You see a customer slipping before the orders stop, not months later.
Gully Sales builds the retention and reactivation campaigns that run on your own sales records: which customers are drifting, what triggers an outreach, what that outreach says, and which lapsed buyers deserve a win-back offer.
- Customers at risk get contacted while they are still customers.
- Lapsed buyers get a reason to return, not another newsletter.
- Repeat revenue is read by cohort, so you can see what the campaigns held.
Gully Sales Private Limited runs retention and win-back programmes for small and medium businesses across India.
In one paragraph
What is Customer Retention and Reactivation Campaigns?
Customer retention and reactivation campaigns are planned outreach to people who have already bought from you. Gully Sales groups your customers into cohorts, finds the behaviour that predicts a customer going quiet, and builds triggered save plays, a renewal cadence and win-back offers for lapsed buyers. Every campaign is read against repeat revenue, not against opens.
The problem
The customer who left never told you they were leaving.
A customer who ordered every six weeks now orders every ten. Then twelve. Nobody raises it, because each individual gap looks ordinary and your team is busy with the accounts that are shouting. Eight months later somebody notices the name is missing from the sales report and calls to ask what happened. By then the decision was made long ago, quietly, and usually for a small reason. This is not neglect. Your records hold the warning, but nobody has been given the job of reading them week by week, or a campaign to send once they do.
You will recognise it as
- You find out a customer has gone only when someone happens to notice the name missing from a report.
- The customer list keeps growing while monthly revenue stays flat, so new business is filling a hole you cannot see.
- Every promotion goes to the whole database, so a loyal customer and a lapsed one receive the same message.
- Renewals, refills and repeat orders are chased in the last week, by whoever remembers, over WhatsApp.
- Somebody says once a quarter that you should call the old customers, and the list is never opened.
- When a customer does come back, nobody can say which message or offer brought them.
What it costs the business
- You buy new customers to replace ones you could have kept, so the marketing budget rises while revenue holds still.
- Your revenue swings with acquisition alone, since the repeat base is not being worked as a source of its own.
- Discounts are handed out at the last moment to save an account, because the conversation began after the customer had already decided.
- Sales time goes to whoever complains loudest, while quietly profitable accounts drift towards a competitor unnoticed.
- Years of customer records sit unused, and the business cannot say what a customer is worth over their full life with you.
Why it persists. Retention has no owner. Marketing is measured on enquiries, sales on new orders this month, and service on complaints closed, so a customer who simply orders less belongs to nobody's number. The data that would show it sits split between an accounting system, a spreadsheet and someone's phone, so joining it feels like a project. Many owners also treat repeat business as a relationship matter that cannot be systemised, when the timing of a call is exactly what a system does well.
If it stays unresolved. The hole widens quietly. Acquisition cost keeps climbing because every new customer replaces a lost one instead of adding to the base, margins fall as last-minute discounts become the standard way to save an account, and the customers who leave take their word of mouth with them. Winning them back later costs more than keeping them would have, and some will not take the call at all.
What changes
What changes in the business when this work is done.
In the first weeks
- You can name, this week, which customers are behind their normal buying pattern and by how much.
- Every customer sits in a cohort with a defined value, order rhythm and risk level, in one place.
- Your team stops guessing who to call and works a list produced by the data.
In how the work runs
- Outreach is triggered by customer behaviour, so the contact happens on time rather than when someone remembers.
- Each save play has an owner, a script and a limit on what may be offered without approval.
- Renewals, refills and service dates are chased on a written cadence rather than in the final week.
In sales and marketing
- Revenue from existing customers becomes a line you plan and track, not a residue of acquisition.
- Lapsed customers become a defined pipeline with a stated value, worked in waves.
- Discounting narrows, because a conversation held early rarely needs the price to rescue it.
In what management can see
- Repeat purchase rate, order gap and revenue retained are reported by cohort every month.
- You can see which trigger, message and offer brought a customer back, and which did nothing.
Over the longer term
- The business learns which customers are worth acquiring, because it can finally see who stays.
- Retention becomes a habit inside the sales week rather than a campaign that runs once a year.
Gully Sales controls the cohort analysis, the risk model, the campaign design, the messages and offers, the tracking and the reporting. Whether a customer returns also depends on your product, service, pricing and why they left. We say plainly when the cause of loss is something no campaign can fix.
Who it is for
This is built for you if you already have customers worth keeping.
The businesses it suits
- Businesses with repeat, refill, renewal or subscription purchase, where a gap between orders means something.
- Distributors, dealers and manufacturers whose top accounts quietly reduce order size before they stop.
