Merged PR 3120: Creation of growth score by deal model for top losers report (intermediate)

# Description

Creates a model to identify deal growth based on YoY performance of Created Bookings, YoY performance of Listings Booked in Month and one month shifted YoY performance of Revenue.
Also added weighted score to account for revenue size.

# Checklist

- [X] The edited models and dependants run properly with production data.
- [X] The edited models are sufficiently documented.
- [X] The edited models contain PK tests, and I've ran and passed them.
- [ ] I have checked for DRY opportunities with other models and docs. **Probably something can be done here, sorry I've not checked.**
- [X] I've picked the right materialization for the affected models.

# Other

- [ ] Check if a full-refresh is required after this PR is merged.

Related work items: #22635
This commit is contained in:
Oriol Roqué Paniagua 2024-10-11 07:20:35 +00:00
parent 52f01adc11
commit 7f9c038fc0
2 changed files with 667 additions and 0 deletions

View file

@ -589,3 +589,443 @@ models:
description: The value or segment available for the selected dimension.
tests:
- not_null
- name: int_monthly_growth_score_by_deal
description: |
The main goal of this model is to provide a growth score by deal and month.
The idea behind it is that each deal will have some business performance
associated to it over the months, and that comparing how it is currently
performing vs. historical data we can determine whether the tendency is to
grow or to decay. This is specially useful for AMs to focus their effort
towards the clients that have a negative tendency.
The computation of the growth score is based on 3 main indicators:
- Created bookings
- Listings booked in month
- Total revenue (in gbp)
The main idea is, for each deal, to compare each of these metrics by
checking the latest monthly value vs. the monthly value of the equivalent
month on the previous year - in other words, a year-on-year (YoY) comparison.
We do this comparison by doing a relative incremental.
The growth score is computed then by averaging the outcome of the 3 scores.
Lastly, in order to provide a prioritisation sense, we have a weighted growth
score that results from the multiplication of the growth score per the revenue
weight a specific deal has provided in the previous 12 months.
However, this is not strictly true for Revenue because 1) we have an invoicing
delay and 2) in some cases, monthly revenue per deal can be negative. In this
specific cases, the YoY comparison is shifted by one month, and an effective
revenue value for the revenue share is computed, that cannot be lower than 0.
In order to keep both a properly set up score and revenue consistency, both
a real revenue value and effective revenue value are present in this model,
while no MoM or YoY value is computed if negative revenue is found.
Lastly, this model provides informative date fields, deal attributes, absolute
metric values and MoM & YoY relative incrementals to enrich reporting.
tests:
- dbt_utils.unique_combination_of_columns:
combination_of_columns:
- date
- id_deal
columns:
- name: date
data_type: date
description: |
Date corresponding to the last day of the month. Given month
metrics are inclusive to this date. Together with id_deal, it
acts as the primary key of this model.
tests:
- not_null
- name: id_deal
data_type: string
description: |
Unique identifier of a Deal. Together with date, it acts as
the primary key of this model.
tests:
- not_null
- name: main_deal_name
data_type: string
description: |
Main name for a Deal, representing the client.
tests:
- not_null
- name: main_billing_country_iso_3_per_deal
data_type: string
description: |
Main billing country for this client. In some cases
it can be null.
- name: deal_lifecycle_state
data_type: string
description: |
Identifier of the lifecycle state of a given deal
in a given month.
- name: given_month_first_day_month
data_type: date
description: |
Informative field. It indicates the first day of the
month corresponding to date.
If date = 2024-09-30, this field will be 2024-09-01.
tests:
- not_null
- name: previous_1_month_first_day_month
data_type: date
description: |
Informative field. It indicates the first day of the
previous month with respect to date.
If date = 2024-09-30, this field will be 2024-08-01.
It can be null if no previous history for that
deal is found.
- name: previous_12_month_first_day_month
data_type: date
description: |
Informative field. It indicates the first day of the
month with respect to date, but on the previous year.
If date = 2024-09-30, this field will be 2023-09-01.
It can be null if no previous history for that
deal is found.
- name: previous_13_month_first_day_month
data_type: date
description: |
Informative field. It indicates the first day of the
