Ask any advertiser if their PPC cost is too high, and you will most likely get a resounding, “Yes!” The cost per click is a result of an auction. The number of advertising slots in Google is limited. Advertisers, who want to buy more and more clicks, place higher and higher CPC bids in order to outbid their competitors, who in turn, do the same.

As a result, the CPC grows until advertisers realize they are paying too much. This very often happens when the bill for PPC clicks is higher than the total sales margin. For this happenstance managers use the euphemism, “overinvestment in ads”.

## When We Spend Too Much

In fact, overinvestment occurs much earlier, even before the company produces losses. Buying more clicks requires higher CPC, and it causes higher conversion costs and smaller profits per transaction, even though there are more transactions and total profit grows. At a certain point, however, the increased number of transactions does not compensate for decreasing margins and total profit starts to decrease.

Mathematical models help to identify the sweet spot where the profit is the highest. Increasing bids makes sense only if **ROI > 1/E**, where E is the price elasticity of clicks. Elasticity is defined as:

For example, if higher bids cause a growth in CPC of 10% and clicks increase by 20%, then elasticity equals 2. If CPC grows by 10% and clicks increase by just 5%, then elasticity is 0.5.

*Detailed derivation of the ROI>1/E formula is presented in the article How to Drive Profit With PPC Campaigns.*

The price elasticity of clicks in PPC systems like AdWords generally decreases as CPC increases:

ROI depends on CPC, because higher cost per click means higher conversion cost. Therefore, every change of bids changes the ROI as well as the elasticity. When CPC grows, both ROI and elasticity decrease, and quickly ROI can become smaller than 1/E.

The PPC optimization in e-commerce practice usually means maximizing the revenue according to the target ROI set by the company management. The ROI > 1/E formula tells us that this approach can be counterproductive. If one wants to maximize profit, the target ROI should not simply be an arbitrary decision of company decision-makers. The optimum ROI depends solely on the shape of the elasticity curve.

Each campaign and keyword may have different elasticity, and this elasticity can differ on different devices (desktop/mobile) and in other dimensions. Therefore, the target ROI cannot be the same across all keywords, devices, etc. Even if the target ROI is set by the company management at a certain level, for example, the company wants to maximize the income at ROI greater than or equal to zero – the elasticity adjustments to target ROI help to increase the revenue without changing the total average ROI. This process is described in How to Maximize the Revenue at a Targeted ROI.

The consequences of the profit-driven approach are sometimes counterintuitive. Instead of moving budgets from campaigns with lower ROI to those with higher profitability, we ought to consider increasing bids for keywords with lower ROI, but with higher elasticity (for example, balancing around “first page bid” value) and then we might decrease bids for the best-converting keywords with lower elasticity (for example, keywords with very high impression share and average position close to the first position at the top of the search results page).

The elasticity should be measured and monitored. AdWords native features, such as bid simulator and A/B experiments, help to obtain necessary data to estimate the elasticity.

In practice, the elasticity of keywords for ads displayed on the top four slots above the search results page, selected by the most aggressive occupants, is usually around one or smaller. So, if they want to increase traffic by 10%, the CPC will also grow by 10% or more. According to ROI>1/E formula, such an increase makes sense only if the current ROI is greater than 100%. In a competitive market, such profitability is rare.

This is the typical situation, but it can differ in a particular case. However, the leading advertisers most often increase their profits when they decide to be less aggressive in PPC, especially in SEM.

## Click Value and Price Optimization

Besides PPC optimization, advertisers usually try to increase the conversion rate by better web usability and site speed, cross-selling, retargeting – basically, anything that may increase the chance that the visitor becomes a client. One of the forgotten points of optimization is the price of the product itself.

Sometimes advertisers offer discounts in order to encourage the visitors to finalize the transaction. In the case of the majority of products, the lower the price is, the easier it is to sell the products. Discounts usually increase the conversion rate.

But does it increase revenue and profit?

Now it’s worth emphasizing the difference between revenue and margin. The revenue is the total price paid by the customer, but it’s not the same as the profit. The goods sold have to be produced or purchased first; therefore, the margin is the revenue less cost of materials, production, packaging, transport etc. This means all the direct costs required to deliver the product to the customer. Therefore, we should always compare our PPC conversion cost, not to our revenue, but to the margin per transaction (M/Conv).

The profit is the difference between the margin per transaction and the conversion cost, multiplied by the number of transactions (conversions):

*Profit = Conv × (M/Conv – Cost/Conv)*

This equation can be also written using the number of visitors/clicks (Clk):

*Profit = Clk × (Vclk – CPC)*

where Vclk is the click value, equal to conversion rate (CR) multiplied by the margin per transaction:

*Vclk = CR × M/Conv*

The conversion rate generally does not depend on CPC and number of clicks, and it is mainly related to everything that happens after the click. Margin per transaction and click value do not depend on the number of clicks either. The higher the click value is, the higher the profits. Therefore, in order to maximize profits, advertisers should maximize neither the conversion value nor the conversion rate. They should *maximize the click value*.

