Decision nightmare: enterprise e-commerce analytics

1 points by purplepatrick ↗ HN
Hi!

I am trying to figure out the best-suited analytics tool for a $500m/year e-commerce platform (web & mobile).

Everyone I ask tells me something else and I am hoping to filter some of that noise by posting here.

The most important things for us are:

(i) must play well with both Tealium and Google Tag Manager. We currently use the former but I want to run the later as well to see if we can eventually switch.

(ii) must work across web & mobile (comes with SDKs for both Android and iOS)

(iii) accurate attribution, funnels, etc. I am especially confused when it comes to aliasing, i.e. matching an identity to a user before they log in. Apparently, only Kissmetrics can do that, which I find odd.

(iv) I’d like to implement Segment and also potentially Autopilot, so we can add functionality through third-parties that is not native. This means we don’t need a tool with kitchen sink features and would connect notifications, email, heatmaps, A/B, etc.

(v) Other than that, we care about everything that would matter for any multi-platform e-commerce. Attribution is a big issue, of course, as are accurate funnels, drilldowns, etc.

I am looking at Heap, Amplitude, Mixpanel, Pendo, Kissmetrics. We’ve already ruled Firebase out. I also started to look at Oribi and Roivenue, but have not talked to them yet.

I’d be grateful for any first-hand experience with the above as demos/talks with some of them have not been that useful. They all say they’re great and can do everything.

Thanks!

2 comments

[ 4.9 ms ] story [ 17.5 ms ] thread
I believe that you should set your expectations clearly.

1. Do you want this tool to have advanced analytics features which give you to analyze the user behavior?

2. Who is going to use this tool, marketing or product people? Do you need a tool which also has marketing automation capabilities?

3. How technical your team is? Do you have data analysts & engineers who would be running SQL & Python on top of this data in order to do some advanced analysis or does this tool need to be plug&play for non-technical people?

Hey - I'm the the VP Product at Mixpanel. Happy to help if you like. Send me a message (neil at mixpanel) and I'll connect you directly with someone technical/non-sales.