previous month with respect to date, but on the previous year.
If date = 2024-09-30, this field will be 2023-08-01.
It can be null if no previous history for that
deal is found.
- name: aggregated_revenue_from_first_day_month
data_type: date
description: |
Informative field. It indicates the first day of the
month from the lower bound range in which the revenue
aggregation is computed.
The aggregation uses the previous 12 months in which we
know the revenue, thus:
If date = 2024-09-30, this field will be 2023-09-01.
It can be null if no previous history for that
deal is found.
- name: aggregated_revenue_to_first_day_month
data_type: date
description: |
Informative field. It indicates the first day of the
month from the upper bound range in which the revenue
aggregation is computed.
The aggregation uses the previous 12 months in which we
know the revenue, thus:
If date = 2024-09-30, this field will be 2023-08-01.
It can be null if no previous history for that
deal is found.
- name: given_month_revenue_in_gbp
data_type: decimal
description: |
Monthly value representing revenue in GBP
for a specific deal. This value corresponds to
the given month. This value can be negative,
but not null.
tests:
- not_null
- name: previous_1_month_revenue_in_gbp
data_type: decimal
description: |
Monthly value representing revenue in GBP
for a specific deal. This value corresponds to
the previous month.
This value can be negative.
This value can be null, thus indicating that no
history is available.
- name: previous_12_month_revenue_in_gbp
data_type: decimal
description: |
Monthly value representing revenue in GBP
for a specific deal. This value corresponds to
monthly amount generated 12 months ago.
This value can be negative.
This value can be null, thus indicating that no
history is available.
- name: previous_13_month_revenue_in_gbp
data_type: decimal
description: |
Monthly value representing revenue in GBP
for a specific deal. This value corresponds to
monthly amount generated 13 months ago.
This value can be negative.
This value can be null, thus indicating that no
history is available.
- name: mom_revenue_growth
data_type: decimal
description: |
Relative increment of the revenue generated in the
current month with respect to the one generated in
the previous month.
It can be null if any revenue used in the computation
is null or it's negative.
tests:
- dbt_expectations.expect_column_values_to_be_between:
min_value: -1
strictly: false
- name: yoy_revenue_growth
data_type: decimal
description: |
Relative increment of the revenue generated in the
current month with respect to the one generated 12
months ago.
It can be null if any revenue used in the computation
is null or it's negative.
tests:
- dbt_expectations.expect_column_values_to_be_between:
min_value: -1
strictly: false
- name: yoy_1_month_shift_revenue_growth
data_type: decimal
description: |
Relative increment of the revenue generated in the
previous month with respect to the one generated 13
months ago.
It can be null if any revenue used in the computation
is null or it's negative.
This field is used for the growth score computation.
tests:
- dbt_expectations.expect_column_values_to_be_between:
min_value: -1
strictly: false
- name: given_month_created_bookings
data_type: integer
description: |
Monthly value representing created bookings
for a specific deal. This value corresponds to
the given month. This value cannot be null.
tests:
- not_null
- dbt_expectations.expect_column_values_to_be_between:
min_value: 0
strictly: false
- name: previous_1_month_created_bookings
data_type: integer
description: |
Monthly value representing created bookings
for a specific deal. This value corresponds to
the previous month.
This value can be null, thus indicating that no
history is available.
tests:
- dbt_expectations.expect_column_values_to_be_between:
min_value: 0
strictly: false
- name: previous_12_month_created_bookings
data_type: integer
description: |
Monthly value representing created bookings
for a specific deal. This value corresponds to
monthly amount generated 12 months ago.
This value can be null, thus indicating that no
history is available.
tests:
- dbt_expectations.expect_column_values_to_be_between:
min_value: 0
strictly: false
- name: mom_created_bookings_growth
data_type: decimal
description: |
Relative increment of the bookings created in the
current month with respect to the ones created in
the previous month.
It can be null if the bookings created in the
previous month are null.
tests:
- dbt_expectations.expect_column_values_to_be_between:
min_value: -1
strictly: false
- name: yoy_created_bookings_growth
data_type: decimal
description: |
Relative increment of the bookings created in the
current month with respect to the ones created 12
months ago.
It can be null if the bookings created 12 months
ago are null.