- Service firms with annual contracts, maintenance cycles or retainers coming up for renewal.
- Clinics, education providers and wellness businesses with a visit or course cycle that customers fall out of.
- Retail and D2C brands with a customer database large enough to split into cohorts.
- Businesses spending heavily on new enquiries while the existing customer base is never contacted.
What usually prompts the call
- Revenue is flat although the customer count is rising, so the base is leaking.
- A large account reduced its orders and nobody noticed until the quarterly review.
- You have several years of customer records and have never sent them a planned campaign.
- Renewal or refill dates are missed often enough that your team accepts it as normal.
- You are about to raise acquisition spend and want to plug the leak before you do.
- A competitor has begun approaching your customers directly.
What Gully Sales does
The work, component by component.
Cohort baseline
We rebuild your customer base from your own sales records and group it by when the customer first bought, what they buy, how often, how much and through which channel. Each cohort gets a normal order rhythm, an average value and a current status: active, slipping, dormant or lost.
- Why it matters:
- Retention cannot be discussed usefully at company level. Averages hide the fact that one cohort is holding steady while another has quietly halved.
- You receive:
- Cohort map with order rhythm, value and status for every customer, signed off by your sales lead.
- Business value:
- You stop arguing about whether retention is a problem and start looking at which group of customers it is a problem in.
Risk drivers
We look at the customers you have already lost and find what they did before leaving: orders stretching out, basket shrinking, a complaint left open, a contact person changing, payments slowing, a service visit missed. These become a small, written set of risk drivers with weights.
- Why it matters:
- Every business has its own tell. Guessing at industry averages produces a list of at-risk customers your team does not believe and will not call.
- You receive:
- Risk driver definition with weights and thresholds, derived from your own lost accounts.
- Business value:
- Your at-risk list is built from how your customers actually behave, so the sales team trusts it enough to work it.
Triggers
Each risk driver is turned into a trigger with a threshold and a timing rule: an order gap that passes one and a half times normal, a renewal date thirty days out, a first order not followed by a second within the usual window, a complaint closed without a follow-up call.
- Why it matters:
- The value of retention work is almost entirely in timing. The same message sent two months late is a different message.
- You receive:
- Trigger sheet listing every condition, its threshold, the channel it fires on and who owns the response.
- Business value:
- Outreach starts when the customer's behaviour changes, not when the monthly report is finally read.
Save plays
For each trigger we write what actually happens: who contacts the customer, on which channel, what is said, what question opens the conversation, what may be offered, what needs approval, and what is recorded afterwards. Plays are written for a phone call as often as for a message.
- Why it matters:
- A flagged customer with no play attached is a report nobody acts on. Teams need the words, the limits and the owner.
- You receive:
- Save play library with call scripts, message drafts, offer limits, escalation rules and logging steps.
- Business value:
- Any salesperson can run the save the same way, so the outcome no longer depends on who picked up the file.
Renewal cadence
For contracts, AMCs, subscriptions, refills and seasonal reorders we build a dated cadence that starts well before the due date, with a reminder sequence, a value recap of what the customer received, an owner for the conversation and a fallback if there is no response.
- Why it matters:
- Renewals lost in the last week are rarely lost on price. They are lost because the conversation started after the customer had already looked elsewhere.
- You receive:
- Renewal calendar and sequence per contract type, with owners, reminders and a written escalation path.
- Business value:
- Renewal conversations open early, on your terms, with the value already stated rather than a discount offered in panic.
Reactivation offer
Dormant and lost customers are ranked by past value and likelihood of return, then worked in waves. We design the reason to come back: a relevant new product, a changed service, a reorder made easy, a limited returning-customer term, or a straight conversation about what went wrong.
- Why it matters:
- A win-back that only says we miss you gets ignored. A lapsed customer needs a reason that answers whatever made them stop.
- You receive:
- Reactivation wave plan with ranked lists, offer variants by segment, message sets and a call list for high-value names.
- Business value:
- Your database turns into a working pipeline with a stated value, instead of a list everyone agrees somebody should call.
Impact tracking
Every campaign is tagged so that a returning customer, a renewed contract or a saved account can be traced to the trigger and message that produced it. Cohort revenue, repeat purchase rate, order gap and reactivation rate are reported monthly against the signed baseline.
- Why it matters:
- Retention spend is the first thing cut in a tight quarter, because most businesses cannot show what it held on to.
- You receive:
- Tracking setup, tagging convention and a monthly cohort report with revenue retained and revenue reactivated.
- Business value:
- You can defend the programme with numbers from your own records, and stop the parts of it that are not earning their place.
