Automatic sales report without the Friday spreadsheet
A workflow pulls every deal out of the CRM overnight and counts what a board actually reads: pipeline by stage, conversion by rep and by lead source, a forecast weighted by probability for this month and the next, and a list of deals with no movement. The report is in the inbox at 7:00 a.m., or it lives as a dashboard. Nobody retypes columns into Excel.
- 4 viewsin one report: pipeline by stage, conversion by rep and by lead source, a forecast weighted by probability, and deals with no movement
- 14 daysa sample threshold after which a deal with no activity lands on the at-risk list; you set it in the workshop and change it yourself later
- 2 to 4 weeksfrom the workshop to a report that runs alongside your own spreadsheet and matches it number by number before anything is switched off
- 7:00 a.m.when the report is in the board inbox, daily or every Monday, no matter who is on holiday that week
The problem
Why is Friday's report already out of date on Monday?
Friday afternoon. Somebody opens the CRM, copies the deals into a spreadsheet column by column, adds up the totals per stage and emails the file to the board. By Monday two of those deals are closed, one is lost, and three new ones do not exist in the file at all. The board takes decisions on a picture from before the weekend.
The second problem is worse than being late. Some reps update their deals once a week, or not at all, so the pipeline total in the spreadsheet and the total in the CRM disagree, and nobody knows which one is true. The report shows the sum of deal values, not the sum weighted by probability of closing. It also says nothing about the deals that have had no activity for 14 days and are dead, only nobody has marked them yet.
The cost sits in three places. The hours of the person who assembles the file every week instead of selling. Hiring and budget decisions taken on a total that was never weighted. And the deals that died quietly, because in a spreadsheet they looked exactly like the live ones.
- One person loses hours every week copying the CRM into a spreadsheet
- The board decides on a sum of deals, not on a forecast weighted by probability
- A deal with no contact for 14 days looks the same in the file as a live one
- The total in Excel and the total in the CRM disagree, and nobody knows which to trust
Why now
What changed, so that the report can count itself?
Every CRM a B2B team runs on hands over the full list of deals through an API, with stage, value, probability of closing and the date of the last activity. HubSpot documents it in its developer reference, and Pipedrive, Salesforce and LiveSpace do the same. Nothing has to be exported by hand. A workflow in n8n or Make pulls that data overnight, at an hour when nobody is working in the CRM.
Where the report is calculated follows the scale. With a handful of reps and a few hundred deals, Google Sheets is enough, and the board already knows it. With thousands of deals, several years of history, or the CRM joined to invoices and ad spend, BigQuery earns its place. Either way the report goes out by email at a set hour, or lives as a dashboard that refreshes on its own.
The third change is the AI agent. The report answers the questions we anticipated in the workshop. The ones we did not anticipate, like which DACH deals have been sitting longest without contact, or who converts worst on leads from Google Ads, go to BOND, our company AI agent. Plain sentence in, same data behind it, answer in 5 to 10 seconds.
What changes
What do the board and the sales director get?
The report is there before your day starts
The workflow runs overnight and at 7:00 a.m. sends a finished report to the board and the sales director. Daily, every Monday or on the first of the month, whatever you settle in the workshop.
Same hour, same format, with nobody who has to remember to do it
A weighted forecast instead of a sum of deals
Every deal enters the forecast with the probability attached to its stage and with its close date. The board sees two numbers side by side: what is in the pipeline, and how much of it realistically lands this month and next.
The raw sum and the weighted sum on one line, so the gap between them is visible at a glance
Dead deals show up before they vanish
A deal with no activity for longer than the agreed threshold, 14 days for example, moves onto a separate list with the rep's name and the last note. You rescue it this week or you close it as lost, but you do it on purpose.
Threshold in days set in the workshop, the sales manager changes it without us
Conversion by rep and by lead source
How many deals from the contact form, from referrals and from Google Ads end in a signed contract, and with whom. The same stage definitions for the whole team, so the comparison is fair and the performance conversation runs on numbers instead of impressions.
