In Practice

What Tattvanet looks like in practice.

A real client engagement with tracked results, followed by illustrative examples of how we approach common business problems. Anything labelled illustrative is a representative scenario, not a specific client.

Property & LettingsAn independent Manchester lettings agency

An AI tenant agent that handles 7 in 10 enquiries

The challenge

An independent lettings firm managing 386 properties with a nine-person team was drowning in tenant communications — 95–130 emails a day plus WhatsApp and web enquiries, the bulk of them the same handful of questions about rent, payments, and maintenance. Admin staff were spending five to six hours a day just triaging inboxes, maintenance requests took two to four days to be assigned, and tenants were repeatedly chasing for updates. The firm was about to hire a third administrator at £28,000 a year to cope.

The Tattvanet solution

We built an AI tenant communication agent across their shared support inbox and website chat, trained on 2,400+ historical tenant emails plus their tenancy agreements, maintenance procedures, and arrears policy. It auto-handles routine rent and payment queries, and runs a structured maintenance intake that gathers the details, categorises each job as emergency, urgent, or routine, raises a complete ticket, and assigns a contractor automatically. Strict escalation rules keep humans in control: anything involving complaints, legal language (council, solicitor, court), or payment negotiation is flagged straight to the right property manager — the AI never touches it.

The outcome

Over the first 90 days the agent fully resolved 71% of tenant enquiries, cutting human-handled emails from around 120 a day to about 35. Average response time fell from 12–24 hours to under two minutes, and maintenance jobs that previously took up to 38 hours to assign were being assigned in roughly 18 minutes — about 52% faster. Admin staff recovered an estimated 4.5 hours each per day, the planned third hire was cancelled, and the firm estimates an annual saving in the region of £54,000. The director reported a noticeable drop in complaints within six weeks.

Under 2 min

avg response time, from 12–24 hrs

~52%

faster maintenance assignment

~£54k

estimated annual saving

Result

71%

of tenant enquiries resolved by AI

Illustrative examples

HealthcareMulti-location dental practice groupIllustrative scenario

Turning missed calls into booked appointments

The challenge

Over 150 inbound calls a day across three locations, with an estimated 30% going unanswered at peak hours — meaning lost appointments and a stream of reviews citing poor phone accessibility.

The Tattvanet solution

We deployed an AI call answering agent trained on the practice FAQ and integrated with their practice-management software to book appointments directly. Clinical concerns and complaints were flagged and transferred to staff in real time with a full transcript.

The outcome

A deployment like this is designed to lift the answer rate to around 94% (from roughly 70% at peak), with the agent taking on a meaningful share of new bookings and easing reception pressure within the first couple of months.

Illustrative target

94%

inbound call answer rate

E-commerceIndependent fashion retailer going online-firstIllustrative scenario

Scaling a 4,000-SKU catalogue without scaling the team

The challenge

A 4,000+ SKU inventory and a new e-commerce platform. Writing individual product descriptions would have taken 600+ hours of copywriting, and supplier-feed copy was generic and off-brand.

The Tattvanet solution

We built a custom content generation pipeline trained on the retailer’s brand voice, ingesting raw product data and producing on-brand descriptions automatically — plus a lightweight review interface for the team to spot-check outputs.

The outcome

A pipeline like this can take 4,000+ descriptions from raw data to reviewed and approved in around eleven days — work that would take an in-house writer over four months — producing on-brand copy designed to outperform generic supplier feeds on conversion.

Illustrative target

11 days

to ship 4,000+ descriptions

LegalBoutique employment law firmIllustrative scenario

Automating the backroom of a growing legal practice

The challenge

Fee earners were losing six to eight hours a week each to manual record updates, copying between systems, generating standard letters, and chasing documents — non-billable work that wasted senior expertise.

The Tattvanet solution

We audited operations, automated five high-frequency workflows (intake capture, document generation, billing updates, follow-ups), and deployed an internal AI assistant trained on the firm’s precedent library for instant document retrieval.

The outcome

Automating these workflows can recover on the order of 20+ hours per week across a small team — over half a full-time role — and cut standard correspondence turnaround from days to hours.

Illustrative target

21 hrs

recovered per week

The scenarios in this section are illustrative examples of the kind of outcomes our solutions are designed to achieve, not records of specific client engagements. Their figures are representative targets, not guaranteed results.

Could this be your story next?

Tell us about the process that's slowing you down, and we'll show you what AI could do about it.

Start a conversation Explore services