It is Wednesday afternoon in a manufacturer’s office on the outskirts of a mid-sized European town. The rep has been at her desk since 7:30. The inbox shows 14 unanswered emails. Three are spec questions from buyers she quoted last week. Two are from a sales engineer asking whether the 5 by 4 louvred pergola needs the 100mm post or the 120mm. Four are price checks from the workshop because the steel rate sheet changed on Tuesday. The remaining five are weekend leads waiting on a first reply. Full days, honest effort, and not a single fresh quote has gone out since Monday.
This is not a careless rep. She is a senior with a decade in the role. The week looks like this because every artefact has to be translated from one format to another. The spec lives in the buyer’s head. The compatibility rules live in the engineer’s head. The current price lives in three spreadsheets last reconciled a week ago. The rep is the bridge between them, which means most of the week is translation and very little of it is selling. The Friday quote is real. It is also not the problem.
We thought we had a quoting problem. We bought software that promised faster quotes. The quotes got faster by maybe twenty percent. The week stayed broken. What fixed it was killing the Monday translation step, the Wednesday compatibility step, and the Thursday price step. By the time we got to the quote itself, there was nothing left to automate.
Salesforce State of Sales puts non-selling activity at roughly two thirds of a rep’s week, and the number has barely moved across editions. In customised-product lines, the bulk of that lives inside the quoting workflow, which is the polite way of saying it lives in the translation work on the way to the quote.
What 60 percent of a sales rep’s week really goes to#
A typical week breaks down like this. Monday, the rep processes weekend contact-form submissions. Each says some version of “interested in a pergola, please send pricing.” She calls eight buyers, reaches three, spends 20 to 40 minutes per call extracting the spec. That is the intake loop. It takes Monday and a chunk of Tuesday.
By Wednesday, two of those buyers are back on the phone because engineering flagged a compatibility question. The 5 by 4 roof needs the 100mm post, not the 80mm the buyer had in mind from last summer. Rep calls buyer, then engineer, then buyer again. That is the compatibility loop. Three or four hours.
Thursday is pricing reconciliation. The aluminium rate changed on Tuesday. The rep recalculates the three quotes drafted on Wednesday and emails the workshop to confirm. That is the price loop. Two to three hours, more during supplier-change windows.
Friday afternoon is the quote itself. Format the PDF, double-check the parts list, attach the drawing, send. Ten minutes per quote, eight quotes, 80 minutes total. By Friday the prior weekend’s leads have gone cold and a fresh batch is queuing for Monday.
Live buyer conversation with intent to close lands between 8 and 12 hours out of 40. The remaining 28 to 32 hours are translation work. The proportion is not a personal failing. It is the structure of the work.
Why “automate the quote” misses the real bottleneck#
Most vendors lead with quote-generation as the headline feature. Auto PDFs, branded templates, one-click pricing. The pitch lands because the quote is the visible artefact. The math, however, is unforgiving.
A quote that takes ten minutes manually, automated to ten seconds, saves ten minutes. On a rep producing eight quotes a week, that is 80 minutes saved. Selling time moves from 10 hours to 11 hours and 20 minutes. A three percent funnel bump. Real, but not the improvement the rep was promised.
The reason is that the quote is downstream. Everything upstream still has to happen. The spec still has to be gathered. The compatibility still has to be checked. The price still has to be reconciled. Faster quote generation saves ten minutes at the end. The upstream work that consumed the rest of the week sits untouched.
Hrovat sat with this honestly. The first pass put a configurator on the website and kept the manual workflow behind it. Quote generation got faster. The week stayed broken. The second pass moved the rule library out of the engineer’s head and into a configurable bill of materials wired to a pricing engine. Buyers’ choices arrived already structured, already priced. The 10,835 hours saved came from killing the upstream loops, not from a prettier PDF.
The 3 translation loops worth killing first#
Loop 1: spec intake. The Monday ritual of turning a contact form into a written spec. Every call runs 20 to 40 minutes. The questions are the same every time. The fix is a configurator on the public website where the buyer designs the product themselves. The configured spec lands in the CRM where a contact-form lead used to. The intake call disappears. YourPergola ran the configurator alongside the rep on every call and recovered 166 hours across the team over four months.
Loop 2: compatibility clarification. The Wednesday ritual of explaining why the 5 by 4 roof needs the 100mm post. The rep is the human cross-reference between the buyer’s wish and the engineer’s structural rules. The fix is the same configurator, wired to the compatibility rules. The 80mm post is unavailable in the menu the moment a 5 by 4 roof is configured, because the rule has already filtered the option. The deeper walkthrough lives in manual quoting errors.
Loop 3: pricing reconciliation. The Thursday ritual of checking whether Tuesday’s price still holds. The fix is a single pricing engine, not a spreadsheet. The aluminium rate changes Tuesday morning, the new rate flows to every open quote by Tuesday afternoon, with no human in the loop. The buyer sees the current price every time, and the rep sees a clean record that does not need a Thursday audit.
