Performance and retention agencies
with up to 50 people who manage several brands at once and need a reporting layer they can put their name on. I build one dashboard per brand, and the agency delivers it to the client as part of its own service.
I build analytics infrastructure for e-commerce brands.
Your shop, marketplace and advertising data come together in one place, with dashboards and a detailed monthly analysis of where the next budget should go. Contracts follow EU practice, including GDPR-compliant data handling.
Revenue per €1 of ads, last 12 weeks€3.40+€1.30 since week 1
Goal €3 per €1Budget moved
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
- 12
This month’s analysisMove €4,000 a month from Meta to Google SearchExpected: +€6,000 in revenue a month
Shopify, Google Ads and Meta each report a different revenue for the same month, and none of them says where the money should move.
An analyst on the payroll takes up a large share of the ad budget, so the question stays unanswered and next month’s spend is split like last month’s.
Who I work with
Brands that sell direct to consumers
with €2–20 million in annual revenue, on Shopify, Amazon, eBay or regional marketplaces. At this size the standard tools stop showing what is profitable after ad spend.
What I offer
Data audit
€4902 weeks
A comprehensive review of your tracking, ad accounts and the reports your team relies on. The result is a prioritised list of what needs fixing, an estimate of what each issue costs you, and the order in which I would resolve them, with a price next to every item. We go through it together on a one-hour call.
Dashboard sprint
from €1,9003 to 4 weeks
A single dashboard built around the figures that run the business: revenue by channel net of ad spend, repeat purchase rate, contribution margin by product, and stock cover against current sales velocity. It is built on your own data, inside the BI tool of your choice, and delivered with documentation so your team can maintain and extend it on its own.
Data pipeline and dashboards
from €4,9006 to 8 weeks
When company data is spread across Shopify, Amazon, marketplaces, ad platforms, Klaviyo and a CRM that do not reconcile with each other, I bring it into a single data warehouse, build the dashboards on top of it, and document the entire setup in detail. The stack is AWS or Google Cloud with your existing BI tool, and the running cost for a company of this size is typically under €300 a month in cloud fees. At the end the infrastructure, the code and the credentials belong fully to you, with no licence fees payable to me and no lock-in.
Monthly analytics
from €900 a month
Ongoing analytical support once the dashboards are live. Every week you receive a short written analysis of what changed in the figures and why, and once a month we hold a one-hour review call to decide what to test or adjust next. Between those you have an analyst who answers questions within two working days. Agencies can run the whole arrangement under their own brand as part of their client retainer.
Stack
Infrastructure
AWS S3raw data storage
AWS AthenaSQL warehouse queries at scale
AWS GlueETL pipeline orchestration
AWS Lambdaserverless jobs for daily crawls and loads
Google BigQueryserverless analytics on Google Cloud
Apache Airflowscheduling and monitoring of data workflows
SQLdata modelling and transformation
Pythonautomated data validation and alerting
Business intelligence
Amazon QuickSightembedded dashboards on AWS
Power BImanagement reporting and financial models
Tableaudetailed visual analysis with drill-down
Looker Studiolightweight dashboards on Google data
Data sources
Shopifyorders, customers and products
Amazonmarketplace sales and digital-shelf tracking
Google Analytics 4on-site behaviour and conversion events
Google Adssearch, shopping and YouTube spend
Meta Adspaid social spend and campaign performance
TikTok Adsshort-video ad spend and sales
Klaviyoemail revenue and retention cohorts
Airbyteopen-source connectors into the warehouse
Your pipeline
Answer three questions, and the pipeline I would build for your company assembles itself, step by step.
What I would build
Package
02 Dashboard sprintfrom €1,9003 to 4 weeks
03 Data pipeline and dashboardsfrom €4,9006 to 8 weeks
- Your 4 sources are connected and refreshed every day.
- The dashboard shows revenue by channel net of ad spend.
- The running cost is typically under €300 a month in cloud fees.
I usually start with the Data audit: €490, 2 weeks.
Book a free 30-minute callData infrastructureAWSS3, Athena and GlueGoogle CloudBigQuery
AWSS3, Athena and Glue- Storage and queries are paid per use, so at your size the monthly bill stays small.
- It is the setup I run every day for the crawls of more than a hundred retail sites.
Google CloudBigQuery- Google Ads and GA4 flow into BigQuery through Google’s own free transfers, with no connector to maintain.
- The first terabyte of queries each month is free, which covers most of what a company of your size needs.
Direct connection
- Your sources are few enough for the BI tool to read them through its connectors, with no cloud setup of your own to pay for or maintain.
- If you later sell in more places, a warehouse can be added and the same dashboard connected to it.
BI toolLooker StudioPower BIAmazon QuickSight
Looker Studio- It is free and reads Google Ads, GA4 and BigQuery directly.
- Reports are shared as a link, so everyone sees the same figures.
Power BI- It is built for management reporting and financial models: margins, costs and stock in one place.
- Many finance teams already have it through Microsoft 365.
Amazon QuickSight- It runs inside AWS, next to the data, with nothing extra to connect.
- Dashboards can be embedded in a portal, under an agency’s own brand if needed.
How the first call goes
Book a free 30-minute call- Before the call
The call is free and takes 30 minutes, and there is nothing to prepare: bring any questions you already have about your figures.
- On the call
You tell me how your business works and where the data is, we open the reports together and I give my assessment as we go.
- By the end
You know the 2–3 biggest gaps in your setup, roughly what each of them costs you, and whether an audit is the right next step.
Daria Ermicev‑Apostol
Founder of ACCADA
From 2022 I was a senior data analyst in banking, building data infrastructure for the retail business and designing customer segmentation and scoring models. Part of my role was translating regulatory reporting requirements into data models that risk, product and engineering teams implemented together.
In 2026 I founded ACCADA, which today runs the data-quality monitoring on AWS for daily crawls of more than a hundred retail sites behind the pricing, digital-shelf and share-of-voice analytics of global consumer brands, alongside Google Ads budget-allocation analytics for a waste management company and digital infrastructure for a medical equipment manufacturer.
- Company
- ACCADA S.R.L. is registered in Chișinău and is a Moldova IT Park resident. It currently works with clients in France, Canada and Moldova and is taking on new analytics projects.
- Contracts
- Contracts follow EU practice, invoices are issued in euros, and clients inside the Union account for VAT under the reverse-charge rule.
- Communication
- ACCADA works in the European time zone, usually with a weekly call and written updates in between, in English, Romanian and Russian.
- White label
- For agencies, every deliverable can carry your branding and be delivered as a white-label service: the dashboards, the written analysis and the monthly review are presented to your client as part of your own offering.
Pick a time for the first call below.
Free call
30 minutesVideo call
The calendar needs JavaScript. You can book on calendly.com or write to daria@accada.io.
Loading the free times…
The calendar is not available right now. You can book on calendly.com or write to daria@accada.io.
There are no free times in the next four weeks. Write to daria@accada.io to agree a time.
That time has just been taken. Please choose another one.
The call is booked
To change the time or cancel, use the links in that email.