9 Step Guide to Increasing PPC Sales by 39.4%

By , President at Hanapin Marketing

183 SHARES

Recently I had the good fortune to be on ta team that produced quite a turn around for an Ecommerce account. In 1 month sales grew by 39.4%. This was during a time when month-over-month sales trend downwards due to seasonality. Below is a brief summary of our to-do list from that month.

 

1. Check your funnel to see what percent of cart abandons you have.

  • If above 80% abandonment = review checkout process and test alternatives such as reducing steps, removing main navigation from the checkout process, and changing button colors and text.
  • If less than 79% abandonment = don’t worry about it, for now.

 

2. Filter your keywords to show all that are 20% below your CPA goal or greater and have converted 3 or more times. Bid them up by 15%.

 

3. Filter your keywords to show all that are 20% higher than your CPA goal and have converted 3 or more times. Bid them down by 15%. (This won’t grow sales, but will create wiggle room to push sales other places that may convert at goal).

 

4. Turn CPA optimizer off, and adjust bids based on the following formula (obviously do not change the bids of the keywords you changed in 2 and 3. The easiest way to do all this is in one bid excel doc and using filters):

1 / Keyword CR for last 7 days  = Clicks to get 1 close

(CPA goal x 1.10) / Clicks = Max CPC

 

Example:

1 / 3% =  33

($30 x 1.10) / 33 =  $1 max CPC

 

CPA optimizer does something more complicated, but similar. Although this will allow you to insure you push every keyword you want to based on how it is doing for the last 7 days. CPA optimizer often has more lag than that.

 

5. Add 500 keywords modeled after your top converters. To do this, take your top converters and put them into a handful of keywords tool like Google’s Keyword Tool and Word tracker. It will kick out a handful of relevant keywords that you can put back into the keyword tool and get even more options.

 

6. If you are only using exact and phrase match keywords, expand to modified broad match. If you are only in broad match keywords, start new campaigns with exact and phrase. Either way, find balance and make sure you’re testing what works best for you.

 

7. Conclude ad tests and leave your top ads running. Determine your top ads as explained in this post on ad testing. Start running only one ad in each ad group, this top-performing ad.

 

8. Add remarketing, and bid it high to start. CPA will start poorly but an overall lift will be felt and CPA will drop over time (hopefully).

 

9. Run a report in Analytics to see if any geo targets provide very high ROAS for you. If they do, break them in to their own campaign and push bids even higher.

 

That was it. It took about 20 hours worth of work, for a significant lift in leads at a CPA that was below goal.

Get more weekly links with our Fast Five newsletter! Five Fast Links in Your Email Every Friday.

Also send me a daily RSS digest

Remarketing Toolkit

Twitter Facebook LinkedIn Google+ Email Print More
  • http://www.deadtreemedia.com Jeff @ Dead Tree Media

    Hey Jeff,

    Nice tips… I especially like the line you’ve drawn on the funnel, has to be done somewhere.

    I think my only problem is on #2, where you up your top performers by an arbitrary 15%. While upping your bids on your top performers to get more volume is certainly a valid strategy, you should really base your increases on a mathematical line. A best of both world strategy is to still increase your bids, but don’t increase the CPC any higher than your CPA goal, using the idea that, even if your CR was 100%, you still wouldn’t bid any more than the CPA, otherwise, you’re over your target.

    A more mathematical increase is similar to your step 4, which I haven’t reversed engineered the math on yet, but I assume is a similar idea, is to just use a target CPC based on an algebraic solving of the ROI formula using PPC-based variables.

    In any case, solid stuff, and thanks for the post!

    • JeffreyAllen

      Fair point.

  • JeffreyAllen

    1. I wouldn’t worry about KW’s the only get a few clicks. But that’s because I work on large volume clients where those terms wouldn’t make an impact and focusing my energy other places would.

    2. That could give you a better QS and thus lower CPC, but it’s not always the case. Again I would look to volume. If the two different queries are generating significant traffic then it would be worth breaking them out.

  • http://twitter.com/EverydaySEMGuy The SEM Guy

    Why limit the use of ROAS to geo reports and not the optimization of the entire account? Does this e-commerce shop only sell 1 item at 1 price point?

    • JeffreyAllen

      Fair question. In the case of this client, they market hundreds of products in 88 countries. For the ROAS/Geo portion of the optimization, I was looking for clear outliers in terms of ROAS by country. There were many countries that were so far away from goal that it made sense to completely axe the entire country as a quick way to reduce CPA. There were also several countries with ROAS so high that it made sense to break them out into their own campaigns and bid up to drive more volume. ROAS by product was a factor, but a small one because the price points of the products were relatively similar. After this set of optmizations, though, a more granular approach was taken by product.