Payment for targeted actions

How does it work? Depending on the business model of the company, the target action is a purchase, an order of services, a call to the company, registration, subscription, an application for consultation. At the start, the agency and the customer make a list of target actions and agree on payment. Most often, the client pays a subscription fee and a reward for leads. Such a scheme guarantees the client that the agency will solve his business problems, and not just raise positions and traffic.

Possible problems. To ensure the growth of leads, a whole team of marketers, SEO specialists, developers should work. Improvements on the site, content optimization, work with product cards, CJM processing, ordering services / shopping cart, and so on will be required. Therefore, the tariffs for this payment model are quite high. On the other hand, the client understands that he is not paying for air, but for specific user actions.

Not every targeted action brings profit to the company. For example, not every call is converted to a customer. Therefore, in some cases, the payment for leads may not pay off either.

Payment per turnover

How does it work? A new payment model linked to turnover growth. This is the next stage in the evolution of the pay-for-leads tariff. The scheme focuses on the main KPI of any business — revenue and revenue.

The agency together with the client analyzes the current traffic, conversion, revenue, looks for growth points, makes a forecast and an action plan to increase the turnover of the company.

For example:

we need traffic growth -> to do this, we need to analyze competitors, rebuild the semantic core and optimize pages

increase conversion -> to do this, we improve the usability on the site and redo the landing pages

change the conditions of return and warranty -> we need a marketing analysis and solution

To determine the cost of services, the agency calculates the cost of its actions and lays marginality. The client is invited to invest X rubles to get a turnover growth of Y%. If the turnover growth is more or less than predicted, then the client will pay, respectively, more or less.

Most businesses do not believe that if they increase the budget for search engine promotion by 1.5-2 times, then returns will increase proportionally.

Our task is to break this stereotype and make the SEO channel predictable.

Dmitry Sevalnev, Head of the Pixel Plus SEO and Advertising Department.

The turnover-linked payment scheme is suitable for medium and large companies that have established business processes and are striving to scale. It is not suitable for startups that are just looking for their business model.

Possible problems. The client has to share with the agency the financial indicators of the business and other information. For those who are not ready to disclose data, a pay-per-turnover tariff is not suitable.

Openness and trust are also required in other matters. For example, in some cases, agency employees need to engage not only in SEO, but also delve into other business processes, configure them together with the client.

For most agencies, switching to a turnover—linked payment scheme is too bold a decision. There are many different “what if…”.

— What if an agency brings a client through SEO, but the sales department fails the deal?

— What if there are problems with the quality of the product?

— What if the agency makes extra efforts, and the turnover does not grow?

Paying for turnover breaks the barrier between the SEO area of responsibility and the business area of responsibility. The client with such a scheme practically does not risk. The agency assumes the main risks.

So, there are five main payment models for SEO services, and not one of them is perfect. The simplest and long—existing one is payment for the amount of work. This is a scheme that is convenient for the agency, but risky for the client, because it does not take into account his business KPIs. The further evolution of payment models is the way from linking to SEO indicators (payment for positions and traffic) to accounting for business indicators (payment for targeted actions and turnover).

Search for growth points for SEO and performance through product segment analytics. Printbar


When you come to a large e-com, there is a strong expectation that everything is already there: beautiful dashboards, structured data, analytics at a glance. But this is not the case — for the most part you need to do everything from scratch. What difficulties are there in a company where any word is a keyword? How to choose analytics tools?

Project specifics:

this is merch, clothing and everything-everything-everything with a pattern, print, image

infinite semantics

there are no physical goods, but in fact there are a million of them

2 major players in the Russian market

TOP10 is either us and a competitor, or partners

What was the problem of the project:

it was not intelligence for traffic channels in conjunction with other categories

there was no semantics, no historicity

reports in Yandex Metrika, Google, and publishers have a strong error

few tools for finding points of growth

the semantics (meaning ideas) formed adjacent departments, including the context, designers (both regular and external) — no single system

you can’t just create +100500 goods

there are copyrights and licenses

Problems in SEO:

at that time there was a drop in traffic by about 2 times

flat structure

a lot of 18+ queries

duplicates of names of prints = repetition of cards on the first listing

the impact of world events

typical SEO problems: technical, logs, meta, cf

weak SEO admin

canonical on traffic pages

uniqueness of content in the subject

The management had clear tasks: to increase organics by 2-3 times, set up processes and build an SEO analytics system (catalog data, semantics, traffic, sales, etc.). At the start of the case, there was a good team — a cool IT department and designers. But at the same time, there was a lack of brand strength, tools for end-to-end SEO analytics, strategy and systematization in general. Many processes had to be built from scratch, and that’s how it was!