Stop guessing whether your customers are happy, and start hearing it in time.
Gully Sales builds a running feedback programme for your business: short surveys asked at the moments that matter, answers that reach a named owner quickly, and a review where what customers say turns into changes.
- A short survey your customers will actually answer, asked at the right moment.
- Every score and comment routed to a named owner, with the customer called back.
- One satisfaction number, with its response rate, that leadership reviews each quarter.
Gully Sales Private Limited works with businesses across India, and the programme is built to run on the tools your team already uses.
In one paragraph
What is Customer Satisfaction and NPS Programmes?
A customer feedback programme is the system that asks your customers how you are doing and then does something with the answer. Gully Sales designs the questions, chooses the moments to ask, builds the survey and the NPS or satisfaction reporting, routes every response to a named owner, and sets the review where findings become changes.
The problem
You only hear from the customers angry enough to say something.
Most businesses measure satisfaction through the loudest signals: a complaint, a payment held back, a renewal that quietly does not happen. The customers in the middle, mildly disappointed and still paying while they look around, say nothing at all. When a survey does go out, it goes to everybody at once, months after the experience it asks about, with fourteen questions and nobody waiting for the replies. A handful of people answer. The scores are read once, discussed briefly, and nothing inside the business is different the following month.
You will recognise it as
- You find out a customer was unhappy when they stop replying, not when the problem actually happened.
- A survey went out some time ago, the response rate was poor, and nobody has opened the file since.
- Sales, service and accounts all ask for feedback separately, so the same customer is surveyed three times.
- Nobody owns the answers, so a low score sits in a spreadsheet instead of reaching the person who can fix it.
- You can describe how customers feel with stories, but not with a number you would put in front of a board.
- Praise from happy customers stays inside an inbox instead of becoming a review, a reference or a referral.
What it costs the business
- Problems that one honest conversation would have settled become lost renewals, and you learn the reason only after the customer has gone.
- Improvement money goes to whatever the last angry customer complained about, not to what most of your customers quietly struggle with.
- Your team only ever hears criticism and never the praise, so the word feedback starts to feel like an accusation.
- Customers who would happily refer you are never asked, so growth keeps depending on new enquiries bought at full price.
- You cannot answer a large buyer, a lender or an investor who asks how satisfied your customers are.
Why it persists. Asking is easy; doing something with the answer is work. A survey can be sent in an afternoon, but a programme needs somebody to own each response, a rule for how quickly a poor score is called back, and a meeting where a repeated theme becomes a decision with a name against it. Most businesses stop at sending. The scores then carry no consequence, the team learns that nothing follows a bad one, and the next round is quietly dropped.
If it stays unresolved. You keep running the business on the opinions of the few customers loud enough to reach you. Quiet dissatisfaction stays unmeasured until it arrives as churn, and by then winning the account back costs far more than the question would have. Meanwhile the competitors who ask, listen and reply are building the reputation you had assumed was yours.
What changes
What changes once customers are asked properly and answered quickly.
In the first weeks
- A working survey reaches real customers, and answers start arriving within days rather than at the end of a project.
- You get a first satisfaction and NPS reading you can trust, with the response rate stated beside it.
- Every response has a named owner and an agreed time to respond, from the first week the programme runs.
In how the work runs
- Feedback stops being a project somebody remembers and becomes a step inside your normal customer workflow.
- A low score reaches the person who can act on it the same day, with the account history already attached.
- One programme collects feedback for the whole business, so customers are asked once and asked well.
- Recurring themes reach your management meeting as a standing item with owners and dates against them.
In sales and marketing
- Accounts at risk are spotted while they are still customers, and the recovery conversation happens in time.
- Satisfied customers are asked for a review, a reference or a referral at the moment they say they are happy.
- Improvement spending goes to the issue most customers raise, not to the complaint that shouted loudest.
In what management can see
- A one-page monthly report: score, trend, response rate, top themes and what was done about them.
- A named owner and a closure status against every single response, visible to anyone who asks.
Over the longer term
- A comparable score history, so you can see whether a change in service actually moved how customers feel.
- A store of customer language you can use in sales conversations, service training and marketing copy.
We control the design and running of the programme: the questions, the timing, the routing, the reporting and the review. Whether scores improve depends on your team acting on what customers say. We report responses honestly, including the uncomfortable ones, and never promise a score.
Who it is for
This is for you if customers are meant to buy more than once.
The businesses it suits
- Service, subscription and relationship-led businesses where renewals and repeat orders carry the revenue.
- Businesses with enough customers that anecdotes are no longer a reliable way to know how service is going.
