The first step of accurately setting up a system for analyzing your sales funnel is making sure that the data you're collecting is unambiguous and un-skewed. Check your configuration for these possible fallacies:
There are instances when a visitor might have to navigate to a third party site before converting, for example, a payment gateway while checking out. Remember to account for these false drop-offs while working through the funnel analysis reports. You can check for the visitor count on the final step of the funnel, say, a payment confirmation/thank you page, to measure the real drop-off between the two pages—i.e the number of visitors who've abandoned the funnel right before converting.
Explicity defining funnel steps:
Visitors do not follow a set path to convert. So, to get data on each possible path, match the order of the steps in the funnel with the exact page-by-page navigation your visitors might undertake. Here are a couple of examples that highlight funnel misconfiguration:
Visitor A might take the home page> product page> payment page> convert route. And visitor B might take a shortcut: landing page> payment page> convert.
You've defined your funnel as: pricing page> click event > convert. But the visitors that land on the pricing page bypass the click event to convert.
If you do not account for the alternate navigation track, you might be missing out on a key behavior stat, and also might be tracking only a fraction of the conversions that occur in this funnel.
Tracking code installation:
Having the same code snippet installed on all the website pages is an obvious but an often overlooked step.
When you find a page in your conversion funnel with high drop offs, check if the page following it has the code installed on it. If the steps you're adding in your funnel have different domains, make sure that they have the same tracking code on all of the pages.