What AI budget for a store under €1M?

Far less than people assume, if the order is respected. A store under a million gets most of the gain from the native and SaaS layer: Shopify Magic (included), a support tool like Tidio or Gorgias, Klaviyo on its free or first paid tier, and one personalisation tool like Rebuy. That delivers most of the result with zero engineering. Advanced on-site personalisation platforms, advertised around $500–2,000/month, are not justified at that traffic level.

Gorgias or Intercom Fin for support?

Gorgias if the store is the whole subject: it natively understands e-commerce objects — order, return, shipment — and wires the agent onto them with no integration work. Intercom Fin if you already run Intercom, or if conversational quality matters more than domain specialisation. Both automate a meaningful share of enquiries; the difference is time to value, which is shorter with Gorgias on a Shopify store.

Is AI personalisation worth the cost?

It depends entirely on your traffic. A personalisation engine needs volume to separate signal from chance: below a few thousand monthly sessions it optimises noise and you pay for an unmeasurable effect. Above that the effect becomes real, especially where intent already exists — cart and checkout — which is why Rebuy often produces faster results than homepage personalisation.

What about GDPR for a European store?

Almost all of these tools are American and process customer data: orders, browsing behaviour, contact details. That is not prohibited, but it requires paperwork — standard contractual clauses, a transfer assessment, customer notice — and a consent banner that genuinely blocks trackers before acceptance. The most commonly neglected point is not the contract but the banner: a personalisation tool firing before consent makes the rest of the file moot.

AI E-commerce Automation: What to Automate First (2026)

The short answer

Deploy in this order: support (Gorgias, or Tidio if you are small), content (Shopify Magic, free and already there), lifecycle (Klaviyo, or Omnisend for something simpler), personalisation (Rebuy at the cart, Nosto if you have the traffic), then operations (Inventory Planner) last. Each layer funds the next.

The deployment order

Comparisons sort these tools by category. Useful for understanding the market, useless for deciding: no store deploys eleven tools, and the ones that try stop at the third. The real question is where to start.

Order	Layer	Why here	Time to effect
1	Customer support	Repetitive, measurable volume: tracking, returns, delays	A few weeks
2	Product content	Free on Shopify, no integration	Immediate
3	Email / SMS lifecycle	Acts on a list you already own	One to two months
4	On-site personalisation	Needs traffic to produce a reliable signal	Two to three months
5	Forecasting and operations	Requires clean sales history	A quarter or more

Deployment sequence observed in stores that see it through. Lumyniq, 2026.

The reason is simple: the early layers produce visible results quickly, which funds and justifies the later ones. The later layers need clean data — reliable sales history, a coherent tracking plan — that the early layers help build. Starting at the end means optimising on data you do not have yet.

The comparison

Tool	Layer	The right case
Gorgias	Support	The first project for almost every store
Klaviyo	Lifecycle	The second project, once support is under control
Shopify Magic	Content	The free starting point for any Shopify store
Rebuy	Personalisation	Raising average order value without touching acquisition
Nosto	Personalisation	Established stores with traffic worth segmenting
Tidio	Support	Small stores that find Gorgias oversized
Intercom Fin	Support	Stores already running Intercom
Omnisend	Lifecycle	Stores starting their lifecycle programme
Octane AI	Personalisation	Catalogues where choosing is hard: beauty, supplements, equipment
Yotpo	Lifecycle	Stores wanting to activate social proof and retention
Inventory Planner	Operations	Stores whose real problem is stockouts or overstock

Positioning read on 15 August 2026 from vendor sites. Lumyniq, 2026.

The eleven tools, one by one

01 · Support

Gorgias

A helpdesk built for e-commerce, with an AI agent handling order tracking, returns and shipping questions.

The right case: The first project for almost every store

Strengths

The vendor claims up to 60% of enquiries handled automatically
Native understanding of e-commerce objects: order, return, shipment
First-class Shopify integration

Limitations

The claimed rate depends heavily on your catalogue and policies
Little relevance outside e-commerce

02 · Lifecycle

Klaviyo

The e-commerce email and SMS benchmark: lifetime-value forecasting, next-order prediction, send-time optimisation.

The right case: The second project, once support is under control

Strengths

Predictions that are genuinely usable, not cosmetic
Free tier to start
Ready-made cart and repeat-purchase flows

Limitations

Billed on profile count
Gets expensive as the list grows

03 · Content

Shopify Magic

Built into the Shopify admin: description generation, image editing, basic replies.