- Owners who suspect quiet dissatisfaction but have no number to test the suspicion against.
- Teams already collecting feedback in scattered ways who want one programme instead of four half-used forms.
- Companies asked for satisfaction evidence by a large buyer, a lender, an investor or a certification body.
- Businesses whose next stage of growth depends on retention and referral rather than on more new enquiries.
What usually prompts the call
- A valuable customer left, and nobody inside the business saw it coming.
- A survey was sent, the response rate was poor, and the results were never used for anything.
- Somebody has asked you for an NPS or satisfaction figure and you had nothing to give them.
- Your service team has grown, and the experience now varies depending on who handled the account.
- You want reviews, references and referrals, but there is no reliable moment where anyone asks for them.
What Gully Sales does
The work, component by component.
Feedback strategy and metric choice
We agree what you actually need to learn, then pick the measure that answers it: NPS for the strength of the overall relationship, CSAT for a specific interaction such as a delivery or a service call, and a simple effort question where the complaint is that dealing with you is hard work. One primary measure is chosen, with a few supporting questions, so results stay comparable between periods.
- Why it matters:
- Businesses often collect three different measures inconsistently and end up unable to compare any period with another, or to say what a movement in the number means.
- You receive:
- A one-page feedback plan naming the primary metric, the supporting questions and what each is for.
- Business value:
- You get one number the leadership team can follow over time, instead of several that argue with each other.
Question design and survey build
We write a short question set in plain language your customers use, test it on a phone, and build it in a tool you already pay for wherever possible. The rating question comes first, an open comment second, and anything you cannot act on is cut. Where your customers are more comfortable in another language, the survey is written in that language rather than translated word by word.
- Why it matters:
- Length and jargon are the two biggest reasons surveys go unanswered, and a survey nobody answers gives you a number built on the opinions of a very small, unusual group.
- You receive:
- Live survey forms, mobile tested, plus the question set and an anonymised sample response.
- Business value:
- More customers reply, so the score describes your customer base rather than a handful of extreme opinions.
Moments of asking and sampling rules
We map where in your customer journey each question belongs: after a delivery, a service visit, an onboarding, a support ticket close, or on a relationship cycle for accounts that have no frequent event. Sampling rules then decide who is asked, how often, and who is deliberately left alone, so a single customer is never surveyed by three departments in the same month.
- Why it matters:
- Feedback asked at the wrong moment measures memory rather than experience, and asking too often trains customers to ignore you.
- You receive:
- A survey calendar and trigger map showing every asking moment and the rule that governs it.
- Business value:
- Feedback arrives while the experience is fresh and while there is still time to put something right.
Routing and closing the loop
Every response is routed the moment it arrives. A low score goes to a named owner with the account history attached and an agreed time to call the customer back; a strong score goes down a short thank-you route where a review, reference or referral can be asked for. Each response carries a status until somebody has closed it, so nothing is left sitting unread.
- Why it matters:
- The customer who told you something and heard nothing back is more likely to leave than the customer who never told you at all.
- You receive:
- Routing rules, a detractor callback script and a promoter thank-you route, with a closure status on each response.
- Business value:
- Feedback becomes a conversation with a customer rather than a row in a report nobody reads.
Scoring, reporting and trend analysis
We calculate the score the same way every period, report it alongside the response rate and the number of responses, and break it down where it matters: by segment, by branch, by product line, by the person or team that handled the account. The report is one page and says what moved, what it appears to be driven by, and what is being done about it.
- Why it matters:
- A score with no response rate beside it can be flattering nonsense, and a company-wide average can hide one branch or one product doing real damage.
- You receive:
- A monthly one-page report template plus the working file behind it, handed to your team.
- Business value:
- You can see where satisfaction is genuinely weak, so effort goes to the part of the business that needs it.
Theme analysis and the action review
Comments are read and grouped into recurring themes, with the volume and the revenue sitting behind each one. Those themes go to a scheduled review with your owners, where a small number are picked up, given a name and a date, and tracked to a conclusion. The next review opens with what happened to the last set before anything new is discussed.
- Why it matters:
- Individual comments feel like opinions and get argued away; the same comment counted forty times, with the revenue attached, becomes a decision.
- You receive:
- A theme log with owners and dates, and a review agenda that starts with the previous actions.
- Business value:
- Feedback changes something in the business, which is the only reason to collect it in the first place.
Handover, training and the calendar
We train the people who will run the programme after us: how the survey is sent, how a response is answered, how the report is produced and how the review is chaired. You get a written runbook and a calendar for the year, so the programme survives a busy quarter, a staff change and the month everybody is chasing targets.
