KPI, “L’origine du monde”

UXMetric - Mesurer l'expérience utilisateur

KPI – Key Performance Indicators

A KPI or Key Performance Indicator is information that allows an individual to assess a situation.

Any collection and use of data-based approach is based on indicators defined by the relationship between the qualification of a quantum and its quantification.

We will use two attributes to define a datum, the dimension to qualify it and the metric to quantify it. For example, on the digital supports one consults contents, these contents are qualified (Dimensions) as: pages, products, music, videos, articles etc.

In order to measure the impact of these contents we will quantify the views, listening, purchases, viewings (metrics).

Thus we can define simple performance indicators such as the number of page views, the number of music plays.

KPIs will measure performance in 4 directions

  • Marketing, piloting sources of audience acquisition,
  • The user experience, evaluate the usability of a service,
  • Customer knowledge, enrich and specify customer data,
  • Technology, evaluate the compatibility of a device with the standards in force,

KPO – Key Performance Objectives

In the continuity of the exploitation of the performance indicators we will try to evaluate or measure more complex “concepts” by combining several KPIs, so we can evaluate

  • The efficiency, the efficacy of a device
  • A degree of optimization, a change of state
  • Measuring customer satisfaction
  • A degree of awareness on a market

How to choose KPIs?

The Influence of Lean Analytics

Based on the approach of the project launch and more particularly the startup launch, the Lean Analytics method can in fact be transposed on any subject, with the starting assumption that the current situation proposes a reference index and that the future work will aim to improve this situation, whether from a marketing, economic, or user experience point of view.

How to choose the indicators?

In the Data driven approach, the focus is on the word “driven”. To control with data, it is necessary that it can vary for different behaviors or actions.

In fact by asking ourselves the following question, we can validate the relevance of following one datum or another. “What am I going to do differently based on this information?”

If we cannot answer this question, we should not give importance and probably not worry too much about this datum (pay attention to “Vanity KPI” i.e.: number of page views does not inform about the number of potential prospects nor on their interests for the brand).

Following few performance indicators will increase accuracy and our decision-making powers (OMTM – One Metric That Matters).

The quality of the indicator (KPI) is in question

  • A good indicator is a rate or a ratio because a ratio intrinsically conveys the notion of comparison (example: number of achievements of an objective on a group / segment of population and / or on a time range)
  • A good indicator should allow you to modify your behavior

For example: number of goal achievements / population segment allows us:

  • to refine the segment of the population to maximize the goal achievement rate and reorient “incompatible” individuals with this goal to a new one.
  • to modify the objective so that it is more in adequacy with this population

Excerpts from the book Lean Analytics (Alistair Croll, Benjamin Yoskovitz)

Data-Driven tactics through Practice

MARKETING: The AIDA Method (Attention, Interest, Desire, Action)

The AIDA method is a model used in marketing to unlock the components of the attitude (Cognitive: judgments, beliefs, knowledge; affective: favorable and unfavorable feelings ;Conative: tendencies of action) hindering the purchase or the commitment.

This model is based on the crossing of 4 successive stages:

  1. Capturing the attention
  2. Generating interest
  3. Provoking desire
  4. Encouraging action

In order to bring value through data, we will try:

  • to evaluate the performance of the method; like a conversion tunnel, we will be able to quantify the users having crossed the 4 steps successively, and thus highlight the break points (abandonment)
  • to measure for each step the brand’s ability to validate the goal of the milestone (such as KPO)


Measuring a brand’s ability to capture users’ attention

  • Number of searches on search engines on a non-brand / time-domain theme
  • Number of new users for sources of organic traffic, campaign, referrer, and Paid / time range
  • Number of users reached by posts on social networks / time range

Measuring the capacity of a brand / device to interest its audience.

  • Number of queries of a brand / time range
  • Number of recurring visitors / total number of visitors
  • Frequency and recency of visits
  • Number of users with a time greater than or equal to the average viewing time
  • Type of pageviews / recurring users
  • Number of pages per repeat visitor
  • Number of registrations or contacts / total number of visitors
  • Number of visits by email or social sources / total number of visitors

Measuring users’ desire for a brand’s products / services (performance goal measure)

  • Frequencies of recent visits in a short time
  • Measurement of the purchase process not finalized
  • Visits of contextualized pages (around the target products)
  • Number of deep page hits on direct source

Measuring the brand’s ability to transform its audience.

  • Number of transactions for direct traffic sources
  • Number of transactions following  an abandoned basket’s revival
  • Number of satisfaction comments / number of product recommendations

Client Experience

A / The “MAYA model”

The acronym MAYA “Most Advanced Yet Acceptable”, defined by designer Raymond Loewy to summarize his philosophical perception of design highlights the objectives of measures to evaluate the success of a launch by accepting the evolution of the product.

Measuring acceptance of the evolution of a product involves the evaluation of:

  • Users (capacity, maturity)
  • The degree of divergence of the product with respect to the market and uses (capacity of the brand to open the field of possibilities)
  • The degree of convergence of the product according to the maturity of the users (brand effort to adapt to the capabilities of its users)
  • The degree of freedom in terms of the creation of use of the brand (adequacy between the side of the possible and towards which it is necessary to trend)

B / Measuring the actual use of a solution, product, or service

Through technical and quantifiable KPIs we can evaluate the use of a solution

  • Recency and frequency & time of use / equipment used (example: consultation terminal)
  • Duration of use / period / material
  • Equipment used compared to public statistical data on equipment
  • Support / Help Research
  • Maintenance contact, update

For more information on the concept of usability of products / solutions, we invite you to read the article “Usability of interfaces”.


The AARRR matrix (KPIs, pirates of growth) from Lean Startup is also discussed in the article “In Search of Performance”

Acquisition (How do users get to you?):

Measuring the efficiency of the acquisition channels.

Indicators to follow, rates of:

Activation (what is the level of satisfaction / commitment of your users towards the presentation of the service / product, the goal here is retention):

Measuring “Activated” users committed to the service or product.

Indicators to follow,  rates of:

  • Registration to a newsletter,
  • Account creation,
  • Users performing a search,
  • Downloading a white paper / number of visitors, etc.

Retention (Are your users returning?):

Measuring the recurrence of use of the proposed service.

Indicators to follow:

  • Frequency of visits / group of users,
  • Visit rates with research use,
  • Positive response rates.

Revenues (How to generate income?):

Measuring the users who bought your product, measuring the turnover by sources of acquisition

Indicators to follow:

  • Turnover by source of acquisition, Turnover / frequency of visits,
  • Turnover per product,
  • Average basket by source of acquisition.

Referral (Measure the snowball effect)

Measuring the recommendation rates of your products or your brand (sponsorship / social networks)

Indicators to follow:

  • Affiliation rate,
  • Sponsorship rate

A broad spectrum approach

To conclude, these “three models” are therefore linked and complementary. They propose a holistic evaluation grid for defining the strategy and optimizing the sales performance of a sales system.

“Holistic marketing is the development and implementation of broad-based, interconnected programs, processes, and marketing actions. Holistic marketing is relational, integrated, internally distributed, performance-oriented and socially responsible. “*

* Kotler, Keller & Manceau, Marketing Management, 14th edition, Pearson, 2012, p. 22


(The lists of performance indicators presented are not exhaustive)