Most people will guess it’s for lack of goals, not having sufficient time/expertise, or having a crappy product.
The creatives will say it’s for poor content– weak landing pages and lame creatives.
The numbers geeks will say it’s conversion rate, quality score, insufficient budget, or some metric that’s off.
These are all good things to optimize, but there’s something even more fundamental than that.
I will let you in on the “secret” I use to get campaigns unstuck.
It works nearly all the time, and when it does, people often pretend it wasn’t so embarrassingly simple.
You ready? Here it is.
Lack of iteration.
It’s that simple– not enough shots on goal.
That’s not the same as Herculean effort, slaving nonstop in building massive campaigns.
In fact, it’s the opposite– many small steps daily, like working out a little daily instead of a monster workout each month.
Social advertising allows us to take hundreds of shots on goal for a few dollars each.
There is so little risk in trying a different target or creative out– worst case, you lose a few dollars.
Underlying all this is fear, which most people, no matter the age, fail to recognize.
When most people go about campaigns, they get stuck in a particular step.
Usually, that reason is not a stopper– it’s one you can easily work around, like a speed bump in the road.
I’ve run online campaigns for 20 years, and lack of iteration is still the number one challenge to get past.
HOW TO DIAGNOSE THIS PROBLEM
We have a simple tool to cut right through the excuses and fear to determine if iteration is actually happening.
It’s called MAA for Metrics > Analysis > Action.
You don’t need a fancy template or spreadsheet– just 3 minutes to go through this logical sequence.
Metrics: Here are a couple of stats on our campaign performance that have gone up or down (usually CPC and conversion rate).
Analysis: Here’s why I think they’ve gone up or down (look for “because”).
Action: Therefore, this is what we can test next (look for the next action).
You’re not looking for how deep the analyst goes into the explanation or how many pages the report is.
You’re looking to see how many iteration cycles they have.
Cycle #1: Our overall CPA is too high, but our remarketing is going well. We need to increase remarketing pools by mid-funnel traffic.
Cycle #2: Landing Page A is doing better than the others, probably because it is relevant to Audience B. Therefore, let’s drive more variants of Audience B to Landing Page A.
Cycle #3: Our Quality Scores are low in broad match terms. Why? Because we don’t have enough negatives or an exact match. Therefore, add negatives and report again in a couple of days.
And so forth.
How many iterations have you done on your business or client in the last week?
Odds are you have just one, corresponding to the frequency you have to report results.
The “final exam” procrastination mentality causes people to wait until the last second to make a tweak.
But there’s no iteration if no time elapses between taking action, observing, and subsequent follow-up.
And then you don’t touch things for another week when it’s time to make that report or attend that meeting.
HOW TO FIX THIS PROBLEM
A smart “test and learn” (trial and error) optimizer runs a few “experiments” at the same time.
While it’s easy to say that, actually, to get change, you have to reason with your campaign folks so they aren’t paralyzed by fear or habit.
Here are some examples:
“I’m too busy.”
This is the most common excuse by far.
You help them break through this mindset by teaching them how to do an iteration cycle in 5 minutes (see below).
“I’m waiting on person X to do something.”
We can always take some other action to get around this or something else to test.
Perfect is the enemy of good. Done is better than perfect.
“We need more creatives.”
This happens in larger companies where multiple agencies are involved.
Use the same image, but tweak multiple versions of the headline and body for each target.
“Not enough data/budget to be statistically significant.”
Look at early indicators instead of waiting to get X conversions: CPC instead of conversion rate or CTR instead of CPC.
If they say they need to wait weeks to let the test run, note that it’s how much data you gather, not the time elapsed, that matters.
“I already did X, and it didn’t work.”
They perhaps ran one ad to one target but didn’t go any further in testing.
Have them break down quantitatively what didn’t work, isolating whether it’s the creative or the targeting since an ad is the combination of these two.
The net result is that you want your folks to complete an iteration cycle in 5 minutes and move on to the next project.
