Funnel experimentation and A/B testing · Marketing
Test whether a shorter form or faster follow-up gets more enquiries to say yes.
A form gets shortened, a follow-up message gets rewritten, an offer gets changed, all based on someone's reasonable-sounding hunch, and nobody checks afterwards whether it actually helped. Six months later, three changes have stacked on top of each other and there is no way to know which one, if any, made a difference.
Scope, fee, reporting and term are agreed in writing after we have read your brief and spoken to you.
In one paragraph
Funnel experimentation and A/B testing is the discipline of testing one change at a time, a form, a follow-up message, an offer, against a clear result. It replaces changing things on a hunch and hoping. GullySales builds a ranked list of what to test, runs each test properly against the current version, and reports honestly on what actually moved the number.
We treat the enquiry funnel as something to test deliberately. Each test changes one thing, against the current version, and runs long enough to give a real answer on one agreed outcome, enquiries or bookings. What wins stays, what loses goes, and we tell you which was which.
For example, a dental clinic gets about 200 form enquiries a month and wants to know whether cutting the form from nine fields to four brings more bookings. We would show the short form to half of the visitors and the current one to the other half over the same weeks, then count bookings, not just form fills, for each. With 30 enquiries a month that comparison cannot settle anything, and the honest first step is fixing how fast you reply. This example is hypothetical.
The problem
Does this sound like your business?
What owners tell us on the first call, in their words.
“A change is made on a hunch and nobody checks if it helped.”
The enquiry form gets shortened because someone felt it was too long, with no before-and-after comparison to confirm it worked.
“Two follow-up sequences run at once with no comparison.”
Different team members send different follow-ups, and nobody has ever compared which one actually gets more replies.
“The team keeps guessing why conversion dropped.”
Opinions are offered in a meeting, one is picked, and the fix is applied without ever testing whether the guess was right.
What we do
Everything that is included.
Each deliverable, and what it gets you. Nothing here is an extra.
Experiment backlog
A ranked list of changes worth trying, based on where the funnel actually loses people, not the loudest opinion in the room.
Test design
For each test, the current version and one changed version, differing in a single thing, so the result actually means something.
Tracking set-up
The measurement in place before a test starts, so the result is not argued about after the fact.
Test execution
A test run end to end until the numbers are enough to trust, not stopped early because someone is impatient.
Results readout
A clear, honest report on what happened, including when a test shows no meaningful difference.
How the result is measured
- Experiments run against the backlog
- Funnel stages with a documented conversion baseline
- Experiments that produced a clear, actionable result
- Ideas on the backlog tested, dropped or still waiting
Recorded as a baseline before work starts, so every later report has an honest comparison.
Who it is for
This is written for you if these sound familiar.
- Businesses with enough website traffic or enquiry volume to test changes meaningfully
- Teams who currently change forms, pages or follow-up messages on instinct with no way to check the effect
- Companies with a clear drop-off point in the funnel and several plausible fixes to choose between
- A dealer with a quote-request form that many people start and few finish, and a follow-up message nobody has ever compared with another
Not for
It is not the right fit if.
- You get only a few dozen enquiries a month; most such firms cannot run a fair test, and replying faster to every enquiry will do more for you first
- You want to change several things at once and see what happens; that does not tell you which change caused the result
How it works
From your first message to the first report.
No open-ended retainer. Every step gives you something in writing.
First
Your brief
You tell us what you need in a few short steps: your business, what you want, where enquiries come from and how they are handled.
After the brief
A conversation and a written scope
We read it, then talk with whoever handles your enquiries. If you want our help, the scope, the fee and the reporting come in writing, and nothing starts before you have read them.
Before work starts
Baseline recorded
The numbers this work is meant to move, such as enquiries, quotations, orders, visits or renewals, written down so every later report has an honest comparison.
After the baseline
The first fix goes live
The fix we agreed to start with is put in place and measured against the baseline, whether that is a reply, a follow-up, a page, a listing or the handover between marketing and sales.
As agreed
Report against the baseline
What moved, what did not, and what changes next, in plain words. How often you get it is set in writing before work starts.
At renewal
Renew on the numbers
The term ends and you decide whether to continue from the results. The length is agreed in writing before anything starts.
How the work runs
- 1
Audit the funnel
We map where enquiries currently drop off and how many people reach each stage, to see whether a fair test is possible at all.
- 2
Build the experiment backlog
We rank ideas worth testing by likely impact and how quickly they can produce a reliable result.
- 3
Design and run the test
The current and changed versions are built, tracked properly, and run side by side until the answer is real.
- 4
Measure the result
We report what actually happened against the agreed measure, honestly, including inconclusive results.
Why us
Why owners pick GullySales over an agency.
Marketing and sales, as one job
Most agencies stop at the enquiry. We also fix what happens after it: the reply, the follow-up, the quote and the CRM.
The person on the first call does the work
No account managers in between. You are never handed to someone you have not met.
A baseline before anything starts
Your numbers are written down on day one, so every later report compares against something honest.
The fee in writing, split three ways
Our time, your media spend and production on separate lines. You always see what goes to us.
No guarantees we cannot keep
The term is agreed in writing and never a default twelve months. We never promise a ranking or a lead count, because nobody controls those.
One office, and we say so
Nagarbhavi, Bengaluru. We work across India by call and WhatsApp and travel when a session needs to be in person.
#257, 3rd floor, Sri Nanjundeshwara Complex, Nagarbhavi 8th Block, Outer Ring Road. How we work.
Engagement options
Ways to work with us.
Pick the size of commitment that fits. Every option starts with your brief and a conversation.
Option 1
The conversation on its own
A call after you send your brief. It says honestly whether you need outside help, and many fixes are ones your own team can make.
Option 2
One fix, scoped
Start with the fix that matters most, such as reply time or follow-up. The fee is in writing before anything starts.
Option 3
An ongoing programme
We run the work, report against the baseline, and you renew on the numbers. The term and the reporting are agreed in writing first.
Option 4
Guidance for your own team or agency
We plan, brief and check the work of your in-house team or current agency, instead of replacing them.
Your brief
Tell us what you need, and we will read it before we call.
Eight short steps about your business, what you want and who decides. Most questions are a tap, and you can skip any you would rather answer on the call.
FAQ
Questions owners ask before they call.
Not here? More answers, or ask on WhatsApp.
How much traffic do we need for this to work?
How soon will we know if a test worked?
What if a test shows no difference?
Can you test more than the website, such as follow-up messages?
How is this priced?
How long is the contract?
Who will actually do the work?
What do you need from us?
What if we are not happy with the work?
By industry
Funnel experimentation and A/B testing, written for your industry.
Related services
Often bought with this, or instead of it.
Your next practical step
Tell us what you are trying to grow, and we will read it before we call.
Eight short steps, mostly taps. Skip anything you would rather answer on the call.