# Description
As raised by Kayla, Retained metrics are not computed correctly in AM Account Margin report, see screenshot:
Issue lies in the fact that Guest Payments contains more dimensions than the rest of the models. This is generating duplicates in the combination_of_sources CTE, but these are effectively hidden because of the sum() applied of the metrics.
In the case of a Deal having in the same day Guest Payments with different dimensions from those selected, AND on the same day having a value on Invoiced Revenue or Host Resolutions, then the Invoiced Revenue or Host Resolutions would be duplicated.
This PR aims to fix this issue, and also adds a few tests on the base model for AM Account Margin to ensure this is kind of error raises an alert.
I'll be debugging other affected models and increase the robustness test-wise; as well as gathering the full impact.
# 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.
- [ ] 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: #29227
# Description
Small changes to exclude "irrelevant" data. All data is relevant I guess but this is less.
* Exclude churned accounts after 3 months.
* Exclude any data after 24 months.
# 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.
- [ ] 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: #28998
# Description
Creates a new model for reporting. It includes any relevant metric at Deal level in a similar format as we do for `mtd_aggregated_metrics`. Additionally, there's few Deal attributes - from Hubspot, segmentations, lifecycles, etc.
In order to dynamically choose which metrics are relevant on a Deal level, I modified the configuration in `int_mtd_aggregated_metrics` so the extraction is under control.
# 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.
- [X] I have checked for DRY opportunities with other models and docs.
- [X] I've picked the right materialization for the affected models. **Worth discussing the possibilities to include indexes in the future**
# Other
- [ ] Check if a full-refresh is required after this PR is merged.
Related work items: #28998
# Description
To be able to compute By Deal KPIs similarly as we do for Main KPIs we need to propagate the "By Deal" dimension first in intermediate. This does not impact reporting since there's a macro that already cuts the dimensions to be displayed, but I modified the name so it's clearer.
Changes:
* In `int_mtd_vs_previous_year_metrics`, I added a new `dimension_list`. This is applied to all initial CTEs that filter by dimension. Note that I added here the `by_deal` dimension.
* Modified the name of `get_kpi_dimensions_for_production` to `get_main_kpis_dimensions_for_production`. Now it's more explicit that this is the configuration for Main KPIs reporting. This affects both `business_kpis_configuration` and it's usage in `mtd_aggregated_metrics`.
* Modified the tests in `int_mtd_vs_previous_year_metrics` and `int_mtd_aggregated_metrics` to include the new dimension `by_deal`.
* It seems by adding this now autoformatting works again on this model! I'll tag all cases that are just because of autoformatting.
MD5 trick has been applied in `reporting.mtd_aggregated_metrics`.
# 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.
- [X] I have checked for DRY opportunities with other models and docs.
- [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: #28998
# Description
Removes old onboarding mrr models, that are no longer used.
dbt compiles correctly and no deprecated warning is showing anymore.
This finishes the refactor
Related work items: #28949
# Description
Puts in production the new KPIs model for Onboarding MRR.
The change has been validated with MD5 trick.
# 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.
- [X] I have checked for DRY opportunities with other models and docs.
- [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: #28949
# Description
Tags as deprecated the following models:
* `int_monthly_onboarding_mrr_per_deal`
* `int_mtd_agg_onboarding_mrr_revenue`
Project compiles.
Related work items: #28949
# Description
Removes the models:
* int_monthly_12m_window_contribution_by_deal
* int_monthly_churn_metrics
as well as their entries in the schema files.
Project compiles and KPIs run works.
This closes stage 3 of the refactor.
# Checklist
- [ ] The edited models and dependants run properly with production data.
- [ ] The edited models are sufficiently documented.
- [ ] The edited models contain PK tests, and I've ran and passed them.
- [ ] I have checked for DRY opportunities with other models and docs.
- [ ] I've picked the right materialization for the affected models.
# Other
- [ ] Check if a full-refresh is required after this PR is merged.
Remove previous Churn models
Related work items: #28948
# Description
Small but important change.
Churn metrics will be propagated in production by sourcing intermediate/kpis rather than the cross models.
# 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.
- [X] I have checked for DRY opportunities with other models and docs.
- [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: #28948
# Description
Churn models being tagged as deprecated for next week, for information.
# Checklist
- [ ] The edited models and dependants run properly with production data.
- [ ] The edited models are sufficiently documented.
- [ ] The edited models contain PK tests, and I've ran and passed them.
- [ ] I have checked for DRY opportunities with other models and docs.
- [ ] I've picked the right materialization for the affected models.