This field is used for the growth score computation.
tests:
- dbt_expectations.expect_column_values_to_be_between:
min_value: -1
strictly: false
- name: given_month_listings_booked_in_month
data_type: integer
description: |
Monthly value representing the listings booked in month
for a specific deal. This value corresponds to
the given month. This value cannot be null.
tests:
- not_null
- dbt_expectations.expect_column_values_to_be_between:
min_value: 0
strictly: false
- name: previous_1_month_listings_booked_in_month
data_type: integer
description: |
Monthly value representing the listings booked in month
for a specific deal. This value corresponds to
the previous month.
This value can be null, thus indicating that no
history is available.
tests:
- dbt_expectations.expect_column_values_to_be_between:
min_value: 0
strictly: false
- name: previous_12_month_listings_booked_in_month
data_type: integer
description: |
Monthly value representing the listings booked in month
for a specific deal. This value corresponds to
monthly amount generated 12 months ago.
This value can be null, thus indicating that no
history is available.
tests:
- dbt_expectations.expect_column_values_to_be_between:
min_value: 0
strictly: false
- name: mom_listings_booked_in_month_growth
data_type: decimal
description: |
Relative increment of the the listings booked in month
in the current month with respect to the ones of
the previous month.
It can be null if the listings booked in month in the
previous month are null.
tests:
- dbt_expectations.expect_column_values_to_be_between:
min_value: -1
strictly: false
- name: yoy_listings_booked_in_month_growth
data_type: decimal
description: |
Relative increment of the listings booked in month
in the current month with respect to the ones of 12
months ago.
It can be null if the listings booked in month of 12
months ago are null.
This field is used for the growth score computation.
tests:
- dbt_expectations.expect_column_values_to_be_between:
min_value: -1
strictly: false
- name: deal_revenue_12_months_window
data_type: decimal
description: |
Total aggregated revenue in GBP generated by a deal
in the months from the period ranging from the
aggregated_revenue_from_first_day_month to
aggregated_revenue_to_first_day_month.
It can be negative if the sum is negative.
It cannot be null.
tests:
- not_null
- name: effective_deal_revenue_12_months_window
data_type: decimal
description: |
Effective aggregated revenue in GBP generated by a deal
in the months from the period ranging from the
aggregated_revenue_from_first_day_month to
aggregated_revenue_to_first_day_month.
All negative monthly revenue values are settled as 0,
thus this value should not be reported.
It is used for the deal contribution share with respect
to the global revenue. It cannot be null.
tests:
- not_null
- dbt_expectations.expect_column_values_to_be_between:
min_value: 0
strictly: false
- name: effective_global_revenue_12_months_window
data_type: decimal
description: |
Effective aggregated revenue in GBP generated by all deals
in the months from the period ranging from the
aggregated_revenue_from_first_day_month to
aggregated_revenue_to_first_day_month.
All negative monthly revenue values are settled as 0,
thus this value should not be reported.
It is used for the deal contribution share with respect
to the global revenue. It cannot be null.
tests:
- not_null
- dbt_expectations.expect_column_values_to_be_between:
min_value: 0
strictly: false
- name: deal_contribution_share_to_global_revenue
data_type: decimal
description: |
Represents the size of the deal in terms of revenue. In
other words, what's the percentage of the global revenue
that can be attributed to this deal. It cannot be null.
tests:
- not_null
- dbt_expectations.expect_column_values_to_be_between:
min_value: 0
strictly: false
- name: deal_contribution_rank_to_global_revenue
data_type: integer
description: |
Represents the ordered list of deals by descending size
in terms of revenue. It cannot be null. If more than one
deal have the same share, the order is not under control.
It cannot be null.
tests:
- not_null
- name: avg_growth_score
data_type: decimal
description: |
Represents the average score of YoY growth of created
bookings, YoY growth of listings booked in month and
YoY shifted by one month of revenue.
It indicates the tendency of growth of the deal without
taking into account its revenue size. It cannot be null.
tests:
- not_null
- name: weighted_avg_growth_score
data_type: decimal
description: |
It's the weighted version of avg_growth_score that
takes into account the client size by using the revenue
contribution share of that deal to the global amount.
It's the main indicator towards measuring both growth
(if positive) or decay (if negative) while weighting
the financial impact this deal tendency can have.
tests:
- not_null