What you will have at the end.
- Cohort map of your full customer base with order rhythm, value, channel and current status per customer.
- Risk driver model built from your own lost accounts, with weights, thresholds and a plain-English explanation.
- Trigger sheet defining every condition that starts an outreach, its timing rule, channel and owner.
- Save play library: call scripts, email and WhatsApp drafts, offer limits, approval and escalation rules.
- Renewal and refill calendar by contract or product type, with a dated reminder sequence and named owners.
- Reactivation wave plan with ranked dormant lists, segment-wise offers and a separate call list for large accounts.
- Message and offer bank covering the whole cycle, written for email, WhatsApp, SMS and phone.
- At-risk dashboard or worksheet your team opens weekly, showing who to contact and why, in priority order.
- Tracking and tagging setup in your CRM or sheets so returns and saves can be attributed to a campaign.
- Monthly cohort report: repeat purchase rate, revenue retained, revenue reactivated, order gap and campaign return.
- A sample anonymised cohort extract and a filled at-risk worksheet, shown before you commit to the full build.
- Handover pack with the weekly running routine, owner list and a review checklist your team keeps using.
How it runs
The engagement, step by step.
- 1
Read the records
We take your sales history from wherever it lives, accounting software, CRM, spreadsheets or invoice exports, and rebuild a clean customer-by-customer purchase timeline. Missing fields, duplicates and merged accounts are cleaned before any analysis, and we tell you what the data cannot support.
- You provide:
- Sales and invoice history, customer master data, contract or renewal dates, and access to any CRM or tool in use.
- We produce:
- A cleaned purchase timeline, a data quality note listing gaps, and the definitions we will use for active, slipping, dormant and lost.
- Done when:
- Your sales lead agrees the definitions and confirms the customer list is recognisable.
- 2
Build the cohort baseline
Customers are grouped by first purchase period, product, channel and value band. For each cohort we establish normal order rhythm, average value, repeat rate and how many have already slipped. This becomes the signed baseline every later report is read against.
- You provide:
- Two working sessions to confirm which groupings match how you actually run the business.
- We produce:
- Cohort baseline document with current status counts and the revenue currently at risk in each group.
- Done when:
- The baseline is signed by you and stored as the reference for the engagement.
- 3
Find the risk drivers
We study the customers already lost and look for what they did in the months before. Order gaps, falling basket size, unresolved complaints, contact changes and payment behaviour are tested, and only the drivers that hold up in your data are kept and weighted.
- You provide:
- Access to service and complaint records if kept, and time with two or three salespeople who knew the lost accounts.
- We produce:
- Risk driver model with weights, thresholds and worked examples from named past accounts.
- Done when:
- Your team reviews a sample at-risk list and agrees the names on it look right.
- 4
Design triggers and plays
Each risk driver becomes a trigger with a threshold, a channel and an owner. Against each trigger we write the save play: the opening question, the script, the message drafts, what may be offered and what needs your approval before it is offered.
- You provide:
- Your commercial limits on discounts, credit and free service, and one reviewer for tone and claims.
- We produce:
- Trigger sheet and save play library, with every message drafted and approved before anything is sent.
- Done when:
- Triggers, scripts and offer limits are approved in writing by you.
- 5
Set the renewal cadence
Contracts, AMCs, refills and seasonal reorders are mapped onto a dated calendar. We define how early the first contact goes out, what it contains, how the value already delivered is recapped, and what happens at each stage if there is no reply.
- You provide:
- Contract and renewal dates, service history where available, and the person who will own renewal conversations.
- We produce:
- Renewal calendar, sequence and value recap template for each contract or product type.
- Done when:
- The next quarter of renewals is loaded with owners and dates against every one.
- 6
Run the reactivation waves
Dormant and lost customers are ranked and worked in waves rather than all at once, so offers and messages can be tested on a small group first. High-value names go to a phone call, the rest to sequenced messaging, and what we learn from each wave changes the next.
- You provide:
- Sales time for the call list, and a decision on what commercial terms a returning customer may be offered.
- We produce:
- Wave-wise lists, message and offer variants, a call brief per large account, and a result read after each wave.
- Done when:
- The first two waves are complete and their results are recorded against the baseline.
- 7
Track, report and hand over
Tagging and tracking are set up so a save or a return can be traced to its trigger and message. We report monthly by cohort, review what worked in a working session, and retire the plays that are not earning their place.
- You provide:
- One monthly review hour with sales and marketing, and confirmation of orders won so revenue can be attributed.
- We produce:
- Monthly cohort report, campaign-wise results, a revised trigger and play set, and the handover pack.
