Online advertising has become more and more technical. On the one hand, the number of platforms is increasing with each of them bringing new challenges. On the other hand, thanks to a large amount of data, online advertising is now the fastest way to scale up a business online. In addition, the insights gained from one platform can be used for other platforms, creating synergy in your advertising strategy. Unlike SEO which often takes longer to implement and which is more time consuming because it relies on content writing and content promoting, online advertising allows you to scale up your business quickly and at the same time control your return on investment.
In this ultimate guide, we are going to cover the A to Z of online advertising, from tracking your results to scaling up your advertising, optimize every last penny and cover each of the platforms and marketplaces.
Disclaimer: We might receive a commission from affiliate links displayed in this article. However, our opinion is never biased because of that.
What is Online Advertising?
Online Advertising is part of Online Marketing, where you need to pay for the traffic that you get. There are 7 channels for Online Marketing:
Unpaid Channels
- Direct channel: This is people who know your website, for instance, amazon.com, and come directly to the website by inputting your website address in their web browser. Having said that, it can be that they know your website because they learned about it through another channel or that they found out about you offline (friends, printed ads, tv…)
- Search Organic: also known as SEO for Search Engine Optimization: people come to your website by searching in search engines like Google, Bing, Yahoo…Most of the time, people are looking for something to solve their problems or to educate themselves. They can be comparing different products or technology or looking for the best way to solve their problems.
SEO Marketing is an art on its own so we will not cover it in this ultimate guide. In a nutshell, it requires keyword research, writing some engaging content, and promoting this content to get some backlinks directing to your website. If this is something interesting for you, you can check SEMrush for instance as it is affordable, easy to handle and allows you to get your competitors’ data and automate your backlinks promotion strategy:
Take a Trial and Get More Out of Semrush
- Email: that goes without saying but for email, you need to have a list of emails first. You can either collect email from people who visited your website and engage with you (for instance on social media, newsletter, account creation, white paper, exit popup). You could also consider getting emails from 3rd party. After collecting emails from customers and prospects, you can re-engage with them and keep them updated on any news of your website…By all means, emailing is the marketing channel with the highest return on investment because the cost is very low and it scales easily.
- Referral: this can be any website, blog, newspaper…with a link directing to your website.
Paid Channels
- Paid Search Ads: Typically, you get charged for each click, mainly for search/shopping campaigns. Those ads appear usually at the top of the search result when someone enters a search query in a search engine.
- Display Ads: Typically, you get charged for each impression. Those ads are displayed when users surf a website, usually between the text or on the side of the content.
- Social: those ads are shown on social media like Facebook & Instagram, or TikTok.
Now even though each channel should not be considered independently, in the rest of this ultimate guide, we are going to focus only on paid advertising because unpaid advertising requires different skill sets and would be part of another guide.
Also, understanding each of the channels individually is great but being able to combine them to increase the number of touchpoints with the customer and nudge him to become a returning customer is a completely different strategy and that’s what we are going to explore.
Growth scaling concept
There are 3 important concepts to understand growth scaling: reach, customer intent, and optimization. Those 3 concepts allow a continuous growth process for our customers.
What is Customer Intent and why is it important?
When someone is looking for something, no matter if it is in Google, Amazon, or any search platform, he or she has something specific in mind. We call this customer intent. The search query used can tell us a lot about the customer’s intent.
For instance, if someone is looking for Shopify this means that this person is already familiar with the brand. In other words, he or she is more likely to buy because he already knows about the value-added, the benefits, and the mission of the company..
However, if he or she looks for Shopify cancelation, it is not likely at all that he would buy.
If he or she is looking for Shopify review, then it is a signal that he is considering buying. So even though all those search terms are similar, they have completely different intents.
Having said that, the same keyword can be a brand keyword for a customer and will be a competitor keyword for another customer. For instance Nike and Adidas.
Because the customer intent defines how likely it is that someone will buy a product, it is very important to structure campaigns based on different customer intents so that each campaign can, later on, be optimized properly.
