What your market says, counted — in their own words.
Every objection, request and question your team already hears, ranked by how often it comes up, with the calls behind each number.
- Fix
- Price doubt rose to 38% of first calls
- Say
- Answer it on the landing page, before the call
- Because
- “It’s a good program, but that’s above what we planned.”11:04
Someone reads a sample and writes a summary.
The summary is honest and it is one person’s reading of twenty calls out of four hundred. What ends up in it depends on which twenty, and on what that person already believed before opening them.
The survey has forty responses and the calls have four hundred
Surveys reach the people willing to fill in a survey. The conversation reaches everybody, and it happened anyway. What buyers said in it is more honest than what they wrote afterwards, because they were talking to a person and not to a form.
Your team ran the research already. Nobody read it back.
What changes
A count, not an impression
Thirty-eight percent is a number you can check. “Clients often say” is not.
Their wording, not your paraphrase
The page you write answers the objection in the words people actually use to raise it.
Product hears it too
Every request your team left open, counted and quoted, with the calls behind the number.
How it works on your conversations
Every screen below is the product itself, running. Try one.
Rank what buyers want, and what went unanswered
Every goal your market raised, counted across the quarter with the sentence behind each count. Alongside it sits the request nobody could answer: when a buyer asks for something you do not have, the rep says what reps say and the call moves on. Here it is counted rather than forgotten.What buyers asked for this quarter, ranked by how many of them asked.What buyers want this week
Buyer goals from the week’s calls, as a share of everyone who brought it up. This is what the ads should talk about.
«Half my team joined this year and every one of them learns by watching whoever is free.»
Most often this week: Getting new people productive faster (41%)
Requests left without an answer
Everything a buyer asked for and did not get, not only “we don’t have it”. Real product gaps are marked separately.
3× real product gap this week
Aug 12Product gap
[40:35] Buyer: «Can we get this in German for the Austrian team as well?»
The rep said they would check and the call moved on. Nobody checked.
Open conversation →
Aug 11Send documents
[20:37] Buyer: «Could you send over the slides from the presentation?»
The rep said they did not have them to hand. They were never sent.
Open conversation →
Aug 9Product gap
[26:04] Buyer: «Does it connect to the system we already use for tickets?»
The rep answered “not right now”, and the buyer did not ask again.
Open conversation →
Turn every objection into two numbers
How often buyers raised it, and how often anybody answered it properly. The gap between those numbers is the part of your market you are losing to a sentence nobody has written yet. Each row opens into how the objection sounded in the buyer’s own words.Raised this often, answered this often. The distance between them is the work.What stops your buyers
Every objection heard this quarter, and how often the team actually got past it.
113calls with at least one objection
40%worked out overall
«Before we go further, how does this work if we start mid-quarter?»
Forty per cent worked out means sixty per cent of interest quietly went nowhere.
Read the week in sentences
A number tells you something moved. The words tell you what to do. Each week arrives with the moment that went wrong and the moment that went well, quoted, with the wording that should have been used instead and why the good one worked.The failure and the win of the week, in the words they actually happened in.Verbatim from the calls of the week
Two failures and one win, pulled out of 218 conversations. Each opens into the recording it came from.
Failure of the week
The discount was offered without a date on it
[07:53] «You are currently getting 30% off the base price after the trial…»
Context The buyer had just reacted to the price. There was nothing in the sentence that made this week different from next month, so the conversation ended in "let me think".
What should have been said "The 30% holds for two days after the trial, so you can decide calmly and still keep it." A discount without a date is a discount without a decision.
Isaac Lane · Open call #108046 →
Failure of the week
Nothing was offered to send after the price talk
— never said —
Context After discussing cost and terms the buyer said "we have not decided yet". The rep agreed and moved to closing. No summary, no schedule, no written offer: the conversation left nothing behind it.
What should have been said "I will send the full program and the payment options in writing, so you can go through it with your team." The buyer leaves with something to reread instead of a memory.
Nora Bennett · Open call #108066 →
Win of the week
The rep quoted somebody the buyer already trusts
[01:16] «Going by the trainer’s own feedback, he actually did quite well…»
Context Right after the trial, before any talk of price. The rep repeated what an outside expert had said about the buyer’s own case, in that expert’s words.
Why it worked The reassurance came from someone with nothing to sell. The buyer heard confirmation of quality without being sold to, and the price conversation started from a different place.
Technique: Third-party proof, in the buyer’s own case
Chloe Barrett · Open call #108635 →
Give every finding an owner and a state
A finding somebody has to remember is a finding that expires. Each row here carries who it belongs to, whether anyone has started, and the count of conversations it rests on. Filter it to one team and you have their quarter, with the evidence attached to every line.What the reading found, split by owner, with the conversation count on every row.Marketing and product tasks
What the week collected: objections for marketing, gaps for product, unanswered requests for both. Ranked by how many conversations each one stands on.