Conversion counted from the same CRM fields for every person and every source
Gaps in the CRM shown as gaps
The report says how many deals have an empty value field, no close date or a stale stage. Instead of quietly pulling the total down, it names what is missing and where. CRM hygiene improves, because for the first time you can see where it is absent.
A separate section of the report: count of deals with empty fields, per rep
Sales questions asked in a plain sentence
What is the forecast for this month, which deals have been stuck longest, who closes worst on paid leads. BOND, our company AI agent, answers from the same data, with no login to the CRM and no waiting for the next report.
An answer in 5 to 10 seconds, in English or in Polish
And in your company
What if our CRM looks nothing like that?
Four sentences we hear most often on a consultation, and what we answer to each of them.
If you are after a neighbouring piece, there are separate pages for the automatic weekly report, for the automatic financial report, and for reporting automation as a whole.
HubSpot and Pipedrive have their own reports, why add anything?
If the native dashboard shows what the board needs, we say so on the consultation and that is the end of it. The difference appears in three places: when you want to compare the forecast against invoices, when you need history, meaning what the pipeline looked like 4 weeks ago rather than only today, and when the report has to reach the board inbox on its own at a fixed hour.
Our reps do not update the CRM, the report will count rubbish.
That is exactly why the report shows gaps as gaps: how many deals have an empty value field, no close date, zero activity for 14 days, and whose they are. It is the first effect clients notice after a rollout: CRM hygiene improves because it is measured for the first time. In the workshop we agree the minimum set of fields without which a deal does not enter the forecast.
We have no BigQuery and nobody who works with data.
With a handful of reps and a few hundred deals the report runs in Google Sheets, which you already know. BigQuery only comes in at thousands of deals, several years of history, or when you join the CRM to invoices and ad spend. In both cases the sales manager changes the day threshold, the recipient list and the send time alone, with no developer and without calling us.
The board wants a spreadsheet, not a dashboard.
Then they get a spreadsheet. The report is a file or an email at a chosen hour, and the dashboard is an extra for the people who prefer to click. In the workshop we check which numbers the board actually reads in today's file, and those go on the first page. The rest we cut, because a report nobody reads looks exactly like a successful project.
- CRM
- HubSpot, Pipedrive, Salesforce, LiveSpace
- Where the report is calculated
- Google Sheets, BigQuery, SQL databases
- Delivery
- Email in Google Workspace or Microsoft 365, Slack, a live dashboard
- Supporting sources
- Fakturownia, iFirma, Comarch Optima or sevDesk to compare the forecast against invoices, GA4 for lead source
- Automation layer
- n8n, Make, custom integrations over the API
- AI agent
- BOND, the company AI agent, sales questions in a plain sentence
What the report counts and where the data comes from
The views that most often make up a sales report. Which of them go into the first version is settled in the workshop, because a board usually reads three or four numbers, not everything that can be counted.
| View | Where the data comes from | Who uses it and what for |
|---|---|---|
| Pipeline by stage | value and stage of every open deal in the CRM | sales director: where in the funnel the jam forms |
| Weighted forecast | deal value times the stage probability, plus the close date | board: how much realistically lands this month and next |
| Conversion by rep | deals won against all deals closed, per person | manager: a performance conversation on numbers, coaching instead of a league table |
| Conversion by lead source | the source field in the CRM, or utm parameters from GA4 | marketing: which source produces contracts and which only produces deals |
| Deals with no movement | date of last activity against a threshold in days, 14 for example | rep: what to rescue this week |
| Data hygiene | empty value, close date and stage fields | manager: who is not filling in the CRM and how far it pulls the forecast down |
| Forecast against actuals | invoices from Fakturownia, iFirma or Optima against last month's forecast | board: whether the forecast can be trusted at all |
The rows are a starting point. A first version usually covers three or four views, and we add the rest after 30 days, once it is clear what the board actually opens.
How we do it
Five steps from a Friday spreadsheet to a report in production
Each step ends with something you can judge before you decide on the next one.
Workshop
We take today's report apart: who assembles it, how long it takes, which numbers the board reads and which it ignores. We go through the CRM with the sales team: whether stages are used consistently, which fields are genuinely filled in and which are empty.