Here is a list of different customer intent on a scale from 0 to 5, 5 being the highest:
(conversion likelihood:4) Brand: Customer searches for the brand directly. Ex. Nike
(conversion likelihood:5) Brand + Specific Attribute: Customers search for a specific product of a brand. Ex. Blue Nike
(conversion likelihood:2) Generic: Customers look for a product. Ex. Shoes
(conversion likelihood:3) Generic + Attribute: Customers look for something more specific. Ex. blue shoes.
(conversion likelihood:1) Competition: Customers are looking for another brand. Ex. Adidas.
(conversion likelihood:0) Competition: Customers are looking for a specific product from another brand. Ex. red Adidas.
Also, the likelihood decreases if the product is not sold. For instance, if we sell only blue and green T-sheets and the customer is looking for red tshirts, which we don’t sell, it is less likely that the customer will buy other colors (it doesn’t mean that it is not likely at all).
Therefore, campaigns need to be separated by customer intent. A combination of generic terms and product attributes is a good start. For instance green shoes, blue shoes, yellow shoes… This allows us also to systematically scale up keywords to reach more customers with the same performances.
How to reach more users without paying more per clicks?
The more people you reach, the more traffic you can generate and the more sales you can get for the same performance. To increase your reach, you need to show your ads to more people, either by adding more keywords to your account and/or diversifying the type of campaign you use and/or by duplicating your campaigns to another platform. You can also do so by increasing the targeting, for instance, if your ads run in the US, you can expand to Canada, UK, Australia to show your ads to more people. Of course, it always depends on the nature of the business. If you sell a service, it is easier to sell internationally. If you sell a product, then it is more difficult to sell globally.
Which match type should you use?
Each match type has pros and cons. Broad keywords have the biggest range but the problem with broad keywords is that it is hard to control the customer intent.
Macth Type Example keyword Match to:
Match to search query like: Broadt shirt misspellings; synonyms; related searches; close variation sthsirt; polo; golf shirt; T-shirt Broad Match Modifier+tshirt +blue search terms that include all words preceded by “+”; close variant tshirt with blue stripes; T-shirt with blue stripes Phrase“blue tshirt”search terms that include every words in the quote; close variant best blue tshirt for men; best blue T-shirt for men Exact[blue tshirt]Search terms that exactly match the keyword; close variant blue tshirt; blue T-shirt
Therefore, it is important to use keywords in a wise way. For instance, having a generic campaign with broad keywords to get traffic. Then analyze the search query report and identify different worlds that are relevant (category, property…) and add systematically a combination of as broad match modifier to the right campaign. For instance:
Blue King size bedsheet:
- +blue +king +sheet
- +blue +king +linen
- +navy +king +sheet
- +navy +king +linen
Which Campaign Type should you use and how to add them in your strategy?
Each campaign is a way to get additional traffic and reach and therefore incrementally grow our customers’ revenue.
For instance, on Google, many advertisers use Search but should also explore DSA, Display, Shopping, Video.
On Amazon, many advertisers have sponsored products but should also consider Sponsor Brand & Display
Google campaigns overview
Google Display (and Video) campaigns
Display campaigns are great to reach people that are not searching for a product (yet or anymore). It can be used for creating brand awareness for instance at the top of the customer funnel. It can also be used to reengage visitors or purchasers after some time.
For prospecting, targeting is very important, that’s why we always advise using audience insights from Search campaigns before creating Display campaigns.
Audience based targeting:
In Google Analytics, we can see which are the top audiences (in-market audience, affinity audience…) and we can add those to the search campaigns. If they perform well we should use those audiences as targeting for display prospecting campaigns.
Keyword-based targeting:
In Google, it is also possible to target customers based on the keyword of the website. If the website contains top converting keywords, then it is likely that people are interested in this topic and therefore it is a good targeting option.
Search intent targeting:
It is also possible to target people who research some specific keywords. This means that we can target people who look for top keywords but didn’t click on our Search Ads. This makes especially sense when competitors’ bids are very high and we can not compete anymore but we still want to reach those customers.
In every case for prospecting, it makes sense to exclude past buyers from the targeting to avoid wasting money.