- Done when:
- Your team runs the weekly at-risk routine without us, and the reporting continues in your own records.
Ways to work with us
You can start with the analysis alone or run the whole programme with us.
Retention baseline
A one-time engagement: cohort baseline, risk drivers and a first at-risk list, with a written recommendation on which triggers and plays are worth building. Suited to a business that wants to see the size of the leak before committing to campaigns.
Campaign build
The full design and build: cohorts, triggers, save plays, renewal cadence, reactivation waves, message and offer bank, tracking setup and handover. Your team runs it afterwards, with our routine and reports in place.
Run and improve
We build the programme and then run it with you on a monthly cycle: the at-risk list, the waves, the message changes, the reporting and the review session, while your salespeople keep the customer conversations.
Reactivation sprint
A focused push on dormant and lost customers only, worked in waves over a short period, for a business with a large old database and a specific reason to reopen it, such as a new product or a service change.
Why Gully Sales
What you are actually choosing when you choose us.
We start from your records, not from a template.
The cohorts, risk drivers and thresholds come out of your own sales history. A retention model borrowed from another industry produces a list your salespeople quietly ignore, which is worse than no list.
Sales and marketing are handled together.
Retention breaks at the handover: marketing sends, sales calls, and neither owns the customer in between. Gully Sales works across both, so a trigger has a message and a named person attached to it.
Built for how Indian SMBs actually sell.
Much of your customer relationship lives on WhatsApp and the phone, and your history may sit in Tally or a spreadsheet. We design for that reality instead of insisting on a stack you do not have.
We say when a campaign is the wrong answer.
If customers are leaving over delivery delays, quality or a pricing decision, we say so and tell you what to fix. Sending a win-back offer into an unresolved problem costs goodwill you may not get back.
Everything is written down and handed over.
Triggers, scripts, offer limits, calendars and the weekly routine are documented in your own systems. If we stop working together, the programme keeps running and your team knows how to change it.
Reporting against a signed baseline.
The baseline is agreed and signed before any campaign runs, so every later number is read against the same starting point rather than against whatever makes the work look good.
Where it applies
The same service, in different businesses.
Industrial distribution
- The situation:
- Two hundred dealers and end customers reorder on their own rhythm. Three of the top twenty have stretched their order gap from six weeks to eleven, and no report shows it.
- How it applies:
- Cohort baseline by account size and product line, an order-gap trigger set at one and a half times normal, and a call play for the account manager with a defined credit and pricing limit.
- Likely benefit:
- Slipping accounts are called while they are still buying, and the conversation is about supply and stock rather than about a discount to win them back.
Multi-speciality clinic
- The situation:
- Patients complete one treatment cycle and do not return for review or the next stage. The front desk calls when it has time, which is rarely.
- How it applies:
- Visit-cycle cohorts, a trigger on the review date passing, a recall sequence over SMS and WhatsApp, and a call list for high-value treatment plans.
- Likely benefit:
- Recall happens on a schedule instead of on spare capacity, and the clinic can see how many patients returned from each recall wave.
Facility and equipment services
- The situation:
- Annual maintenance contracts are renewed in the final fortnight, often with a price concession, and a few lapse each year without anyone noticing until the site calls with a breakdown.
- How it applies:
- Renewal calendar starting ninety days out, a value recap of the service delivered during the year, staged reminders, and an escalation path for silent accounts.
- Likely benefit:
- Renewal conversations begin early with evidence of the value delivered, and lapsed contracts are caught before the site is uncovered.
D2C and retail brand
- The situation:
- Thousands of one-time buyers sit in the database. Every promotion goes to the whole list, so loyal customers and people who bought once two years ago receive the same message.
- How it applies:
- Cohorts by first purchase period and value, a second-purchase trigger inside the normal window, and reactivation waves with segment-wise reasons to return.
- Likely benefit:
- Repeat purchase rate is measured by cohort, and dormant buyers are worked as a ranked pipeline rather than as one undifferentiated list.
B2B software and services
- The situation:
- Subscriptions and retainers renew quietly until a client with falling usage and an unanswered support ticket gives notice at renewal, and the account team is caught unprepared.
- How it applies:
- Risk drivers built from usage, support history and contact changes, a trigger for accounts with two or more drivers active, and a save play owned by the account lead.
- Likely benefit:
- Accounts at risk surface a quarter before renewal, giving time for a real conversation rather than a last-minute concession.
Education and training
- The situation:
- Students finish one course and never enrol in the next, although the progression is the obvious step and the institute has their full contact history.