Three loops, six hours per quote, eight quotes a week. That is the 60 percent.
- 10,835 hours saved Hrovat (Slovenia, modular houses); €133M in quotes processed after the translation step was removed.
- +40% revenue in 3 months Armat (Slovenia, carports); 1,873 inquiries worth €12.3M, no new sales hires.
- 166 hours saved over four months YourPergola (Hungary, louvred pergolas); 331 quotes generated across four months with the same team.
- ~60% of the rep's week Translation work in customised-product lines, before the upstream loops are automated.
Named numbers from public case studies. The 60 percent figure is from internal diary studies plus Salesforce State of Sales.
What changes when the translation step is gone#
The before and after shows up in three places.
The rep’s day. Before, she arrives at 7:30 and starts intake calls. After, she opens an inbox of configured PDFs. A buyer who spent 12 minutes designing a 4 by 5 in anthracite with motorised screens is not the same buyer as one who typed “interested, send pricing.” She skips the intake call and goes straight to the close. The Wednesday clarifications never start. The Thursday reconciliation never happens. The Friday PDF collapses to the moment the buyer hits submit.
The team’s pipeline. Before, the pipeline grows as leads come in and shrinks as the team works through the translation backlog. The two flows run even and the team runs to stand still. After, the pipeline grows continuously because translation is no longer the rate-limiter. Armat lifted revenue 40 percent in three months with no new sales hires, because the team finally had time to close the buyers who used to ghost on the Monday queue.
The headcount conversation. Before, the team debates whether to hire a third rep because the current two cannot keep up. After, the question becomes whether the existing team can absorb the next product line without growing. The configurator does not pay for itself by reducing labour cost. It pays for itself by lifting the lid on volume, which is a much larger number. The structural argument for who needs which kind of tool sits in sales configurator vs CPQ.
The line to take home is short. The Friday quote is real, and it is also the smallest part of the cost. The 60 percent of the week that goes missing is upstream of the quote, in the three translation loops a person runs because the rule library lives in distributed human memory. Move the library into software, kill the loops, and the week reshapes around closing instead of translating. The named numbers sit in Hrovat, YourPergola, and Armat.
People also ask.
How much time does the average sales rep spend on quoting?
Industry research from Salesforce State of Sales puts non-selling activity at roughly two thirds of a rep's week. In customised-product lines we have audited, the quoting and pre-quoting block alone runs 50 to 60 percent of working hours. The quote itself takes ten minutes. The work around the quote, the spec clarification calls, the back-and-forth with engineering, the price checks against the spreadsheet, takes the remaining six hours per quote on average.
What counts as the translation step in a quote?
Every step where a person turns an unstructured buyer signal into a structured artefact that another system can read. The intake call that turns a contact form into a written spec. The engineering note that turns the spec into a parts list. The price check that turns the parts list into a number. The PDF that turns the number into a document the buyer recognises. Each step is a translation, each one is performed by a person, and each one can be done by software once the rule library exists.
Why does faster quote generation alone not fix this?
Because the quote is the last step, not the bottleneck. If the rep still spends three days collecting specs, clarifying compatibility, checking prices, and chasing engineering before the quote is generated, shaving the final step from ten minutes to ten seconds saves ten minutes. The compounding savings only land when the upstream translation loops are automated too, so the buyer's choices arrive already structured and already priced when the rep opens Monday's inbox.
What are the three translation loops to automate first?
Spec intake (replace the contact-form-plus-discovery-call with a configurator the buyer drives), pricing reconciliation (move the pricing rules from spreadsheets into a single live engine), and engineering hand-off (resolve the bill of materials automatically when the buyer hits submit). These three account for most of the 60 percent that disappears. The fourth, document generation, is the easy win that vendors lead with, but it is the smallest of the four.
Does automation reduce headcount on the sales team?
In the customised-product lines we have watched, no. Armat added 40 percent of revenue in three months with the same team. Hrovat processed €133M in quotes with the team they had. What changes is what the rep does with the recovered hours. Closing calls instead of spec intake. Field visits instead of price reconciliation. The bottleneck the configurator removes is not labour. It is the translation step between the buyer and the build, which used to consume the rep's day.
How do I measure rep selling time versus admin time?
A two-week diary study is the cheapest instrument. Have every rep log 30-minute blocks against four categories: live buyer conversation, quote preparation, spec clarification, internal coordination. The selling-time number is the first category alone. In manual operations the first category usually sits between 25 and 40 percent. The remaining 60 to 75 percent is where the leak is. Repeat the study 90 days after automation to see where the hours moved.
What is the quickest 30-day win for rep productivity?
Replace the spec-intake step. The configurator on the public website, even before any pricing or PDF wiring is in place, removes the longest leg of the translation chain. Buyers arrive with a configured spec instead of a contact form. The rep skips the discovery call and moves straight to the close. Inside the first month, the typical reduction is two to three hours per rep per day, which compounds before the rest of the chain is automated. The pricing engine and the PDF wiring follow in months two and three.