7 of 7
Because, on the calls Half the buyers postpone without naming a reason. Nothing on the site says what waiting costs them, so the rep has to invent it live every time.
How it lands next to the research you already do
- Customer interviews
- An agency charges for a hundred interviews what your team runs in a week. Those conversations already happened and are already recorded.
- Surveys
- A survey asks people to remember what they thought. A recording holds what they said, unedited, to a person.
- Win and loss reviews
- The reason comes from the conversation rather than from a dropdown filled in after the quarter closed.
- Sample bias
- Reading every conversation removes the question of which ones got picked, and with it the argument about whether the finding is real.
- Segments
- Segments come from what buyers said on calls, not from a form somebody filled in at signup.
What happens, and when
Week one
A quarter of conversations is read at once, not from today forward.
Month one
The vocabulary your market uses is ranked, with the calls behind each phrase.
Quarter one
A new worry shows up the week it appears, not in the next research cycle.
Where to start
Each product is bought on its own. Start with the one that hurts.
Spellit Client Truth
Start with it!Spellit Actions
Spellit Academy
Spellit Notes
See the demo now. Then run it on yours.
A live demo of our products and a real conversation
about the growth problem you need to solve.
FAQ
What does Spellit count in our calls?
For voice of customer work, Spellit Client Truth counts what buyers raise in calls your team has already recorded: their goals, objections, questions and requests, plus the requests nobody answered. Each count is measured against the base that fits it. Objection shares come from calls that contained an objection, and question topics from calls that contained a question, so calls without that moment do not dilute the number.
The Client wants screen opens with the scale in plain view: how many conversations had questions, how many had requests and how many requests went unanswered. Different spellings of one objection are merged into one tag, so a single worry does not split into two smaller ones. When a slice rests on fewer than five calls, Client Truth marks it as thin data instead of presenting it as a confident share.
How does a number link back to the calls behind it?
Every count in Spellit Client Truth opens into the buyer's own words and the call they came from, so anyone can check a number in a few clicks. Each finding is tied to a word-for-word buyer line with two messages of context on either side, plus the call and the timestamp.
When one call has to stand for a pattern, Client Truth picks the call with the longest buyer line on that point, so the evidence is the buyer talking rather than the rep. From an objection, a request or a persona, you can open the call at the right moment. This helps most when someone doubts a finding. The team can play the conversation and skip the argument about impressions.
How do we know whether reps answer an objection well?
Spellit Client Truth shows the share of calls where an objection came up and reps handled it well, with the all-time figure next to the selected week. The gauge is green at 70% or more, yellow from 50% and red below that.
Below the gauge is a weekly breakdown of the top objections. For each one you get the buyer's situation, how reps handle it now, how buyers react, a recommended answer and two ready phrases for the script, and each item opens into a buyer quote. There is one condition. The handling gauge appears only if your call methodology scores how objections were handled. Without that score, Client Truth hides the block, because a row of zeros would wrongly suggest that objections are almost always handled.
How do findings turn into work for marketing and product?
Spellit Client Truth turns its voice of customer findings into a task list addressed to marketing or product and ranked by how many conversations back each item. The list is built from the data. The week's most frequent objections, up to six, become marketing tasks. The most frequent unanswered requests become standing tasks that stay until someone closes them, and requests about a specific product go to product while the rest go to marketing.
Priority follows the conversation count: high from five conversations, medium from two. Each task has a status (new, in progress, done or dismissed), a discussion thread and an example call. Nobody has to sign off on a task. The marketer or product manager takes it and marks it. Statuses also survive data rebuilds, because a task keeps the same identity when the numbers are recalculated.
What counts as an unanswered request, and when is it a product gap?
In Spellit Client Truth, an unanswered request is any moment when a buyer asked for something on a call and got no answer, and only a tagged part of that group counts as a real product gap. The wider group includes ordinary misses too, such as documents that were never sent, a promised callback or a technical problem.
That is why the screen labels the block "Requests without an answer" and marks product gaps separately with a red frame. These requests are counted across all time instead of week by week, because in a single week the number is nearly always zero and the signal gets lost. A gap stays on the list until someone closes it. For a product team, this keeps a real feature request apart from a sales follow-up that slipped.
How do we see what changed in buyer conversations this week?
Spellit Client Truth opens each week with what got worse since the week before, so a new buyer worry shows up in the week it appears. The customer voice screen starts with an AI summary of the week. It then shows what held buyers back, a few tasks for the week and weekly trend lines for objections, trust and lead scoring.
An alert card appears when a metric moves by two percentage points or more, with the previous value shown next to it. The screen closes with the loss and the win of the week, quoted word for word, each linked to its call. Problems come first on purpose. An earlier version that opened with good news was dropped, because it told teams everything was fine and gave them no reason to act.