- Why it matters:
- Feedback programmes usually die from neglect rather than disagreement, and the ones that survive have a named owner, a written routine and a date in the diary.
- You receive:
- A written runbook, a training session for the team, and a twelve-month asking and review calendar.
- Business value:
- The programme keeps running whether or not Gully Sales is still engaged.
What you will have at the end.
- A feedback plan naming the metric, the questions, the asking moments and the customers you will ask.
- Survey forms built in your existing tool, tested on a phone, in the language your customers actually use.
- A sample question set and an anonymised example response, so you see the output before anything goes live.
- A trigger map and survey calendar showing every asking moment and the rule that decides who is asked.
- Routing rules stating who receives which response, in what time, and what they are expected to do about it.
- A detractor callback script and a short promoter thank-you route for asking for reviews and referrals.
- A one-page monthly report template: score, trend, response rate, top themes and actions taken.
- A theme log grouping comments into recurring issues, with volume, revenue exposure and an owner against each.
- An action review agenda plus a record of what was committed, by whom, and what changed afterwards.
- A written runbook and a training session so your team can run the programme without us.
- A baseline report recording where satisfaction, response rate and closure stood before any change was made.
How it runs
The engagement, step by step.
- 1
Listen to what you already hold
Before asking a single new question we read what your customers have already told you: complaints, support tickets, online reviews, sales notes, cancellation reasons and any past survey. This tells us what themes exist, what language customers use and which parts of your journey are already suspect, so the survey asks about things that matter rather than things that are easy to ask.
- You provide:
- Access to complaints, tickets, reviews, past surveys and a list of customers lost in recent periods.
- We produce:
- A listening summary naming the existing themes and the gaps in what you currently know.
- Done when:
- You agree the themes we found match what you see, and the open questions worth measuring.
- 2
Choose the metric and design the questions
We decide which measure fits your business and each moment, then write the question set: one rating question, one open question, and only the extra fields you will genuinely use. Wording is tested with a few real customers and adjusted, because a question that reads clearly to your management team often reads very differently to the person who buys from you.
- You provide:
- An hour with the people who deal with customers, and a few customers willing to test the wording.
- We produce:
- The final question set for each moment, with the scoring method written down.
- Done when:
- The questions are signed off and the scoring method is agreed so results stay comparable.
- 3
Map the moments and the sample
We place each question at a point in your journey where the experience is fresh and something can still be done about the answer, then set the sampling rules: who is asked, how often, who is excluded, and what happens when two triggers fire for one customer in the same week. The result is a calendar rather than a campaign.
- You provide:
- Your customer list with segments, and the events your systems can reliably trigger from.
- We produce:
- A trigger map, sampling rules and the twelve-month asking calendar.
- Done when:
- Every asking moment has a trigger, an audience and a frequency limit agreed in writing.
- 4
Build, test and pilot
We build the surveys in your tools, test delivery on the channels your customers use, and run a pilot with a small group before anything goes out widely. The pilot checks that messages arrive, that the form works on an ordinary phone, that responses land where they should and that the report calculates correctly.
- You provide:
- Access to your CRM, helpdesk, email or messaging tool, and a pilot group of customers.
- We produce:
- Live surveys, working delivery on each channel, and a pilot result with the response rate.
- Done when:
- The pilot returns usable responses and the routing has been proven end to end.
- 5
Launch with routing and closure
The programme goes live across the agreed moments. Every response is routed to its owner, low scores trigger a callback within the agreed time, strong scores go down the thank-you route, and each response is tracked until it is closed. We sit with the owners through the first cycle so the habit forms while the work is still unfamiliar.
- You provide:
- Named owners for each response type and their time during the first cycle.
- We produce:
- Live routing, closure tracking and coaching for the owners through the first full cycle.
- Done when:
- Responses are being answered by your own people inside the agreed time.
- 6
Report, review and act
The monthly report is produced and the action review is chaired. Themes are ranked by how many customers raise them and what revenue sits behind them, a small number are picked up with owners and dates, and the following review opens by checking what happened to them. Nothing is added to the list until the previous items have been accounted for.
- You provide:
- Attendance from the people who own service, sales and operations, and decisions on what to change.
- We produce:
- The monthly report, the ranked theme log and a written record of every action agreed.
- Done when:
- At least one change has been made and traced back to the feedback that prompted it.