Report design
Definitions of each view, the thresholds such as the number of days without activity, and the minimum set of fields without which a deal does not enter the forecast. The tool choice: n8n or Make to pull the data, Google Sheets or BigQuery to count it. We settle who receives the report, when, and in what form.
Rollout in parallel
The report runs next to your spreadsheet with nothing switched off. We compare pipeline and conversion number by number against what the CRM shows by hand, and explain every difference. The first version runs in parallel about 14 days after the workshop.
Switch and training
Once it checks out, the Friday spreadsheet stops being made. We show the board and the reps where every number comes from, so nobody treats the report as a black box. The sales manager learns to change the day threshold, the recipients and the send time alone.
Observation
For the first 30 days we watch what the board actually opens, whether the day threshold is realistic for your sales cycle, and whether last month's forecast matched the invoices. We add views to what is being read, not to what we predicted.
Before and after
How the week changes for a sales director
Before
- On Friday afternoon somebody on the team copies the CRM into Excel instead of making calls
- At the board meeting they defend a pipeline total they are not sure of themselves
- They hear about a dead deal when the client signs with a competitor
- Asked how much lands this month, they answer with an impression, not a number
After
- The report is in the inbox at 7:00 a.m., same format, same hour, nobody has to ask
- The board sees the raw sum and the weighted sum side by side, so the conversation is about the forecast
- The list of deals with no movement for 14 days arrives with the rep's name and the last note
- They change the day threshold and the recipients themselves, in one place
The first parallel report is visible about 14 days after the workshop. Thresholds and the set of numbers are reviewed after 30 days, and the forecast against invoices after the first full month.
For the person signing this off
What a sales director gets out of it
How much lands this month stops being a question aimed at you, because the answer is already in the report the board read this morning. You no longer defend a total you are unsure of. You defend a weighted forecast you can open, pull out any single deal and show why it is in there.
The second thing matters more in a conversation about hiring and budget. You know how many deals die after 14 days without contact and whose they are, and which lead source produces contracts rather than meetings. That is an argument for a new rep and for an ad budget that you do not have today.
- A weighted forecast instead of a total that needs explaining
- Stalled deals caught after 14 days, not after they are lost
- The report runs in parallel, so the old spreadsheet works until the tests are done
- You change thresholds and recipients yourself, with no request to us
What it costs
The workshop has a published price, the rollout is quoted after it
The process workshop costs EUR 3,500 net per week. For a sales report one week is usually enough, because the process is narrow: from a deal in the CRM to a number in the board inbox. We quote the rollout after the workshop, once it is clear how many sources are involved, what state the CRM is in, and whether the report has to reach invoices as well.
You pay the cost of running the workflow directly to the tool vendor, meaning n8n, Make or BigQuery, or you pay nothing at all when the report is calculated in Google Sheets and n8n runs on your own infrastructure. Where the AI agent is part of the scope, model queries cost the vendor EUR 0.005 to 0.07 each. No commission from us on any of it.
Process workshop
EUR 3,500 net
per week, on a sales report usually one week
A self-contained piece of work. It ends with written definitions of every view, the thresholds and the CRM fields, plus a plan you can execute without us.
- Map of today's report: who, how many hours, which numbers get read
- Stage, metric and threshold definitions written down with the sales team
- Tool recommendation: the native CRM report, Google Sheets or BigQuery
- List of gaps in your CRM today: empty fields, stages used inconsistently
- The minimum set of CRM fields and the metrics to measure after 30 days
- Most popular
Workshop and report rollout
Quoted after the workshop
scope follows from the number of sources and views
The option most companies choose. A working report within 2 to 4 weeks of the workshop, checked number by number against your CRM.
- Everything in the workshop
- A workflow in n8n or Make wired into the CRM, counting in Google Sheets or BigQuery
- Pipeline by stage, weighted forecast, conversion, stalled deals, data hygiene
- Pipeline history: a nightly snapshot, so you can see what the funnel looked like 4 weeks ago
- Delivery by email at a chosen hour, or a live dashboard
- Parallel rollout, the switch, and training the board on reading the numbers
Report and an AI agent for sales
Quoted after the workshop
for boards that ask about sales more than once a day
Everything in the rollout plus BOND, the company AI agent, answering pipeline questions in a plain sentence. An agent rollout takes about 2 months.
