One of my favorite clients is an older man who has a highly successful ecommerce website. His PPC was done in-house until we took it over, and he’s been absolutely delighted by the increase in sales and decrease in cost per conversion. He calls me every week to go over his reports. I love the fact that he wants to understand his reports and what it all means and I enjoy talking to him. He’s coming to the realization that nothing stays the same and one week of data can’t predict the rest of the month, that his conversion rate and everything else fluctuates, sometimes dramatically, and that a term that converts one week, might not convert the next. I love that he gets this, because that can be one of the hardest things to get my clients to understand.
Sometimes there’s a reason why a search term converts well and then stops converting as well. But a lot of times, there’s no reason that we can figure out and if you just wait a bit, it’ll turn around again. Sometimes you’ve just got to “Let It Ride.”
I have a few clients that freak out after one bad day – they want me to increase or decrease bids or pause an ad or stop advertising a search term all together. One of the biggest lessons I’ve learned with PPC is to look at the long view and not react to every little dip and swell. I tell my clients that PPC Advertising is like the stock market – you can’t react to every little change, but you’ve got to let it ride and look at the big picture.
And just a few days ago, a client informed me that I should be writing new ads for him every two days, as well as changing his bids. And I told him that’s one of the worst things you can do. Now don’t get me wrong, writing new ads is great – I’m constantly trying to beat my best performing ad in each AdGroup, but if I’m changing ads and bids every few days, how am I supposed to see what’s working and what’s not? How do I make decisions based on sound data and not guestimates? Sometimes you have to let it ride, collect the data and THEN make the decisions about what to do. If something’s working and working well, don’t mess with it, just for the sake of saying something’s been done.
Even though it is really tempting to look at one day of data and say that this keyword or AdGroup or ad will never work, resist that urge. Look at how that keyword, AdGroup or ad performs over time. And don’t get caught up in trying to figure out why a search term worked one day or one week or one month and not the next. You could drive yourself crazy, because sometimes there is no reason that’s within your control. Proceed with caution and keep a watchful eye but don’t give up hope.