As this was a new account, in a brand new niche, with a goal to validate an MVP and we had a limited ad budget and one landing page – we know we only have room in the budget for Exact match keywords.

But it wasn’t even only about the budget – We didn’t know keywords that’ll work, we had no negatives and we had no past data to start with Target CPA smart bidding strategy.

The service we provide is relevant only to one type of tax; We had to pay extra attention not to go too broad with our keywords, or to show our ads to any other type of tax refund.

In our research, we were focused mostly on the bottom of the funnel, long tail keywords, but at the same time, we were building out a list of short tail relevant keywords.

Since we were going to start with manual bidding, we decided to group our initial keywords in Single Keyword Ad Groups (SKAGs). Search volume per ad group wasn’t the factor at the moment, since we weren’t running smart bidding strategies.

With SKAGs, we had a high level of control over the search terms which are triggering our ads, and we had full alignment between the search query and the ad that shows.

It’s a great way to structure an account in starting phase, while we’re sorting out negatives,  learning bids, and collecting enough data to switch to TCPA bidding.

As our budget was limited at the start, it was crucial to reach our desired CPL as quickly as possible, since it was the first milestone towards scaling.

We ended month one with a 13% lower CPL than desired.

Everyone was happy with accomplishing, but we know it was just the start and that we need to do much more to make this account grow consistently.

When you start a new search campaign, the best thing for it you can do is focus on search terms and eliminate irrelevant search queries. It becomes even more important when you’re working with a new account which has newly set up conversion tracking and smart biddings strategies & AI aren’t helping you.

As we started with manual and max clicks bidding strategies, we had to make sure our bids are neither too high, nor too low. The goal was to gradually increase the aggressiveness of our bidding, as we’re strengthening our negative keywords.

We analyzed search terms daily, looking for long tail keywords which are relevant, and were adding them to our exact match campaigns. At the same time, we were always on the lookout for anything that was either irrelevant or too broad for us, at this stage.

We had no content to work with, nor the website and social media to make a funnel on.

We had to be very careful to target only bottom of the funnel keywords and maximize the number of leads from the given ad spend since the business was just starting out.

Quickly after the first campaigns started running, we implemented Hotjar – a user behavior analysis tool.

Watching closely, we analyzed heat and scroll maps, and screen recordings. The goal was to understand how good a job of explaining what we do, our landing page actually does.

We spotted two main problems: lack of social proof and lack of informative content on the page.

The landing page was fairly short, and since the service itself was new, we suspected this was going to be a case. We saw users browning back and forth throughout the page, reading things over and over again, trying to understand what we’re all about.

We asked our client to reach out to a few people who used our service and ask for testimonials. We wanted to have that on the landing page, and they proved to be the crucial element that will significantly increase the trustworthiness of our service.

We also wrote additional sections which mainly explained how the process with DutyRefunds goes, what the main benefits are, and for what items they can reclaim their import/VAT charges. 

As soon as the landing page got deployed and conversion tracking was done, we created an experiment in Google ads, splitting our search traffic 50%-50% and testing the original landing page vs. the one with additional content.

The first iteration of the original landing page was live 2 weeks after the account started.

Over the course of the next 4 months, we had 3 more iterations of the same page, improving it based on the findings and inputs we were getting from Google Analytics and HotJar. The most significant improvement between 2 iterations was when we were testing v3 vs v4.

We did more significant element organization, layout, and copy changes, which resulted in 25% higher CR and 27% lower CPL.

The first big structural change was at the beginning of our third month in the account when we switched to Phrase match keywords.

Our exact match keywords surpassed 90% Search Impression Share and we couldn’t grow much more, relying only on what’s currently live in the account.

Our SKAG structure run on maximize clicks and goal with it was to get to the point where we have 50+ conversions in the last 30 days, so we can start transitioning to smart bidding strategies.

But SKAGs and the way account was structured wasn’t optimal for machine learning. We had to change our approach if we want an account to grow. We grouped some of our top performing exact match keywords in themed groups, as phrase match keywords.

Unlike SKAGs, STAGs (Single Themed Ad Group) support automated bidding strategies much better, because the search volume is higher on the ad group level and the machine has more room to learn.

STAG structure allowed us to optimize the budget a lot easier, and a higher number of impressions per ad group created the perfect surroundings for machine learning.

We had a strong list of negative keywords and quite a nice number of conversions recorded in the pixel.

Still, the first two weeks of phrase keywords performance, on TCPA were weaker than our 2-month average, with manual bidding.

It started improving in the second half of month one, only to reach our previous performance.

We kept our head down, focused on the increased number of search terms we were now getting with broader match type, and negatived everything that was too broad or irrelevant.

Month two was when we felt the first more significant change – Our phrase campaign had 42% lower CPL at the end of month two.

But it didn’t take us long to see we’ll be plateauing on phrase keywords soon, at the pace we’re scaling.

The second big structural change happened at the beginning of our sixth month in the account – 3 months after we switched from Exact to Phrase match keywords.

At that time, we were spending 3x our starting budget and we quickly got to the point where we couldn’t grow much more, relying only on what we already had in the account.

We already had every relevant short and long tail keyword in our account. Since we needed to scale more, conversion volume was the priority over the CPL, next step for us was broad keywords.

As with everything, we always – start small, then scale.

We picked a few best-performing keywords, long tail keywords and created a broad match campaign. We know that, with our conversion tracking and the number of leads we’re getting monthly, TCPA will take this to the next level – But the question was, how long will it take?

Broad keywords can be dangerous, and we had a few occasions where it didn’t go as planned, but we knew we have super strong negative keywords lists and our pixel is fed with the correct data for 9 months – and we were right.

We started seeing incredible performance even before the 7-day learning phase was completed.

June vs May 2022 – We had 137% more leads with 19% lower CPL for 91% higher ad spend.

And it only got better in July:

July vs June 2022 – We had 63% more leads with 12% lower CPL for 43% higher ad spend.

The key to our scaling in June and July while maintaining and improving profitability was avoiding resetting the learning phase as we optimize.

We changed TCPA and budget once a week – changes no higher than 10%-15%

We started with one Unbounce landing page, which we developed and perfected over the months, and that landing page only collected leads. Until the website was done, leads were processed manually.

On the website, we had a multi-step form that automated the rest of the process.
The form on our landing page would now take the user to step 3 of the website form.

We did the tracking for the entire website, including all steps of the new form, but we still had all our campaigns learn and optimize towards the original form for months; We couldn’t just switch it out, we would lose all performance.

We created a new conversion as Google ads conversion, but no campaigns were optimizing towards it. It was there to collect the data, and now, we have 100 hits in the last 30 days on the new conversion.

The next step is to slowly transition the machine via experiments to start optimizing the performance towards our new end conversion.

The first major structural change was when we expanded our keywords from Exact to Phrase match.

Our new, phrase match campaign was even slightly more expensive in the first month. Smart bidding strategy needed time to learn and we needed time to negative new, irrelevant search terms.

Month two was when we felt the first more significant change – Our phrase campaign had 42% lower CPL at the end of month two.

The second major structure change was when we switched to Broad match keywords.

At this point, we were running TCPA bidding strategy for three full months. We expected June to be much better than May, but at the same time, we expect July to be even stronger when the machine stabilizes.

We ended up having 137% more leads with a 19% lower Cost per Lead.

We spent 91% more in June vs. May.

July 2022 was unbelievable!

A month was needed for the machine to learn, and with us assisting with negative keywords, we saw another record-breaking month!

In July we spent 43% compared to June, and we got 63% more leads with a 12% lower Cost per Lead.