If you work out at the gym, you know some people spend 2 hours there but get in only 5 or 6 sets.
They spend so much time between cycles resting, getting water, chit-chatting with others, admiring their muscles in the mirror, and so forth.
But folks who are efficient can get in and out in 30 minutes by working another muscle set while one is resting.
Maybe they alternate bench with lat pulldowns or quads with abs.
I learned this from Paul Sokol, by the way, who is the top mind in automation (not just campaign building), in my opinion.
It’s how many sets you do; not how long you’re in the gym, or in Paul’s own words:
“Time does NOT equal value and most people miss this. Especially when building and iterating their marketing! Rome wasn’t built in a day and neither are ROI producing campaigns. Once version 1 is out the door, that’s the hardest part. From there, if you focus on measuring the right metrics, you can progressively make tiny, informed non-time-consuming tweaks and build toward something amazing!”
THE ITERATION CYCLE
Let’s break down MAA a bit further into the mechanics to show how a set can take 5 minutes instead of 5 hours.
Sort by total spending in the last week by spend descending, clicks descending, and conversions descending.
Look at the top 5 each time or the top 10 if you have more time– we call this Top N.
Why are certain keywords/targets or creatives/pages performing well or not?
Look at metrics in pairs since there is always a counterbalance– CPC vs. conversion rate, CTR vs. average position, audience size vs. performance, etc.
What’s the one thing to tweak which is most likely to address the WHY of analysis?
In increasing order of impact but also effort are these tweaks: bidding, targeting, ad creative, landing page, and strategy.
When people are getting used to iterating quickly, they might take 30 minutes at first, but that’s okay.
The more “corporate” mindsets are conditioned to believe that everything happens in chunks of hours.
Might be a one-hour meeting or 8 hours to build a campaign.
Break them of that habit.
STEADY OPERATIONS VIA CID
Your company is only as good as the various folks you have iterating (optimizing) together.
Besides the basics of project management (tracking tasks according to project plans, following up on meeting action items, etc.), use CID.
CID is Communicate > Iterate > Delegate.
Communication is the foundation of CID that people are reliably responding.
When a thread drops, you have a “weakest link” problem, causing a sequence to break and the project to fail.
As basic as a reliable response is, this is the most fundamental and common issue in organizations.
People feel overloaded, ignore messages, and don’t have general organizational skills.
We use boomerang, a Gmail plugin, to ensure we don’t drop things or lose cycles chasing people.
The RACI project management model is also a powerful tool to ensure not everyone is copied on everything.
Role clarity ensures the right people are in the right conversations– so we don’t have either overflow or blockage.
We use Basecamp for project management and Skype for communication.
Every team member must update the internal team roster with their contact info and add other team members at the start of any project.
Every iteration step requires communication– that way, the chain isn’t broken.
Then it’s easy for anyone to tell where projects are– there is no hiding.
But more than a monitoring tool to catch slackers, iteration/communication eliminates wasteful status meetings.
Use the meetings to make decisions, not read updates to one another.
When communication and iteration are solid, then delegation can take place.
It’s the logical “next step” from the Action of MAA, where that action requires someone else to assist.
So when CID breaks, you generate company politics.
We define politics as the withholding of communication, intentional or not.
People get cut out of the loop, parasites can hide, good people don’t get rewarded, team members aren’t sure who’s doing what, and confusion ensues.
Take the advice of Geoff Woods, who coaches people on finding their mentors and entrepreneurship:
At the end of the day, you will have questions implementing these strategies. The question is, what are you going to do about it? Will you be like most people who just give up, or will you do what the most successful people do? Reach out to those who have done it before and ask for help!You are not alone in this journey. Surround yourself with like-minded individuals and ask for help. After all, in today’s market competition is dead. Collaboration is the new way of doing business.
When CID and MAA are working in sync, you’re resolving dependencies directly– not needing meetings to discuss each time.
You’ve got the optimization engine of MAA in tune with the project organization cycle of CID to drive your company.
About the Author