# Other
- [ ] Check if a full-refresh is required after this PR is merged.
Deprecation of Churn Models
Related work items: #28948
# Description
Decouples Churn/Account Margin model of `int_monthly_aggregated_metrics_history_by_deal_by_time_window` from the model `int_monthly_aggregated_metrics_history_by_deal`.
Changes:
* Removes dependency and reads from KPIs models.
* Removes unused deal stage variable
* Jinja template for time windows with dynamic selection to DRY
This has been tested with dbt audit tools and the md5 for pure refactoring, so output is exactly the same.
# 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.
- [X] I have checked for DRY opportunities with other models and docs.
- [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: #28947
# Description
Adds audit helper.
Removes computation of Total and Retained Revenue from cross models.
I've tested audit helper and the trick with the hashes, as discussed in the demo.
# 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. **Except for today's outlier detector!**
- [X] I have checked for DRY opportunities with other models and docs.
- [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: #28946
# Description
Fixes Blank stuff
# Checklist
- [ ] The edited models and dependants run properly with production data.
- [ ] The edited models are sufficiently documented.
- [ ] The edited models contain PK tests, and I've ran and passed them.
- [ ] I have checked for DRY opportunities with other models and docs.
- [ ] I've picked the right materialization for the affected models.
# Other
- [ ] Check if a full-refresh is required after this PR is merged.
Fix Business Targets
Related work items: #28958
# Description
Ensures Resolutions Payouts (in amount paid) is timely
# 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.
- [ ] I've picked the right materialization for the affected models.
# Other
- [ ] Check if a full-refresh is required after this PR is merged.
Same freshness in MTD and YTD/target for Resolutions Payouts
Related work items: #28560
# Description
I noticed this morning that after the name changes some exclusions have been messed up.
I think is about time we handle this properly. Following what we did for `ytd_mtd_aggregated_main_metrics_overview`, I apply the same logic: the metric display configuration is handled for each metric in the configuration rather than by some crazy-complex-name-logic.
I also took the opportunity to make Host Resolutions timely after a discussion with Chloe yesterday. Finance handles resolutions payments twice a week, and the "delay" in time is mostly coming from Resolutions accepting to pay someone vs. Finance handling the payment. In any case we're talking about a few days maximum difference, which I believe it's 1) process-related freshness, rather than data-related and 2) much better to have visibility in a timely manner rather than waiting for the 20th on next month.
# 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. **Logic is common among YTD and MTD models. At some moment this can be improved**
- [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: #28560
# Description
There's misalignment in the different areas in the business, but even within our own report...
**I'm not saying this is perfect** - but at least it forces a common Data Glossary within Main KPIs. I'd suggest later on reviewing naming - same as we need to do for revenue anyway -, but for the meantime, at least have consistency on our side.
Changes:
* Est. Billable Bookings -> Billable Bookings. We have other metrics stated as Estimated in the Data Glossary.
* Waiver and Resolutions payments to host are now called Payouts. This is not perfect but at least is clear we're paying out, so it's a cost.
* Host Resolutions Payment Rate is now explicitly mentioning how it's being computed to avoid confusion with the Payout Rate, that does not appear here yet, but appears in the YTD.
Additional Changes:
* Tests are also aligned with new names.
* Re-order booking display, so Cancelled Bookings (inclusion and exclusion) are shown below. Having them the first is weird.
# 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.
- [ ] 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: #28560
# Description
Changes:
* Creates Booking Fees in YTD/MTD
* Creates Booking Fees per Billable Booking in YTD/MTD. This is exposed forward.
* Removes code for Onboarding MRR in YTD/MTD.
Additional fixes:
* Small mistake - Revenue Churn is now considered as NEGATIVE
# 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.
- [X] I have checked for DRY opportunities with other models and docs.
- [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: #28560
# Description
Main Changes:
* Computes the total booking fees in `int_kpis__metric_daily_invoiced_revenue`
* Propagates within KPIs, including schema and tests
* Propagates within cross in both "by deal" and "by dimension/global"
* Propagates to main kpis tests
Small fix:
* In `int_mtd_vs_previous_year_metrics`, manual formatting because sqlfmt is broken on this model...
# 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.
- [X] I have checked for DRY opportunities with other models and docs.
- [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: #28560
# Description
Main changes:
* Modifies target seeds to include FY2025 and FY2026. This includes now absolute values for equivalent rate metrics. This also includes Booking Fees per Billable Booking, despite this metric is still not computed. This will be done in a separated PR.
* Removes Onboarding MRR as this requirement is no longer needed, and there has not been any target computed for this metric. This is modified to include Revenue Churn as an absolute figure.
