Conversion research for low Traffic websites

[gh_bennaceur (Hichem Bennaceur)]

I have a low-traffic educational website with 1800-2000 users per month, and I want to start implementing qualitative and quantitative data tracking. Google Analytics is implemented, and Hotjar is to get user recordings as well as heatmaps. I want to add surveys to collect users feedback to improve the user experience on the website based on them, as well as get more data on input to extract valuable insights in the conversion research step. The problem is that I couldn’t find

I want to ask if there is any complete guide to doing CRO on low-traffic websites, to understand :

• if it’s important to do conversion research or punish major website changes?
• What’s the sample size needed for launched surveys that it’s realistic in low-traffic site cases?
• Does the analysis of the low-coverage data from Hotjar as well as google analytics data is accurate to generate valuable insights?

Thank you!

[hesh_fekry]

@gh_bennaceur I would like to point you in the direction of the Speero ResearchXL playbook.

Do it like Speero: Identify website issues with the ResearchXL framework

To improve the effectiveness of your website, you need to identify the issues and friction points that are negatively affecting your website user.

This is quite complete and has many techniques that do not rely on high traffic.

@gh_bennaceur

• What’s the sample size needed for launched surveys that it’s realistic in low-traffic site cases?

I had this discussion with Ben Labay from Speero at one point. I believe 300 responses should give you consistently reliable answers. You could work with less, but this seems like a possibility with your user base. Depends on their level of engagement.

@gh_bennaceur

• Does the analysis of the low-coverage data from Hotjar as well as google analytics data is accurate to generate valuable insights?

This is definitely covered in that playbook.

@gh_bennaceur

• if it’s important to do conversion research or punish major website changes?

I believe it is always good to validate and test assumptions when possible. Sometimes big swings are needed. In what context do you ask this question? Perhaps the answer would be more helpful if we understood the type of changes?

[gh_bennaceur (Hichem Bennaceur)]

Thank you so much @hesh_fekry, I will deeply read the playbook and try to understand the best things I can do.

For the latest question, one client that I’m working with made big changes to their landing page in such a wrong way, this affects negatively the conversion, and now I don’t want to make changes without gathering data, this is why I want to collect data and try to do changes based on generated insights, even if the landing page now is bad in term of conversion.

[hesh_fekry]

@gh_bennaceur

atively the conversion, and now I don’t want to make changes without gathering data, this is why I want to collect data and try to do changes based on generated insights, even if the landing page now is bad in term of conversion.

Could be many things leading to that drop off indeed. Copy, placement, UX anything.

Let me know if you need some more specific here and we go from there.