Sign up for weekly recaps of the ever-changing search marketing landscape.
How Best Practices Differ In Small & Large Accounts
There are a lot of great articles written about paid search and best practices. However, almost all these articles ignore one important aspect of paid search: account size.
For instance, it is a best practice to bid at the keyword level, and every large account should be doing that on most of their keywords.
However, that is an unrealistic expectation for a small account that does not generate enough keyword level data to set keyword bids. Small accounts must bid at the ad group level based upon combined stats of all the keywords within an ad group.
In this article, we’re going to examine some of the differences and best practices of creating and measuring accounts based upon size.
What Is Your Account Size?
It’s almost impossible to layout every account size. The easiest proxy for account size is spend.
Here are the overall ranges I usually see when looking at grouping accounts together. Within each range, there is a sub range as most accounts within a sub-range can be managed similarly based upon spend and industry type.
These are based upon monthly spend:
- Small Accounts:
- Under $500
- $500,000-$1 million
- Huge: $1 million – $3 million
- Massive: More than $3 million
There are always exceptions to a list like this. For instance, I work with one company who spends $40,000/month on 5 keywords. From a spend perspective, they have a healthy spend. From a management perspective, they can treat some of their action items like a small account because they only have 5 keywords. This is an exception, and most companies fall more in line with these spends.
Once you understand what it takes to manage an account in the $26,000 spend range in the lead generation industry, you can usually use those same techniques to manage an account that spends $90,000/month.
However, those techniques must occur more frequently if the account is spending $300,000/month. By grouping accounts by spend and industry, you can determine your account’s project management style.
Once you’ve determined the type of spend you are working with, the first step is to consider how to create the account and this is where the small versus large account differences become obvious.
Account Creation Differences
The difference in account size is evident from the very beginning, the account planning state.
For small accounts, often you’ll have either one campaign that is just search, or two campaigns one that is search and another one for display placements. There isn’t a huge need to plan out all types of campaigns.
If you are doing a lot of geographic targeting, you might end up with a few additional campaigns because of your geographic targeting, but beyond location targeting you usually don’t have many campaigns.
There are two reasons for this:
- Each campaign adds complexity to the management of the account, and small accounts don’t need to take up your entire day
- Small accounts usually have hard cap account budgets, and the more campaigns you create, the most management of budgets that occur – an unnecessary time waster in many cases
Larger accounts need to plan their accounts and campaigns before they even look at keywords. I find that making a chart or spreadsheet of all the accounts and campaigns first helps to plan targeting, budgets, and responsibilities.
Here’s a very basic chart (often there are thousands of campaigns in huge accounts, so these can become quite elaborate):
While a small account campaign structure can be determined in minutes; a massive account might take a few days just to plan the campaign structure before you even start thinking of keywords.
Ad Group & Keyword Selection
For small and mid-sized accounts, you will end up doing quite a bit of keyword research to determine the keywords you want to add to your paid search account. This process can take a few hours to a few days; but it is necessary to find the best keywords for the limited spend.
Often in large account, the account is created without doing any or a limited amount, of keyword research using external tools.
In some cases, the entire initial keyword research is done between the site, sitemap, product feed, and other backend data. If you sell clothes, you’ll take a list of your clothing types, adjectives, brands, sizes, colors, and other attributes and lay out an ad group strategy.
Next, you’ll fill in each ad group with the appropriate keywords. You will most likely add more keywords in the future; but you can launch an account with several hundred thousand keywords without ever using a keyword tool.
Keyword Match Types
Usually, small accounts want to start with lots of control and large accounts want to start with lots of data. This leads to a difference in match type selection.
It’s not uncommon to plan a large account to start with every match type. Since you’re building the account in spreadsheets, and then often using a 3rd party bid management system post launch, there isn’t a time difference between using one match type versus four.
The exception usually applies to those not wanting to use a 3rd party bid management system. In that case, you will usually see a lot of exact and modified broad match. Once the keywords have data, then the bid management will take over.
With this approach, you can expect some keywords to make money and others to lose money early on in the account lifecycle.
A small account does not want to lose money from the start. Often, the $1,000 spend is a significant amount of money for that company and is a big investment; therefore, they expect to make money immediately, or at least see progress and leads flow within a few days to weeks of starting a campaign.
Due to this level of control, you often see a lot more phrase match being used if the account is national.
Many small accounts are just local. When you target a small area, you usually see many keywords with low search volumes. In this case, you often want to show for many variations of keywords just to try and hit volume targets while you determine what is converting.
In that case, its common to see a lot more broad match being used for local accounts, especially if they only serve a single city or a small radius around a single location.
Once an account is created and running, then it needs to measure success. I find that small and large accounts often have to measure success differently. A large account can generate so much data, that by using the PPC engine conversion scripts along with their own analytics, they will be able to measure success quite easily.
There are often two major problems measuring small accounts.
First off, many small accounts are local businesses. These businesses want phone calls and in-person shoppers. This means that to measure success, you need phone call tracking, which is an added layer of complexity that many businesses don’t want to attempt.
For in-store visits, you need to coordinate some type of online offer, such as a coupon, that can be redeemed in the store that is then coordinated back to a keyword or ad. This level of complexity is often too time consuming for many businesses.
Therefore, small businesses are often left with either measuring online contact requests that make up only a small percentage of their total leads or by using interaction goals within analytics to measure success. Neither of these are ideal as they don’t always measure true revenue. However, this is still better than not measuring any goals and bidding blindly.
While some large accounts do measure phone calls, often the increased cost of sophisticated phone call tracking is only a small percentage of their overall paid search spend, so it is worth the investment to track calls.
The challenge of measuring many large accounts is usually getting a true revenue picture. When you add in phone calls, newsletters, ecommerce sales, lead generation, and brand measurement, the amount of conversion data and decisions to be made can be overwhelming.
You need to put an attribution and revenue measurement system in place to see a true revenue and profit picture of the paid search accounts. While this can take a while to put into place, the long term gains are worth the effort.
A large account can generally get to a point of measuring true revenue from paid search. A small account often cannot (due to time, complexity, cost, or data points) get a true revenue picture.
A large account makes bid decisions based upon revenue by keyword or placement. A small account makes educated guesses about revenue at the ad group level.
Best practices are called best practices in paid search because they have provided better results than alternate routes of account creation and management. However, best practices are not always possible based upon resources, cost, and available data.
The same is true for processes. For a large account, you must consider match types, ad group, and campaign organization before you start creating your account. If you don’t, then you risk having to redo hours or days of work. That is time that could have been saved with a comprehensive planning phase before the account was created.
For small accounts, you can change your mind as you go if you need to because there’s little additional time involved to add or change a match type when you only have a few hundred keywords. While you want to plan out how the account will look, you don’t need to triple check the structure before the team gets to work creating the account.
The same is true for ongoing management processes. The time intervals between examining data, such as search queries, for a small and large account are much different based upon data accumulation and the amount of time spent on the account each month.
If you’d like to read more about the differences between managing small and large accounts, please let me know in the comments.
The next time you hear someone declare that something is a best practice or a process everyone should follow; consider your account size and spend, and then qualify those thoughts based upon the size of the accounts you manage.
Some opinions expressed in this article may be those of a guest author and not necessarily Search Engine Land. Staff authors are listed here.