We’ve already written and talked about how we structure search campaigns . We use keyword planner, internal data, data from search engines, open refine, pencil and paper, and spreadsheets.
But how do we work with matches in the structure? How to do it most effectively in today’s machine learning world? We’ll share with you how we do it and why. Plus, we asked several top experts who also shared their modus operandi.
So how do we proceed?
In short: we make a mini-keyword list, we get to know the keywords in open refin, and then we create a mind map of what loan database campaigns and reports we need (instructions in the link above):
if the terms need an extra landing page – we have a few cases where we chose a URL at the keyword level, but those are really a few cases.
When we divide into separate reports:
if keywords need specific advertising – we divide even if we don’t have a specific landing page. Also for data collection, why create a separate LP.
if people searching for create a clear and concise message the terms are at different points in the purchasing process and the keywords have dramatically different conversion rates
some words may have
A low volume and still deserve specific advertising – especially if you are the only one who is able to solve a specific case. Let me give maldivian lads you an example. About a hundred years ago, we were running campaigns for travel insurance.
We found specific keywords like “travel insurance for pregnant women” or “travel insurance for cars” that only had 100 searches per month (I’m exaggerating a bit now, but it was really low).