The right case: The free starting point for any Shopify store

Strengths

Free and already there: no argument for not trying it first
Nothing to integrate

Limitations

Limited depth next to dedicated tools
Shopify only

04 · Personalisation

Rebuy

Behaviour-driven recommendations and upsells, from cart through checkout.

The right case: Raising average order value without touching acquisition

Strengths

Acts where purchase intent already exists: cart and checkout
Measurable effect on average order value fairly quickly

Limitations

Centred on the Shopify ecosystem
Overuse degrades the buying experience

05 · Personalisation

Nosto

Full-site personalisation: recommendations, content, A/B testing, behavioural pop-ups.

The right case: Established stores with traffic worth segmenting

Strengths

The most complete on-site personalisation on this list
Built-in A/B testing

Limitations

Advertised around $500–2,000/month: outside a small store's budget
Needs traffic volume to produce reliable results

06 · Support

Tidio

Chat and an AI agent for modest stores, at a low entry price.

The right case: Small stores that find Gorgias oversized

Strengths

Accessible entry cost
Quick to get running

Limitations

Less e-commerce-specialised than Gorgias
Ceiling reached at high volume

07 · Support

Intercom Fin

Intercom's support agent — generalist, but very strong on answer quality.

The right case: Stores already running Intercom

Strengths

Conversational quality among the best
Careful human handoff

Limitations

Fewer native e-commerce objects than Gorgias
Per-resolution billing worth watching

08 · Lifecycle

Omnisend

E-commerce email and SMS, simpler and cheaper than Klaviyo.

The right case: Stores starting their lifecycle programme

Strengths

Quick to pick up
Ready-made flows

Limitations

Markedly weaker predictions
Ceiling as the store grows

09 · Personalisation

Octane AI

Product quizzes and zero-party data collection, so recommendations stop being guesses.

The right case: Catalogues where choosing is hard: beauty, supplements, equipment

Strengths

Declared data beats inferred data for recommending
Feeds email segmentation afterwards

Limitations

Mostly relevant in certain verticals
Requires real work designing the quiz

10 · Lifecycle

Yotpo

Reviews, loyalty and SMS in one suite.

The right case: Stores wanting to activate social proof and retention

Strengths

Covers reviews, loyalty and SMS without stacking three tools
Good Shopify integration

Limitations

Broad suite, uneven depth across modules
Cumulative cost once everything is switched on

11 · Operations

Inventory Planner

Demand and replenishment forecasting, with ERP integration.

The right case: Stores whose real problem is stockouts or overstock

Strengths

Attacks a cost line the other tools ignore entirely
Integrates with existing management systems

Limitations

Quote-based pricing
Without clean sales history the forecasts are worthless

By store size

Revenue	Recommended stack	What to avoid
Under €1M	Shopify Magic + Tidio or Gorgias + Klaviyo + Rebuy	Advanced on-site personalisation: not enough traffic
€1–10M	Add Nosto or Octane AI, and Yotpo for retention	Stock forecasting before you have clean history
Over €10M	Add Inventory Planner and custom orchestration	Stacking tools that do the same job without removing any

Scoping reference points, not rules. Lumyniq, 2026.

The trap on the third row deserves a word: growing stores accumulate tools and never retire any. You end up with three systems sending email, two recommending products, and nobody able to say which produced which sale. A tooling audit is as profitable an exercise as adding one more tool.

When the stack gets too heterogeneous to live inside SaaS products, a custom orchestration layer starts to make sense — that is what Lumyniq does. Disclosure: we publish this site and do not appear in this comparison, because we are not an e-commerce tool.

Frequently asked questions — AI and e-commerce

Customer support, in the overwhelming majority of cases. It is the area where volume is most repetitive — order tracking, returns, delivery times — so it is where automation produces a measurable effect within weeks. Then content (product descriptions), then email and SMS lifecycle, then on-site personalisation. Analytics and forecasting come last, whereas that is often where stores begin, because it is the most impressive topic in a meeting.

AI e-commerce automation: what to automate first

The short answer

The deployment order

The comparison

The eleven tools, one by one

Gorgias

Klaviyo

Shopify Magic

Rebuy

Nosto

Tidio

Intercom Fin

Omnisend

Octane AI

Yotpo

Inventory Planner

By store size

Frequently asked questions — AI and e-commerce

Read next

Sources

Let's talk about your project

Contact details