AI in SMEs: concrete use cases to automate and save time now

Reworking software means taking over existing code to correct it and make it evolve.
We start by accessing the code, then carry out a free maintainability analysis.
Depending on the quality of the code, improvements are made using the same technology or part of the code (often the back-end) is overhauled.
Changing service providers allows you to unlock fixes when the current team is no longer available.
Updating a language or migrating a technology improves performance and security.

Using AI in SMEs is no longer reserved for large companies: the aim is to show you where artificial intelligence can be applied right now, with concrete examples (emails, invoices, customer support, marketing, HR) and a simple method for making the transition a success.

This page accompanies the video of Rodolphe (co-founder of Iterates), an agency which develops software and applications for companies, and which is increasingly solicited for projects involving AI-enabled applications.

Why AI has become accessible to SMEs (and what it's really good for)

In the video, AI is described as software capable of think according to a given logic, and of’learn to deliver a result. The idea is not “AI for AI's sake”, but AI for AI's sake. serving the company's objectives Save time, improve the customer experience, or support over-stretched teams.

An important point for SMEs: AI can provide skills where you don't have a complete department (example given: a 4-person SME doesn't necessarily have someone dedicated to marketing).

Where to start: list, prioritise, choose a small IA project

The recommendation is very pragmatic:

Put on paper your needs and “automatable” tasks”
- including tasks that you don't do today because of a lack of time/skills (e.g. publishing a blog).
Prioritise
- identify a small project easy to set up.
Test on a reduced perimeter
- achieve initial success internally, then expand.

Suggested example: onboarding new employees. You choose a tool, incorporate the company's data and test it with the first arrivals.

General insight: starting small facilitates adoption, reduces risk and accelerates internal learning (what works/what needs to be adjusted).

Emails: read, sort and propose personalised draft replies

The first “painless” use is to let the AI do the work for you. pre-chew working on repetitive tasks.

Example presented: AI can read your emails and offer you a draft reply :

personalised,
based on the way you already respond,
to save time on routine requests (customers, suppliers, follow-up, etc.).

The idea is not necessarily to send automatically, but to save you time with a draft “just like you”.

Invoices & accounting: data extraction, accounting integration, invoices from estimates

The video explains that there have long been technologies capable of reading text on a document (e.g. PDF invoice / scanned document) and extracting data from it. The contribution of AI here: process and exploit this data in a single stream.

Case study :

you upload a supplier invoice,
the AI reads the data,
and can integrate into your billing/accounting system.

Key point: you don't have to change your accounting software. The approach described consists of link your existing software to an AI, for example by :

monitoring a “suppliers” mailbox,
sending emails and attachments to the AI,
then send the result to your accounting tool.

Another case cited: generate an invoice from a accepted quote (with your template).

For integrate AI into your business tools
For tailor-made AI applications

Chatbot 24/7: customer support, order tracking and reduction of redundant questions

Le personalised chatbot is presented as a highly profitable lever because it :

answers 24 hours a day, 7 days a week,
covers requests when your support is unavailable (e.g. a customer arriving on your site at 9pm),
and takes a lot of pressure off the teams.

Difference highlighted: we're no longer talking about a simple “decision tree”, but a chatbot that takes advantage of AI to give more “cognitive” and "intelligent" answers. personalized (provided it is powered).

Case in point:

a customer gives a order number,
the chatbot is connected to your e-commerce (or internal software),
it checks the status and explains the situation to the customer.

Expected result on the organisational side: if 80% of questions are redundant (example reported), the chatbot handles the majority and human support concentrates on the remaining 20% with more time and better quality.

AI call centre: make appointments connected to your diary (dental example)

The video features a very real case study: a dental practice that automates the booking of appointments by telephone.

What AI can do in this example:

connect to the’agenda,
be familiar with typical times (descaling 30 minutes, crown 1 hour),
identify the caller (number, existing patient),
adapt the language (e.g. answer in Dutch),
save the appointment in the diary.

The benefit: automate a repetitive task with little manual added value, and free up time for more useful tasks.

General point: as soon as a process depends on stable rules (times, availability, customer information), automation becomes easier to make reliable.

