Case study · Funds management · Australia
The fund assistant that will not guess.
India Avenue is a Sydney and Mumbai fund manager focused only on India. Its small support team answered every investor email by hand. We built an assistant that answers investors at any hour, using only the fund documents. Today it resolves 80 to 85% of enquiries. The rest go straight to the team.
- Client
- India Avenue Investment Management
- Sector
- Funds management, Sydney and Mumbai
- Status
- Live in production
- What we built
- A customer support assistant
Change the product. The job stays the same.
- Fund managers
- Wealth platforms
- Super funds
- Insurers
- Lenders
- Private banks
Small support team. Numbers that must be right. Your assistant should answer only from documents you approve.
One inbox. One small team. Every question.
Before
4steps
Investors wanted a NAV, a year of returns or a statement. The figures were on the site, but hard to find. So they emailed, and the team answered each one by hand.
- 1Search the site
- 2Compare factsheets by hand
- 3Email the team
- 4Wait for a reply
When the team was busy, investors waited.
After
2steps
Investors ask in plain words and get the answer straight away, at any hour.
- 1Ask in plain words
- 2Get the answer, chart and source
- Compare factsheets by hand
- Email the team
- Wait for a reply
Requests that need a person go to the team, details ready.
Before the build
How this started.
There was no failed chatbot and no vendor shortlist. India Avenue had not started anything. We came to them.
Nothing came before it
No chatbot, no in house build, no vendors to compare against.
We came with something working
A working proof of concept, built for another Australian firm.
The demo did the explaining
Ask in plain words. Get an answer with a chart and its source.
They wanted to be early
Their goal was to be among the first Australian fund managers with AI for investors.
The two things they asked for
- Sound like India Avenue, not like a chatbot.
- Be right. And say so plainly when you are not.
Underneath both sat a simpler need. A small team could not keep answering every email by hand.
It did not start as a cost problem. It ended up giving the team its time back.
Why the bar was set where it was
What was at stake.
India Avenue holds an Australian financial services licence. A wrong return or a stray forecast can mislead an investor and put that licence at risk. So the assistant can be helpful, but never loose with a number or an opinion.
Every answer had to be:
Sourced
Every figure links back to an approved fund document.
From documents, not memory
Answers come from the factsheets and research, never the model's memory.
No advice, no forecasts
It shares data and explains strategy. It never recommends or predicts.
Honest about limits
If a figure is not in the documents, it says so and passes it to the team.
Discovery to production
How we worked.
- 01A few days
Discovery
Found the one job worth doing first. Learned the funds and the documents investors ask about.
- 02A few weeks
Proof of concept
One assistant, built only on their factsheets and research notes. Nothing from outside the documents.
- 03Two rounds
They marked it
They sent fifty real visitor questions. They marked every answer right or wrong, with the reason. We tuned it and ran it again.
- 04Ongoing
Production
Live as the Avenue Virtual Assistant. It answers investors and hands the rest to the support team.
What came out of round one
It was saying no. They wanted it to say where to go instead.
In round one it refused flatly when it had no answer. India Avenue wanted it to keep the investor moving. So now it says what it can do and points to the right next step.
The hard part
The figures were not text. They were pictures.
Three funds. About 200 documents. Factsheets, NAV sheets, research and white papers. Every answer starts with finding one figure in that pile.
What went wrong
Early answers were often wrong. The prompt was not the cause.
Many key figures sat inside images of charts and tables. A normal document search cannot read images. So the assistant answered from nearby words instead.
What fixed it
We split the pile by document type
Better prompting would not fix this. So we split the documents by type and gave each type its own search. Now the assistant looks for a figure where it actually lives.
If the assistant and the factsheet disagree, the investor has two answers. For a fund manager, that is brand damage.
In production
What it does now.
Fund questions
Strategy, holdings, fees and risks for the Equity Fund, India 2030 Fund and Active ETF.
Performance and positioning
Returns, distributions and sector weights, charted against the benchmark.
Research made short
Factsheets and research notes summarised in plain words, with a link to the full piece.
Open all hours
Answers investors when the team is busy or offline.
Hands off to a person
Statement requests go to the support team with the details captured. ETF holders go to their broker.
Sourced answers only
Answers only from India Avenue's documents, never from a number a visitor types. Every figure links to its source.
“I can certainly provide the detailed historical performance data and strategy documents for our funds, which many investors use to inform their decisions. For personalised guidance I recommend speaking with your own financial adviser, and our support team can provide any fund documents they may require.”
One real question, in the order it follows.
Watch the order, not the answer.
An illustrative recreation, not a live chat. It replays one real question in the assistant's own theme and chart component, and is not interactive. The figures in the chart are sample values until the real factsheet numbers are signed off.
Read the full transcript
Question: How is the India Avenue Equity Fund positioned by sector against its benchmark?
1. What it works out first
- Fund: India Avenue Equity Fund
- Share class: H Class, the hedged one
- Reads: Sector weights, fund and index
- Benchmark: MSCI India (Net) in AUD
2. Reads the latest factsheet
3. Chart: Sector positioning (Fund against benchmark, per cent of portfolio)
India Avenue Equity Fund against MSCI India (Net) in AUD across Financials, Industrials, Consumer, Info Tech, Health Care, Materials, Energy. Illustrative weights, not a live portfolio. Real factsheet figures go here once signed off.
Answer
Here is how the India Avenue Equity Fund sits against its benchmark, sector by sector. The weights come straight from the latest factsheet, linked below.
The fund holds 60 to 80 companies and leans towards mid and small caps, which is where it differs most from the index.
Suggested follow-up questions
- Why does the fund hold more industrials than the index?
- What are the top holdings of the India Avenue Equity Fund?
- Can you send me a copy of my holdings statement?
- Show me the sector positioning of the India 2030 Fund.
Scope
What we left out on purpose.
The brief included emailing statements, like tax reports. Their portal has no API, so we did not build it.
The assistant takes the request and passes it to the team. It never pretends to send the file.
80%+
of enquiries resolved without the team
Fund, performance and research questions are answered on the spot. Only the rest reach the team.
24/7
answers, even when the team is offline
Investors no longer wait for someone to be back at their desk.
50
real questions it had to pass
Supplied and marked by India Avenue. Every change still runs against them.
0
extra support hires
The same small team now handles only what needs a person.
“AvestaLabs worked closely with us to identify a high-impact AI opportunity that aligned perfectly with our business. The AI agent they delivered — trained on our internal data and research notes — has added real value to our business. They also helped us build customer support workflows to handle a wide range of inquiries effectively. Their collaborative approach, technical expertise, and strong command of agentic AI made the entire engagement seamless and impactful. We appreciate their professionalism and the tangible value they brought to our operations.”
Six months from now
Most AI projects quietly drift.
A figure goes stale or an instruction gets edited, and nobody notices until an investor does. So India Avenue's fifty marked questions became a test suite. Every change runs against them before it goes live.
- 1Does the figure match the fund document?
- 2Is this the right chart, and only one of it?
- 3Is the source link there?
- 4Has it stayed clear of advice and forecasts?
- Goes live
One it caught
Early on, it drew the same chart twice in one answer. We fixed the instructions, and the suite confirmed the fix held. It also lets India Avenue move to a newer model and prove nothing changed.
Start with the questions your team answers most.
A short discovery workshop, then a pilot on your own documents. You leave with a working assistant and the checks that keep it right.
Book an AI KickoffOr email hello@avestalabs.ai