I'm trying to write up a UX project plan for a client and I have included a web analytic review in there. My thought is that it would be sufficient to review their web analytics once a month to make sure the site is still working for the users.

I would then encourage them to poke around once a week and let me know if they see something that they have questions on and to make sure everything is running smoothly.

Any freelancers out there have a different approach or frequency? I'm currently an in-house UX/UI professional so setting a schedule for a client is new territory. I suspect it's the same but would like confirmation. Most everything I've come across focuses on in-house employees. Thanks for the help.

3 Answers 3


With one client I have actually a 'deal' that I continuisly monitor their website (both manually and with some link checking tools etc.) and once a month I actually do write a small 'review' about their web analytic results.

Both, client and me, know that of course they should a) look into their analytics themselves and b) draw their own conclusions from that. And of course they do look into their analytics themselves and b) draw their own conclusions from that.

However – they still do like the idea of somebody having 'outside view' with an 'inside perspective'. This may sound a bit confusing, but it just means that basically I don't know their specific content – but still I know what's important to them and what not. So to some degree I can tell them that e.g. in the time period A the content X was a strong entry page and that a teaser box on position Y was probably well placed bacause it 'clicked' better than on position Z etc.

  • 1
    I think I'm going to go this route. I don't know enough about analytics to give them an in depth review, but as you said, I do know enough to see when something I helped them with isn't performing as well as they might like. I'll just give them a short, high-level review to add to their own. It'd also be good for me to make sure that I sent them in the right direction while we were working together. Thanks.
    – Fletchling
    Commented Apr 25, 2015 at 18:00

I would highly recommend against setting a review for analytics. Obviously it depends on what they're doing, but if you have to delegate time to reviewing numbers for a client, I'd have to wonder what they're actually doing.

Here's why: unless you're a data scientist or your job is to analyze the data from the collected analytics, that's something the client should do. If they're willing to trust your analysis, that's different, then fine. But if the client wants analytics set up and then decides that the numbers mean one thing but you think it means something else, then you're in for troubling times. I've dealt with this a few times and have given up on teaching clients on analytics because even with a degree in data science, egos get in the way. This number means this, that means that, and nothing you can say will change what they think. Not all clients will react that way, but depending on the type of work you do, many will.

Also, I don't know if you really want to spend the time reviewing analytics unless this is a long-term project and you want to use that data to determine future work. If that's the case, fine, but just be very careful with how you word the contract. That said, I'd suggest reviewing once a week for a total of 2 hours. In my experience I've spent a lot of time understanding the analytics data, but more importantly I'd bounce in and out of the reviewing when the need arose. So it was never an hour here and an hour there, but five minutes at a time, four times a day, etc.


Analytics isn't necessarily about whether you are hitting your targets and meeting the requirements. The metrics are just another source of information when you are trying to understand the changes in the behaviour of the users in response to either trends, changes to the website or anything else that might have an influence on what they think.

You should monitor rather than review analytics, so part of the plan should include having someone or a team that looks over the information and data for patterns and trends.

  • 1
    You're right and I would make the point more strongly about decoupling targets and measures. Something is either a target or a measure, it can't be both. Commented Apr 21, 2015 at 16:25

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.