For retargeting, remarketing, we can either target people who visited the website and left without purchasing (for instance cart abandoners) or people who bought in the past. If so then we should take into consideration the buying frequency. For instance, if people buy every month, we can show an ad one month after their previous purchase.
Google Shopping Campaign
Shopping is a great way to get more traffic for eCommerce websites. It is slightly different from other campaign types because Google allows only bidding on product groups and not keywords. Products are grouped into product groups and we can set one bid per product group. For each product, Google will automatically associate search terms.
Therefore it is important to segment the shopping campaign by category. If you have a large number of SKUs (>500), it doesn’t make sense to create either one ad group or one campaign for each category
Because we can not bid on keywords and because the performance for different customer intent are different, it makes sense to duplicate the campaigns as follow:.
You should have 2 campaigns:
- Brand (low priority – high bid)
- None Brand (high priority – low bids) with negative keywords for brand terms
In this way, if someone search for something generic, is less likely to convert and therefore the click price will be lower. If he is looking for a specific brand, he is more likely to convert, and then it makes sense to spend more because the return will be higher.
Amazon campaigns
Sponsored Products
Sponsored Product is the main type of campaign on Amazon Advertising. With Sponsored Product campaigns, you can either target some keywords for instance shoes, and appear at the top of the search result on Amazon or you can target competitors’ products and appear on their product page.
One thing to keep in mind is that on Amazon, Advertising sales and Organic sales are not independent. Indeed the more you sell through Advertising, the better will be your ranking and the more you will sell.
Sponsored Brand
Sponsored Brand increases your brand awareness by showing a banner at the top of the search result. Usually, sponsored brand return on ad spent is lower than sponsored products but it can still bring you incremental sales.
Sponsored Display
Finally, Sponsored Display helps you to leverage audience information of Amazon. You can target users on and off Amazon based on their audience, for instance, if they visited your product page or if they are amazon prime users. The level of targeting is not as precise as with Amazon DSP but this is still a good alternative if you want to start small.
Platform
Each additional platform is an opportunity to reach new customers and also a way to mitigate the risk of relying only on one source of income.
Here is the platform we came across and can offer to our customer:
Search
- Google Ads
- Microsoft Ads (campaigns can be imported automatically from Google, also windows users use quite a lot Microsoft ads search engine, especially in the US)
Social
- Facebook & Instagram
Social
Optimization cycle: Define, Measure, Analyze, Improve, Control
Thanks to the large amount of data that is now available, Online Advertising has become a data-driven practice. The main advantage of this data-driven process is that you don’t need to create a persona or a profile of your ideal customer before getting started. You can gradually scale up your advertisement as you get more insights into your customer, for instance by knowing what they are looking for, how old they are, or where they are based.
We are going to see how to implement the full optimization cycle:
Define your KPIs for advertisment
First of all, you need to define the KPIs which are relevant for your business. Here you should focus on what is important for your business model and not get into the trap of using out-of-the-box advertisement metrics like ACOS/Return on Ad spent or Cost per Conversion.
If you are an eCommerce business, you probably have different profit margins for the products you sell. Therefore 10 % ACoS might be good if you have a 20% profit margin but it is bad if you have 5% profit margin.
If your business model is affiliate or you get a commission for instance if you are a price comparison platform or a platform connecting users like Airbnb, Blablacar…Then your commission is probably a rate of the product or service you sell on your platform.
Finally, if you sell a service or a subscription and you have no way to know what would be your customer lifetime value upfront, then cost per conversion would make more sense for you.
Something to keep also in mind :
- customers don’t only buy once. If customers are frequently buying from you, you should know exactly what your customer lifetime value is. You would prefer a customer buying 5 times from you each time for $50 than a customer only buying once for $100. This being said, different audiences have different LTV.
- That cross-selling might increase your profitability. Because many of your costs are fixed costs, you might be more profitable if you use cross-selling, meaning that you could spend more to get incremental profitable conversions.