Maximization of click value is a process independent from PPC optimization, because click value does not depend on the number of clicks and CPC.

How does one maximize the click value? Conversion rate and margin per transaction are not independent values. In order to increase margin, we increase our prices. But that decreases the conversion rate because some clients find the new price too high. The optimization of click value is a trade-off between the margin and conversion rate.

In order to find the sweet spot, we conduct A/B tests and observe the reactions of clients to different prices. Also, we can use the term “elasticity” here. In this case, it’s the *elasticity of demand*. Profits are highest if the elasticity of demand is equal to -1. For example, if we increase our margin by 10% and the number of transactions drops only by 5%, then the elasticity is -0.5. That means that we should increase the prices. If a 10% increase of margin results in 15% loss of transactions, the elasticity is -1.5 and means that the prices are too high. If an increase of margin by 10% decreases the number of transactions by 10% too, the elasticity is -1 and our prices are close to optimum.

In practice, e-commerce businesses are reluctant to raise prices and margins. The consensus is that if you raise prices, customers will go somewhere else. This is partly true, but what if higher margins compensate with a vengeance for the loss of the most price-sensitive customers? What are the results of previous discounts? If the decrease of margin by 30% increases the conversion rate by 10%, the discounts are not profitable.

The elasticity curve is continuous. Therefore, in this case, we may expect that doing the opposite, i.e. increasing the price, should make our profits higher, even though we may have fewer transactions and probably less revenue.

It is also possible, that your prices are too high, but the only way to check that is by testing it. The most common mistake is that after the revenue grows as result of discounts, managers decide that lower prices deliver better results. They don’t count the profits. They maximize revenue instead of click value. A large share of e-commerce businesses increase their profits when they decide to raise prices and accept that they will sell a little less.

## The Miracle of Higher Click Value

Does it all mean that in order to increase profits, we should raise prices, decrease CPC spend, and accept lower revenue? Not necessarily. Even if the maximization of click value requires higher margins and, in consequence, a lower conversion rate, this optimization may open up new opportunities in PPC traffic acquisition.

This is because increased click value increases our ROI. According to the ROI > 1/E formula, we are able now to expand the PPC campaign to areas of lower elasticity.

See the example below, column A: The advertiser spends $100,000 on PPC ads buying 100,000 clicks at $1 CPC. With a conversion rate of 1%, it produces 1000 transactions. At a unit price of $1450, the margin per transaction is $150, and the total margin on all transactions is $150,000. The profit after PPC cost is $50,000.

In order to increase the traffic, the advertiser decides to increase PPC bids (column B). As a result, the CPC grows to $1.10 (+10%), and the number of clicks grows to 120,000 (+20%). Transactions also grow by 20% to 1200 and so does the total margin (growth to $180,000). However, after the PPC cost, the profit decreases to $48,000. This is not a surprise since the elasticity was 2 (20% more clicks with 10% CPC increase). Therefore, the ROI required to increase the bids should be greater than 50%. The actual ROI was equal to 50% (column A). So the campaign already was operating at an optimum level, and the increase of bids produced losses.

So the advertiser decides to reverse the bids to previous values and optimize the price of the product (column C). The clicks, CPC and ad spend are therefore the same as in column A. The unit price increases by $49, from $1450 to $1499, and the margin per transaction grows by $49 from $150 to $199 (+32.7%). The higher price means the conversion rate decreases from 1.0% to 0.9% (-10%), because one in ten customers decides to buy somewhere else. So, the elasticity of demand is -0.31 (far from the optimum of -1) and the price change has increased the value of click. The total margin grows to $179,100, and the profit also grows to $79,100.

It means that the ROI is now 79%, and, as we already know (from PPC data in the column B) that the elasticity of clicks is 2, it means that the ROI > 1/E (79% > ½). Therefore, we can think about PPC expansion. The advertiser changes the campaign (column D), again increasing CPC to $1.10, and the number of clicks grows to 120,000 again. This time, however, the total margin grows to $214,920 and the profit after PPC cost grows to $82,920.

If we compare columns A and D, we can see that not only the profit grew, but also the total margin and number of transactions are higher. The optimization of prices allowed buying more expensive PPC traffic that the advertiser could not previously afford.

The example above shows how important the optimization of click value is. Although this process is independent from the PPC campaign parameters, it can open new opportunities in advertising and allow expansion that could not be previously considered.

If PPC clicks are too expensive for you, you might need to change your prices. Although a big part of e-commerce businesses tend to keep their prices too low and over-invest in PPC, it can be different in each particular case. It’s worth to stress that the only way to determine if the CPC bids and prices are too high or too low is to measure the elasticity of clicks and demand and then modify CPC bids and prices up or down accordingly.

*Image Credits*

*All images by Witold Wrodarczyk/Adequate Interactive Boutique*