Small note. Targets for FY 2025 rely on the actuals from Finance. Up to January 2025 included, these should match with our YTD values, but it's not the case due to small differences in Billable Bookings / Deals; as well as accrued revenue. Accrued revenue is very troublesome specially on Guesty side, since this affects Total Revenue, and this affects also Resolutions Payout Rate, and so on...
# 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.
- [X] I have checked for DRY opportunities with other models and docs.
- [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: #28560
# Description
Booking fees is widely used with different meanings, for old dash, for new dash, for both, etc. This is painful. First step to align on a proper naming is ensure that what we report in Main KPIs is clearly stated, which in this case, Booking Fees are now called Old Dashboard Booking Fees.
Changes:
* Modify `stg_seed__accounting_aggregations` seed to rename Booking Fees to Old Dashboard Booking Fees. This is for us to clarify. This is only applied for KPIs compute. I also added an empty space that I mistakenly forgot in the past for `financial_l3_aggregation`.
* Modify KPIs source, i.e., `int_kpis__metric_daily_invoiced_revenue`. Here I forcefully modify the name of the field to `xero_old_dashboard_booking_net_fees_in_gbp`.
* Propagate changes of downstream usages of `xero_booking_net_fees_in_gbp` to `xero_old_dashboard_booking_net_fees_in_gbp`. This affects all models, including the reporting model. On this one we still have both names to avoid breaking it. I will need to modify the data glossary in PBI anyway so I'll do this change as well.
* Modify displayed metric name from Booking Fees Revenue to Old Dashboard Booking Fees Revenue.
* Modify schema so it reflects the proper names, descriptions, and tests.
* Ensure outlier and completion tests still work after this change.
I confirm the field `xero_booking_net_fees_in_gbp` does not exist anymore in the rest of DWH after these changes, except for the abovementioned comment on the reporting line.
# 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.
- [X] I have checked for DRY opportunities with other models and docs.
- [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: #28560
# Description
Propagates Client Type to KPIs by Deal and AM models. This is just a distinction between Platform and APIs.
# 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.
- [X] I have checked for DRY opportunities with other models and docs.
- [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: #28181
# Description
Adds number format for display purposes
# 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.
- [X] I have checked for DRY opportunities with other models and docs.
- [X] I've picked the right materialization for the affected models.
# Other
- [ ] Check if a full-refresh is required after this PR is merged.
Adds number format
Related work items: #27609
# Description
Changes:
* Adds Onboarding MRR
* Refactors exclusion code for ongoing month / invoicing cycle
* Adds sign format on relative differences
# 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.
- [X] I have checked for DRY opportunities with other models and docs.
- [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: #27609, #27805
# Description
_Describe your motivation and changes here._
# Checklist
- [ ] The edited models and dependants run properly with production data.
- [ ] The edited models are sufficiently documented.
- [ ] The edited models contain PK tests, and I've ran and passed them.
- [ ] I have checked for DRY opportunities with other models and docs.
- [ ] I've picked the right materialization for the affected models.
# Other
- [ ] Check if a full-refresh is required after this PR is merged.
Revert "Adds Onboarding MRR"
Reverted commit `dab210c6`.
# Description
_Describe your motivation and changes here._
# Checklist
- [ ] The edited models and dependants run properly with production data.
- [ ] The edited models are sufficiently documented.
- [ ] The edited models contain PK tests, and I've ran and passed them.
- [ ] I have checked for DRY opportunities with other models and docs.
- [ ] I've picked the right materialization for the affected models.
# Other
- [ ] Check if a full-refresh is required after this PR is merged.
Revert "Refactor display exclusion for clarity purposes"
Reverted commit `fdec616f`.
# Description
_Describe your motivation and changes here._
# Checklist
- [ ] The edited models and dependants run properly with production data.
- [ ] The edited models are sufficiently documented.
- [ ] The edited models contain PK tests, and I've ran and passed them.
- [ ] I have checked for DRY opportunities with other models and docs.
- [ ] I've picked the right materialization for the affected models.
# Other
- [ ] Check if a full-refresh is required after this PR is merged.
Revert "Relative increments with sign"
Reverted commit `f5718ded`.
# Description
Changes:
* Adds Revenue Churn Rate in YTD/MTD Overview. This has several implications, I finally understood how to properly compute a YTD.
The problem is that Revenue Churn Rate is a % of the Total "Churned" Revenue in a 12 m period vs. the Total Revenue in the same 12 m period. This is a bit tricky because it's not really additive, because of the Churn definition. Total Churned Revenue is the Revenue that the churned deals in a month generated on that past 12 months prior to churning.