Measure improvemet of your ads and sales
At Growth Scalers, we have seen many customers who try to scale up their accounts before making sure that the tracking works properly. Here are the most common signs that your tracking is not working:
- You don’t get any conversion even though you have a lot of clicks (>>100)
- Every day the number of conversions is an even number
- Your bounce rate is 0
- You have more sales in analytics than in your CRM
In what follows, we outline how to accurately track conversions on your website.
How to track conversion with Google Analytics?
First, sign up in Google Analytics, and follow the wizard to create your account and a property. Generally speaking you need one property per website.
Then, to get the tracking code of Google Analytics, go to Admin > Property > Tracking Info > Tracking Code.
Copy this code and you can now paste it on your WordPress or Shopify website. For WordPress you can use plugins like MonsterInsights. If you are using Divi themes you can go to Divi > Integration and paste it to the code of the of your website. Finally, save changes.
To check if the tracking is working, go back to Google Analytics > Admin > Property > Tracking Code and click on “send test traffic”. It will open your website homepage and if you see at least one active user, it means that the tracking is working.
How to activate Ecommerce transactions?
Tracking conversion is great but you also want to know which product performed best and how much revenue your advertising is generating.
Therefore, you should activate eCommerce transactions.
Go to Admin > View > Ecommerce Settings and then enable eCommerce.
You will see on the left a new eCommerce menu:
You can now analyze check-out behavior, avg. order value and product performances.
How to get your Google Ads data in Google Analytics?
You might want to see advertising data in Google Analytics to know your cost of advertising. Go to Admin > Property > Google Ads Linking and click on “new link group” and select your google ads account.
How to import conversion from Google Analytics to Google Ads?
Having your KPIs in Google Analytics is nice but having them in Google Ads and use them for optimization is even better. Now that everything is setup in Google Analytics, let’s import the data in Google Ads. In Google Ads, go to Tools & Settings > Measurement > Conversions then click on the blue + to create new conversion action. Finally, click on Import and select Google analytics
Analyze your ads performance
After making sure that the tracking is working correctly, we can start analyzing data.
Understand your customer intent
The search query reports give us a lot of information about customer intent. We know what people are actually looking for and which keywords triggered which search query.
It is important to make sure that each campaign or adgroup is targeting the right queries as intended. It can be for instance that broad keywords in generic campaigns trigger brand search terms. If so we need to add brand search terms as negative phrase to make sure that each campaign is targeting the intended search queries.
We can also aggregate the performance of each word and analyze the performances to see if some search terms are not related to our business. For instance, if we sell yellow T-shirts and we don’t sell black T-shirts, we can aggregate the performance for the word black, and most probably it will have a low return on investment.
Check if your ads Ad Running?
Quite often, ads stop running, meaning that they have no impression. This is a missed opportunity as those ads could easily bring you additional sales. You can check this easily by filtering every ad or adgroup with 0 impression in the last weeks.
If you find some ads that are not running, check if:
- your bid is too low
- ads are not approved (maybe because they are not complying with some policies)
- targeting is too small (keywords too specific or audience targeted too small)
Title and ads performance
There is only a limited number of people looking for what you have to offer. Therefore, the better is your clickthrough rate (CTR), the more click you will get and the more sales you will generate.
The CTR =clicks/impressions
If you want to improve your CTR, check out how to write good ads?
Pay per click performance
If we consider PPC optimization (keywords but also any kind of segment like age range, parental status, audience, devices…)
Then the important output depends on the business model:
If the business model is about generating leads then the output KPI which needs to be considered is Cost Per Conversions (CPA). The input metric that needs to be considered is the conversion rate (number of conversions/number of clicks).
Let’s assume we have only mobile and desktop, if the conversion rate for mobile is 2% and the conversion rate for desktop is 4%, then without optimization, the cost per conversion for mobile will be double of it for desktop. For instance $100 per conversion for mobile and $50 for desktop. Now if we can reduce the click price for mobile and/or increase the click price for desktop so that we have the same cost per conversion, then mobile and desktop will be optimized. For instance, we can apply -30% on mobile and +30% on desktop.