- 7
Hand over and set the rhythm
We write the runbook, train the people who will run each part, and put the year's asking and review dates in the calendar with owners against them. We also agree what should trigger a redesign later: a new product line, a new branch, a merger, or a score that stops moving because the questions have gone stale.
- You provide:
- The internal owner who will run the programme, and time for the training session.
- We produce:
- A written runbook, a trained internal owner and a twelve-month calendar.
- Done when:
- Your team runs a full cycle without us, and the review happens on its scheduled date.
Ways to work with us
Ways to work with us on customer feedback.
Feedback programme design
A short engagement producing the plan: metric, questions, asking moments, sampling rules, routing and the report template. Your team builds and runs it from there, with a review call once the first cycle has finished.
Design and launch
We design the programme, build the surveys in your tools, run the pilot, launch across the agreed moments and stay through the first full cycle, coaching the owners on callbacks, reporting and the action review before handing over.
Run and review
An ongoing arrangement where we operate collection, routing oversight and reporting, chair the action review on an agreed rhythm, and keep the question set current as your products and customers change.
Programme repair
For businesses already surveying customers with disappointing results. We look at response rate, timing, wording, routing and reporting, fix what is weak and leave what works, rather than starting the whole thing again.
Combined with retention work
Where feedback is being set up as part of a wider retention effort, the programme is scoped alongside health scoring, churn analysis or renewal work so one set of signals feeds every decision instead of three separate views.
Why Gully Sales
What you are actually choosing when you choose us.
We start with what your customers have already said.
Complaints, reviews, tickets and lost-customer reasons are read before a single question is written. It costs you nothing extra, it stops us asking about things you already know, and it means the first survey lands on the themes that matter.
Short questions, because long ones go unanswered.
We cut every question you cannot act on. A rating and a comment, answered on a phone in under a minute, will tell you more than a fourteen-question form that only your most patient customers finish.
Every answer gets an owner and a reply.
Routing and callbacks are designed before launch, not added later. A customer who takes the trouble to answer and then hears nothing is worse off than a customer who was never asked in the first place.
We report the response rate, not only the score.
A comfortable number from a small and unusual group of respondents is not evidence. Every report we produce states how many were asked, how many replied and who they were, so you know what the score can and cannot be used for.
We connect feedback to the rest of your revenue system.
Gully Sales works across marketing, sales, service and revenue operations, so a satisfaction theme can be traced into the sales promise that created it, and a happy customer can be routed to your referral and review work.
Built so it survives without us.
The runbook, the calendar and the training exist because a feedback programme that depends on an outside agency to remember it will stop the month that agency does. Your own owner runs a full cycle before we step back.
Where it applies
The same service, in different businesses.
Industrial and equipment suppliers
- The situation:
- Sales are made through dealers and direct accounts, and the only feedback that reaches the head office is a complaint about a delivery or a warranty claim.
- How it applies:
- A short survey after each delivery and each service visit, plus a relationship survey twice a year for key accounts, with low scores routed to the regional manager the same day.
- Likely benefit:
- Service problems at one branch or one dealer become visible early, instead of being explained away as a difficult customer.
Clinics and healthcare providers
- The situation:
- Patient experience is judged by whatever appears in online reviews, which arrive after the visit, in public, and usually from the extremes.
- How it applies:
- A brief post-visit question set covering waiting, explanation and staff courtesy, with a private route for a poor experience and a review request where the experience was good.
- Likely benefit:
- Problems are heard privately and answered, while satisfied patients are invited to say so publicly at the right moment.
Software and subscription businesses
- The situation:
- Renewal conversations are the first time anyone asks how the customer is finding the product, by which point the decision is often already made.
- How it applies:
- An onboarding satisfaction check, an effort question after support tickets, and a relationship survey timed well before the renewal window opens for each account.
- Likely benefit:
- Renewal discussions start with a known account sentiment rather than a hopeful call in the final week.
Professional services firms
- The situation:
- Client relationships depend on one or two partners, and nobody knows how a client felt about a project until the next brief goes elsewhere.
- How it applies:
- A short project-close survey for every engagement and a relationship survey for retained clients, reviewed by the partner group each quarter with actions recorded.
- Likely benefit:
- Delivery issues surface while the relationship can still be repaired, and strong feedback becomes references for new pitches.
Training institutes and education providers
- The situation:
- Feedback is collected on the last day of a batch, filed, and never compared between batches, trainers or centres.
- How it applies:
- A consistent question set at the same points in every batch, with results compared across trainers and centres, and themes taken to a monthly academic review.
- Likely benefit:
- Teaching and facility issues are seen at the batch level while the next batch can still be improved.