Marketing: blog, SEO, LinkedIn, email marketing connected to your CRM

Blogposts & SEO: produce faster, more regularly

The video focuses on a very widespread use: generating LinkedIn posts, blogposts and marketing content using tools such as ChatGPT.

The idea behind it: regularly publishing targeted articles in your field can improve your visibility on Google. At iterates, it is said that they generate numerous articles (with AI) to position themselves on searches related to customer needs (mobile applications, starting a project, subsidies, CRM, etc.).

Figure given : approximately 30% of prospects of Iterates would come from blog articles, with a CTA at the end of the article (“contact us”).

Operational reasoning :

an article can take 1 to 1.5 days to complete,
with AI: writing in a few minutes, proofreading + personalisation + visuals, aiming for around 30 minutes' work.

For the content generation with AI
For the’optimising your website's SEO

LinkedIn / personal branding: learning your tone

Another case explained: AI can analyse your LinkedIn posts (e.g. the last six months) and try to rethink the way you talk to publish 2, 3, 4 times a week.

General lighting: even if AI can imitate a tone, human touch-up is still useful to maintain an authentic voice and avoid an overly generic rendering.

Email marketing: personalised reminders via CRM

The video describes a scenario in which a CRM is implemented in a company :

you detect that a customer usually orders every two months,
but this month he's not ordering,
you trigger a personalised reminder based on their preferences,
and you can differentiate emails between an existing customer and a prospect (or abandoned basket).

HR & onboarding: CV scoring, automatic responses, internal “knowledge base” chat”

Recruitment: CV/job description compatibility

Example presented: an SME that receives 40 to 60 CVs a week. AI can :

read the job description,
analyse the CVs received,
allocate a compatibility note (e.g. 20%, 82%, 94%),
and help filter quickly.

It is also described that you can automatically send a response email to irrelevant applications (with a polite message), to let the HR team concentrate on the interesting profiles.

Onboarding: internal chat with company data

Another example: improve the onboarding experience and save time for HR via a chat-type space (visually similar to ChatGPT), but fed only with your internal data (holidays, canteen, timetables, internal rules, etc.).

Benefits described: time saved by HR + happier employees, because they find the answers quickly and get off to a good start.

Adoption: communication, training and rapid ROI

Two conditions are highlighted:

Communicate
If AI automates a process (quotes, invoices, etc.), the person who was doing the work may feel threatened. The “why” needs to be explained (e.g. reducing the time taken to send quotes in the construction industry, because a quote sent after 1 month means losing customers to faster competitors).
Train
An AI tool requires a period of adaptation. The video mentions the possibility of external support to ensure that the tool is used properly.

In terms of ROI, the idea is that an AI project can get off the ground quickly, with initial results in a short space of time - projects can be launched in “around two months” (meetings, implementation, training).

Tools to test (and what they're for)

Tom Generate presentations with text, images and layout/design (saves time on PowerPoint).
Flicklich Generate videos from an existing text/idea/content (voice-overs or texts).
Make.com Linking software and automating tasks (e.g. Gmail + invoicing + social networks). 5 free automations mentioned.
VAPI An example of an AI call centre to be tested.
Vitamin AI (Iterates) all-in-one tool (structured “SEO boost” blog, meeting minutes from audio, knowledge base, and other needs including onboarding).

The next stage

If you want to identify your 2-3 priority use cases and see how they can be integrated (existing tools or bespoke AI application):

Take Meet our AI expert for SMEs:
Or contact our AI creation agency in an SME :
Discover the services of our web agency that creates AI for SMEs iterates :

To remember

AI is already usable and accessible for SMEs, especially via existing tools and integrations.
Start with list your tasks (repetitive ones + the ones you don't do because you don't have the time), then prioritise.
Launch a small project simple, test it, achieve initial success, then extend it.
Case studies: email drafts, invoices/accounts, 24/7 chatbot, call centre RDV, blog/SEO, CRM/emailing, CV scoring, onboarding.
Success also depends on communication and training in-house.

Tell us about your project

One exchange, a thousand possibilities.

Describe your vision to us using this form: we'll analyse your request and get back to you within 24 hours with personalised advice and a concrete action plan.

We have the team and the resources to help you with your projects. Give us the details in this form and we'll get back to you as soon as possible to discuss them together.

Name

Company

Telephone

Subject

Message

Attachment