n

Q3: Let’s collaborate #<\/s>ppcchat<\/a> what is one of the best A\/B tests you have run that really moved the needle?<\/strong><\/h4>\n

 <\/p>\n

We had pricing added to the site, & instantly saw a drop in leads, so we duplicated the page, one with pricing, one without, and did a A\/B LP test to confirm the drop off was solely tracked to the pricing addition. Spoiler: Despite a drop in lead rate, CVR improved.\u00a0\u00a0–\u00a0@JonKagan<\/span><\/a><\/p>\n

super important to define the hypothesis in advance, get enough data to get statistical significance and record the results. Iterate from there. Doing a split test on mobile for using images of credit cards vs. just having the credit card number. Just the cc number won hands down (people were trying to tap their credit card type) –\u00a0@jimbanks<\/span><\/a><\/p>\n

My best A\/B test results have almost always come from simplifying the LP. Focus, focus, focus. –\u00a0@robert_brady<\/span><\/a><\/p>\n

Used Google Optimize for an SEO addiction treatment center client that increased the size and color of the phone number on a location page. 22% increase in conversion rate and 50% more click-to-call goal completions. For PPC, definitely removing most navigation and focusing the page’s CTA and funnel on one ask. For PPC, definitely removing most navigation and focusing the page’s CTA and funnel on one ask –\u00a0@heidinksmith<\/a><\/span><\/p>\n

I’m just starting a test of local tracking numbers on LPs and ad extensions.\u00a0Patience is required for any type of testing. You want to be annoyingly methodical about it too, which isn’t generally super fun. You want to know definitively what influenced the results. It should be 1 thing at a time.\u00a0One general note about testing and even digging through GA stuff – it only tells you what happened, not what else could happen. Important to keep this top of mind! –\u00a0@NeptuneMoon<\/a><\/span><\/p>\n

If you want an A\/B test to actually move the needle, you have to stop messing with small changes and completely change up the B in the test. Stop changing button color and change the design itself. –\u00a0@PPCKirk<\/a><\/span><\/p>\n

Risky answer, here we go: I haven’t had a lot of success with A\/B tests. My clients find the tools expensive, the tests time consuming and would rather do CRO with their organic campaigns rather than put more $$ on media. Few tests, with fewer results. –\u00a0@JuliaVyse<\/span><\/a><\/p>\n

 <\/p>\n

 <\/p>\n

Q4: Similar question but how about multivariate testing? Does anyone have some success stories using mutliple pages and what was the strategy?<\/strong><\/h4>\n

 <\/p>\n

I had great success in testing messages in out of market areas. The location was itself a test, with several different messages. I captured a whole new audience! –\u00a0@JuliaVyse<\/span><\/a><\/p>\n

In #<\/s>hubspot<\/a> we can have short forms for mobile, long forms for desktop. Nothing worse than complex forms to fill in on mobile device. It’s called Smart Content, it’s generated on the fly. Effectively every page can be different based on context, but it’s 1 template. –\u00a0@jimbanks<\/span><\/a><\/p>\n

We really haven’t done multi-variate much. I’d love to hear what others have ran and what their results were + reasoning behind doing that type of test. –\u00a0@heidinksmith<\/a><\/span><\/p>\n

 <\/p>\n

 <\/p>\n

Q5: When approaching a CRO project what does your strategy session look like? How far in advance are you planning new tests?<\/strong><\/h4>\n

 <\/p>\n

planning vs deployment. We plan test for deployment within 3 weeks. But, that deployment time often gets adjusted for multiple reasons. –\u00a0@JonKagan<\/span><\/a><\/p>\n

Boss says that form can have a higher conversion rate right.\u00a0–\u00a0@gravymatt<\/span><\/a><\/p>\n

It’s better to plan your tests when it’s not mission critical time of year, albeit test should be running all the time. Big tests could kill your best time slots. If you sell at Christmas that is a bad time to test something “risky”. –\u00a0@jimbanks<\/span><\/a><\/p>\n

Usually plan a few weeks out. Agreed with <\/s>@jimbanks<\/a> – we run them all the time, focusing on company performance goals and working to improve the conversion actions that will get them closer to achieving them. –\u00a0@heidinksmith<\/a><\/span><\/p>\n

after discovery\/initial analysis, I will plan where I want to start and it will evolve from there based on what that data shows me.\u00a0 –\u00a0@NeptuneMoon<\/a><\/span><\/p>\n

 <\/p>\n

 <\/p>\n

Q6: For the eComm experts. What tests do you run to improve sales and average order value?<\/strong><\/h4>\n

 <\/p>\n

We actually try onsite recommendations to show what other products compliment the product we deeplink to, to help pad the cart.\u00a0\u00a0–\u00a0@JonKagan<\/span><\/a><\/p>\n

 <\/p>\n

 <\/p>\n

Q7: How do you influence post conversion metrics utilizing PPC?<\/strong><\/h4>\n

 <\/p>\n

We are B2B-focused, so checking CRM for post-conversion activities tells us whether we converted the right kind of leads. –\u00a0@heyglenns<\/span><\/a><\/p>\n

Simple, remarket\/remarket\/remarket –\u00a0@JonKagan<\/span><\/a><\/p>\n

 <\/p>\n

PPCChat Participants:<\/strong><\/p>\n