- Everything in the report rollout
- An AI agent wired into the CRM and the report: forecast, stalled deals, conversion by source
- Alerts on deals past the day threshold and on the forecast drifting from invoices
- Access for the board, the manager and the reps, in English and in Polish
- An answer in 5 to 10 seconds, with no login to the CRM and no waiting for a report
If the workshop shows that the native report in your CRM is enough, we say so and show you how to set it up. The workshop still ends with written definitions of stages, metrics and thresholds that stand on their own, with no rollout at all.
Frequently asked questions
What is an automatic sales report?
It is a workflow that pulls every deal out of the CRM through an API overnight, with its stage, value, probability and date of last activity, and counts pipeline by stage, conversion by rep and by lead source, a weighted forecast and a list of deals with no movement. The finished report goes out by email at a chosen hour or lives as a dashboard. Nobody copies data into a spreadsheet.
How long does an automatic sales report take to implement?
One week of workshop and 2 to 4 weeks of build with a single CRM and the counting done in Google Sheets. The first version runs alongside your own spreadsheet about 14 days after the workshop. When the report joins the CRM to invoices and ad spend in BigQuery, the build can stretch to 6 weeks.
Do we need BigQuery to have an automatic report?
No. With a handful of reps and a few hundred deals the report is calculated in Google Sheets, and that is our first recommendation. BigQuery makes sense at thousands of deals, several years of history, or when you join CRM data, invoicing and GA4 in one place. We make the call in the workshop, on your own numbers.
What if our reps do not update the CRM?
The report says it outright: how many deals have an empty value field, no close date or zero activity for 14 days, broken down by person. A deal without the minimum set of fields does not enter the forecast, so it cannot pull it down quietly. At our clients this is the first visible effect of a rollout: CRM hygiene improves because it is being measured for the first time.
How is the probability-weighted forecast calculated?
Every open deal enters the forecast as its value times the probability assigned to its stage, with the close date deciding which month it falls into. Probabilities come from the CRM, or we set them in the workshop from your history of won deals. After the first full month we compare the forecast against invoices and correct the weights if they have drifted.
Can the report join CRM data with invoices and ad spend?
Yes, and it is the most common reason a native CRM report stops being enough. Invoices from Fakturownia, iFirma, Comarch Optima or sevDesk let you check whether last month's forecast came true. Lead source from GA4 shows which channels produce contracts and which only produce deals. Joining several sources is usually the argument for BigQuery.
How is this different from the reports built into HubSpot or Pipedrive?
The native dashboard shows the state as of today, and if that is enough we will tell you so. An automatic report adds three things: history, meaning what the pipeline looked like 4 weeks ago, a comparison of the forecast against invoices from outside the CRM, and delivery to the board at a fixed hour with nobody logging in. On top of that the sales manager changes thresholds and recipients alone.
Is sales data safe outside the CRM?
We pull only the fields the calculation needs: stage, value, probability, dates, source and owner. Customer personal data stays in the CRM unless the board needs it in the report and we agree that in the workshop. The report is calculated in your own Google Workspace, Microsoft 365 or BigQuery project, and you control access to it yourselves. GDPR duties as controller stay with you, and the workshop writes down which fields leave the CRM and why.
Let us count how many hours the Friday report costs you
The free consultation takes 30 minutes. You tell us which CRM the team runs on, who assembles the report today and which numbers the board actually reads. We show you which views can be counted from your fields straight away and which need the CRM tidied first. You do not have to give access to any system at this stage.
We run at most two rollouts at a time, so that someone is actually sitting on each of them.
Book a free consultationPS. If somebody at your company opens the CRM and a spreadsheet side by side again this Friday, that is exactly this problem. A report a person assembles by hand is out of date on the day it is sent, and it cannot be automated until you write down what it is meant to count. That is where the workshop starts.