So - in order to aggregate it properly, we need to do the sum of the Total Churned Revenue and retrieve the Total Revenue on these 12 months, and THEN compute the Churn rate.
This PR mainly retrieves the necessary input from the Churn models and then follows a similar computation as for the rest of YTD/MTD converted metrics.
I'll handle Onboarding MRR in a separated PR as this one is quite dense already.
# 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.
- [X] I have checked for DRY opportunities with other models and docs.
- [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: #27609, #27805
# Description
Removed `id_accommodation` from athena models
I remembered why I had add it before, it was to unite it with the edeposit model with the same number of fields but I fixed it already
# 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.
- [ ] I've picked the right materialization for the affected models.
# Other
- [ ] Check if a full-refresh is required after this PR is merged.
Update athena models to remove id_accommodation
Related work items: #27938
# Description
Reduce metric name length to fit in PBI
# 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.
- [ ] I've picked the right materialization for the affected models.
# Other
- [ ] Check if a full-refresh is required after this PR is merged.
Reduce metric name length to fit in PBI
Related work items: #27609
# Description
Changes:
* Adds template for targets -> These need to be revised
* Adds comparison of values vs. targets
* Allows for multi-year last date computation (allowing multiple years to be selected, not just the last one)
# 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.
- [X] I have checked for DRY opportunities with other models and docs.
- [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: #27609, #27805
# Description
Adapts YTD/MTD for visualisation purposes
# 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.
- [X] I have checked for DRY opportunities with other models and docs.
- [X] I've picked the right materialization for the affected models.
# Other
- [ ] Check if a full-refresh is required after this PR is merged.
Visualisation changes
Related work items: #27609
# Description
First version of the model int_ytd_mtd_aggregated_main_metrics_overview
It aggregates data at the level of:
* Date
* Dimension
* Dimension Value
* Metric Name (or Id Metric)
This computes differences and relative differences in:
* current MTD vs previous month EOM
* current MTD vs previous year MTD
* current year YTD vs previous year YTD
This is applied to 13 metrics, namely:
* Total Revenue
* Revenue Retained Post-Resolutions
* Guest Revenue
* Invoiced Operator Revenue
* Invoiced APIs Revenue
* Billable Bookings
* Live Deals
* New Deals
* Churning Deals
* Damage Waiver Payout Rate
* Host Resolutions Payout Rate
* Invoiced Operator Revenue per Billable Booking
* Waiver Revenue per Billable Booking
This still does not handle:
* Comparison vs. targets (to be done later, need input)
* Missing metrics on Onboarding MRR / Revenue Churn (to be done later)
* Invoicing dependant data exclusion (to be done in reporting)
# 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.
- [X] I have checked for DRY opportunities with other models and docs.
- [X] I've picked the right materialization for the affected models. **At the moment I keep everything as views and seems ok. We'll see if this needs to be materialised as tables later on**
# Other
- [ ] Check if a full-refresh is required after this PR is merged.
Related work items: #27609
# Description
New model for the purpose of Main KPIs - Overview.
It computes:
* Current MTD value
* Previous Month EOM value
* Previous Year MTD value (12 months ago)
* Current YTD value
* Previous Year YTD value
for the metrics:
* Total Revenue
* Revenue Retained Post-Resolutions
* Guest Revenue
* Invoiced Operator Revenue
* Invoiced APIs Revenue
* Host Resolutions Amount Paid
* Damage Host Waiver Payments
* Billable Bookings
* New Deals
* Churning Deals
* Live Deals (dedicated logic handling)
* Waiver Revenue
Additionally it properly computes the following derived metrics:
* Waiver Payout Rate
* Resolutions Payout Rate
* Operator Revenue per Billable Booking
* Waiver Revenue per Billable Booking
Only for dimension = 'global', though should be easy to extend to other dimensions.
This does not handle (yet) Onboarding MRR nor Revenue Churn Rate, mostly because I need to think how this can be properly attributed in a YTD basis.
This does not compute variations (abs. diff. or rel. diff.) yet.
This does not handle the "hiding" of invoicing metrics due to the invoicing cycle yet.
This does not handle targets... yet!
# 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.
- [X] I have checked for DRY opportunities with other models and docs.
- [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: #27609
# Description
Changes:
* On Est. Billable Bookings, keeps previous logic for Old Dash while assuming the first billable service for New Dash
* Adds a Billable Check Out Bookings in Check Out bookings, as defined in the requirements.
# 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.
- [X] I have checked for DRY opportunities with other models and docs.
- [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: #27619