If the business model is about generating sales then the output KPI will be ROAS (also called ROI return on ad spend, or ACOS=1/ROAS). The ROAS is the conversion value * conversion rate / Cost. So the input will be both conversion value=product price and conversion rate (assuming that every product has the same profit margin). In the same way, to be optimized, each segment needs to have the same ROAS even though the conversion rate and the conversion value are different.
Improve your ads to maximize sales
How to write good ads?
A good ad should engage the reader and create the desire to learn more about what you have to offer. To write good ads, here is our secret sauce:
- Relevance: Your ads need to match with what customers are looking for. Let’s say, for instance, that your customers are looking for a blue T-shirt and your ad is showing a red T-shirt. That would be a no-go: probably people would not click on your ad, resulting in a low Clickthrough rate. You would leave more money on the table because you would get fewer visits than what you could and your sales would be also lower. In addition, low CTR will have a negative impact on your quality score and therefore you will pay more per click.
- Talk to your customer: Imagine that you want to sell directly to a customer who is contacting you or entering your store. How would you address his needs? Probably you would listen to his needs (analyze search query report), show him the most suitable product, and tell him why this product is what he is looking for. If you need to write ads, direct communication is extremely more powerful than generic ads.
- Outline benefits over features: You know the saying features tell but benefits sell. To illustrate this point, remember the slogan of Apple when they launched the iPod. Steve Jobs could have mentioned all the features in the world, the 5 GB hard drive, the Firewire connectivity, and the synchronization to iTunes but features are not solving any of the customer’s problems. Instead, he just stated one benefit, that you could carry 1000 songs in your pocket.
When possible, always showcase the benefits of your product or service instead of features.
How to improve conversion rate?
This is a one-million-dollar question. Because in the end, if you could improve by 10% the conversion rate, you could save 10% on advertisement every month. Compounded for a couple of years, it becomes very interesting to optimize conversion rate. Here are some tips that have proved to work.
- Make sure your site loads quickly: if your website is not loading quickly enough and your visitor are just bouncing, it doesn’t make sense to move forward and try to increase advertising spends. We have seen so many websites that didn’t convert, just because people were leaving before the site was even loaded.
- Target the right audience: First thing first, not all audiences are equal, if your ideal customer is between 25 and 45 years old and most of your visitors are over 65, chances are that your conversion rate is low. You can either limit the targeting to the right target audience or decrease on Google/ Microsoft ads the bid adjustment by demography, location, audience…
- Use social proof: for instance testimonial, review, case study, number of clients or well-known brands. In some way, you need to prove that your solution worked for others and that it will work again with your prospect. Influencers is also an alternative way to create social proof.
- Make purchasing as quickly and as easy as possible: with each additional step in your purchasing process, some users will drop the funnel. Don’t ask for users to register before cashing out, make sure they can pay without having to search for their credit card, for instance with PayPal, Amazon Pay, or Google Pay. If you want them to register, you can let them create an account after they buy.
- Leverage data analytics to see where people are getting stuck in your funnel and where they are dropping. You can use for instance a solution like Mouseflow to understand how visitors are interacting with your website, where they are clicking, and where you make the most money. If you make most of your money at the bottom of the page, just rearrange your website to have the most profitable items at the top and you can thank us later.
Control your result with this simple check list
Follow the following checklist to make sure that you are not missing out anything:
- Check weekly the performance of the last week and the last 30 days. Is there some permanent change in performance, is it due to customer behavior (conversion rate, conversion value) or to competitors (increase in click price, loss of impression share)
- Check weekly the search query report to see what people are looking for. Did you identify new customer intent? If so, you need to create new ad. Is there some unrelated intent to your business, then use negative keywords.
- Check bi-weekly/monthly the performances of each demographic (gender, age, parental status, income range), audience, location. If needed, adjust bid adjustments.
- Check out keyword recommendations from the platform but also from other channels, including SEO, and see if you can add new keywords.
- Run A/B test to optimize ads for at least 1 month.
Thank you for reading this ultimate guide until the end. We spent time keeping it as concise as possible and bring you as many insights as we could. Don’t hesitate to share it with anybody who could benefit from it. Please reach out if you have any further questions. We look forward to talking with you.