- How it applies:
- Cohorts by batch and course, a trigger at course completion, a progression sequence with counsellor calls for high-intent learners, and a reactivation wave for older batches.
- Likely benefit:
- Progression enrolment becomes a planned campaign with a measurable rate, instead of depending on which counsellor calls whom.
Questions buyers ask
Before you enquire, the answers you will want.
How early can you tell that a customer is about to stop buying?
It depends on how often they buy. For a customer ordering monthly, a gap stretching past about one and a half times their normal rhythm is usually visible within six to eight weeks. For quarterly or annual buyers the signal comes from other drivers: falling order value, an unresolved complaint, a changed contact person or slower payment. We set the thresholds from your own lost accounts, so the warning arrives while a conversation can still change the outcome.
How is this different from lead nurturing?
Lead nurturing works on enquiries who have never bought from you, moving a prospect towards a first order. This service starts after the first order. The people involved are customers, so we already know what they bought, how often and what they paid, and the campaigns use that history. The messages, offers and owners are different, and success is measured in repeat revenue rather than in new qualified leads.
How long does the engagement take?
The baseline and risk driver work typically comes first, then trigger and play design, then the campaigns going live. How quickly that happens depends mostly on the state of your customer records and how fast approvals move on messages and offers. We agree a stage-wise plan in the scope with dates you can hold us to, rather than a single delivery date fixed before we have seen your data.
What do we need to give you to start?
Sales or invoice history covering at least two of your buying cycles, a customer master list, contract or renewal dates if you have them, and access to any CRM, billing tool or sheets in use. Service and complaint records help but are not essential. Beyond data we need time with two or three salespeople who knew the customers you have lost, and one person who can approve messages and offer limits.
Our records are in Tally and a few spreadsheets. Is that enough?
Usually yes. Invoice-level history with a customer name, date, product and value is enough to build cohorts, order rhythms and most risk drivers. We clean duplicates and merged accounts as part of the work and tell you honestly what the data cannot support. If something important is missing, we say what to start recording now so the next quarter's analysis is stronger than this one.
How do you measure whether the campaigns worked?
Against the baseline your sales lead signs before anything runs. We report repeat purchase rate, save rate, reactivation rate, revenue retained and reactivated, renewal completion and campaign return, all by cohort. Campaigns are tagged so a return can be traced to the trigger and message that produced it. We also report what did not work, and retire plays that are not earning their place.
Will win-back discounts damage our pricing?
They can, which is why the offer is designed rather than improvised. We test reasons to return that are not price first: a new product, a changed service, an easier reorder, or a direct conversation about what went wrong. Where a commercial term is needed, it is bounded, time-limited and tied to a segment, and your approval limits are written into the play so nobody offers more under pressure.
Do you send the messages, or does our team?
Either, and it is set in the scope. Under a build engagement we design everything and your team sends and calls. Under a run engagement we operate the campaigns, lists and reporting each month while your salespeople keep the customer conversations, because a save call is better made by the person the customer knows. Phone calls to large accounts always stay with your team.
4 more questions
How many dormant customers should we try to bring back?
Not all of them at once. We rank dormant customers by past value, how recently they bought and how well they match what you sell today, then work them in waves. The first wave is small enough to test offers and messages properly. What that wave teaches changes the next one, which protects your database from being burned through on an untested approach.
Does this work for a B2B business with only fifty customers?
Yes, and often better, because each account matters enough to justify a proper conversation. With small numbers we lean less on statistical patterns and more on named account review: order rhythm per customer, a written risk view, and a save play owned by the person who handles the relationship. The reporting is simpler, but the discipline of triggers, owners and a signed baseline is the same.
What if customers left because of a service or quality problem?
Then we say so before designing any campaign. Reaching out to a customer who left over late delivery or a quality issue, without acknowledging it, usually confirms their decision. We tell you what the pattern in the data suggests, recommend what to fix, and design the reactivation message to address the reason honestly once the fix is real.
What is not included in this scope?
We do not rebuild your product, service delivery or pricing, though we report when those are the cause of loss. Loyalty programme design, customer success operating models and deep churn root-cause analysis are separate services. Paid advertising to existing customers, contact centre staffing and the licence cost of any messaging or CRM tool sit outside this scope and are quoted separately if you want them.
Talk to us
Let us audit the customers who have quietly gone.
The free audit is a working session, not a pitch. We look at how much of your revenue comes from repeat customers today, how many have already gone quiet, and whether you need the full programme or only a reactivation sprint.
- No obligation and no sales script
- A reply from someone who does the work
- Your details are never sold or shared