Facility, maintenance and field service companies
- The situation:
- Work is done at the customer site by technicians, and quality is only ever discussed when a customer escalates to the owner.
- How it applies:
- A single-question survey sent after each job close, aggregated by technician, route and client, with a callback rule for any low score before the next visit.
- Likely benefit:
- Contract renewals stop turning on the last bad visit, because problems are caught and answered job by job.
Questions buyers ask
Before you enquire, the answers you will want.
What information and internal involvement do you need from us?
We need your customer list with segments, access to complaints, tickets, reviews and any past survey, and the events your systems can trigger from. From your side it is a few hours in the design stage with the people who deal with customers, a named owner for each response type, and their attendance at the action review. Without owners the programme collects data and changes nothing, so that commitment matters more than any tool.
How long does it take before we have a number we can trust?
Design and build move quickly; the honest answer is that the number is only trustworthy once one full asking cycle has completed, and that cycle length depends on your business. A field service company asking after every job has readings in weeks. A firm with an annual relationship survey waits longer. We report early results with the response rate attached so you know how much weight they carry.
How is this different from a customer research or voice of customer study?
A research study is a deep one-off look: long interviews with a small number of customers, answering a specific question about your offer or market. A feedback programme is permanent infrastructure: short questions asked continuously at fixed moments, producing a comparable number and a flow of comments. They work well together, and the research usually tells us which themes the programme should keep watching.
NPS, CSAT or customer effort score: which one should we use?
NPS suits the overall relationship and is what outside parties usually ask for. CSAT suits a specific interaction, such as a delivery or a support call, and is easier for customers to answer honestly. An effort question is right when the complaint is that dealing with you is hard work. We usually pick one primary measure for the relationship and one for interactions, rather than collecting all three.
Our last survey got almost no responses. What would you do differently?
Usually four things: ask far fewer questions, ask much closer to the experience, send from a person your customer recognises rather than a generic address, and use the channel they already reply on, which in India is often WhatsApp rather than email. We also make sure the previous round was answered, because customers who were ignored last time have learned there is no point replying.
How is success measured on this engagement?
Against the baseline we record at the start. The programme measures are response rate, satisfaction and NPS movement, and how quickly responses are answered and closed. The business measures are renewal rate, churn, expansion revenue and the volume of reviews and referrals from satisfied customers. We separate the two honestly, because we control the asking and the routing while your team controls what changes.
What if the results are bad and the team takes it personally?
Assume the first read will be uncomfortable, and plan for it. Scores are introduced as a measure of the process rather than of individuals, results are shared with the team before they reach a board pack, and praise is circulated as carefully as criticism. Where results are reported by person, that is agreed openly in advance and used for coaching, not as a disciplinary instrument.
Do we need to buy a survey or feedback tool for this?
Usually not at the start. Most businesses already have something that can send a form and record replies inside their CRM, helpdesk or email tool, and we build there first. If volume, languages or routing genuinely outgrow what you have, we will say so and explain what a tool would add. Buying software is never the first step in this work.
4 more questions
Can surveys be run in languages other than English?
Yes, and it usually lifts the response rate noticeably. Questions are written in the language your customers are comfortable in rather than translated word for word, since a literal translation of a rating question often reads oddly. Comments come back in that language too, and are read and grouped into themes in the same way as the English ones.
Will this annoy our customers?
It does when everybody asks separately and nobody replies. The programme prevents that with sampling rules capping how often any customer is asked, one owner for all feedback across the business, and a rule that every response gets acknowledged. Customers rarely object to a short question after a real event. They object to long forms, repeated requests and silence afterwards.
What is excluded from the scope of this work?
We do not answer your customers for you, run your helpdesk, or make the operational changes that feedback calls for; those sit with your team and, where needed, with separate engagements. Market research into people who are not yet customers is a different service. We also do not remove or filter negative reviews, and we will not present a score in a way the response rate does not support.
How does this fit with customer health scoring and churn work?
Feedback is what customers tell you when asked. Health scoring is what their behaviour tells you without being asked, such as usage, payment or contact patterns. Churn analysis explains who left and why. The three are strongest together: a health score flags a quiet account, the survey asks it directly, and churn analysis checks afterwards whether the warning signs were read correctly.
Talk to us
Find out what your customers would tell you if somebody asked them.
Request a Customer Growth Assessment and we will look at how feedback reaches you today, what your existing data already shows, and whether a programme is worth building. It is a working conversation, not a sales pitch.
- No obligation and no sales script
- A reply from someone who does the work
- Your details are never sold or shared