+50% more revenue from moving conversion 1% → 1.5%
5 pillars of post-launch growth
5× a new customer costs more than a returning one
Post-launch growth and CRO for online stores

Launching an online store isn't the finish line. It's the start. The real work begins here — and it's the more enjoyable part, because for the first time you're working with real data instead of guesses.

This article is a playbook for what to do from the first day of going live to your first million in monthly revenue. Three pillars: retention, CRO and scaling. Each of them can double your revenue without adding a single koruna to your ad budget.

If you've read the pillar guide to starting an online store, this is a deep dive into Phase 10.

Pillar 1: Customer support (the foundation for everything else)

Bad customer support breaks everything else. Great advertising brings the customer in, a beautiful checkout nudges them into an order, and then no one replies when they ask about sizing. The customer cancels the order, gives you one star and never comes back.

This isn't just an annoying experience — it's the peak-end rule in action. In Thinking, Fast and Slow (2011), Daniel Kahneman showed that people remember an experience by its most intense moment and by how it ends, not by its average. A frustrating exchange with support gets fixed in memory as the defining moment of the entire brand interaction, even if everything else was excellent. That customer leaves with an emotional injury that, according to prospect theory (Kahneman & Tversky, 1979), is roughly twice as powerful as the pleasure of a good experience.

The minimum set you have to have

Channel

Response time

Tool

Live chat

< 5 minutes during business hours

Smartsupp, Tawk.to

Email

within 24 hours

Helpdesk or info@ directly

Phone

during business hours

+ voicemail outside them

Social media (DM)

within 1 hour

Meta Business Suite

Help centre / FAQ

self-service

Shoptet FAQ page

Live chat: the highest-ROI tool in customer support

Smartsupp is the de facto standard for online stores. It plugs straight into Shoptet, integrates easily, and has a free plan to start with.

What to switch on right after launch:

A rule from experience: In the first month after launch, reply to everyone personally. Not through an assistant, not through a chatbot. You'll learn from the questions what people really want to know. And these insights are the most valuable marketing research you'll ever get — because what customers ask is exactly what belongs in the product description, the FAQ and the ad copy.

A chatbot for the FAQ (only later)

In the Shoptet Market you'll find several AI/chatbot solutions for automated support. Only deploy them once you have a real volume of questions you can no longer handle personally (typically dozens of enquiries a week). Before that it's overkill, and customers can tell they're talking to a robot — which, paradoxically, erodes trust more than a short wait for a human reply would.

Pillar 2: Retention — because a new customer costs substantially more than a returning one

A rule every e-commerce specialist learns in their first week: the cost of acquiring a new customer is, in long-term analyses, typically several times higher than the cost of retaining an existing one (Frederick Reichheld, Harvard Business Review, 1990; The Loyalty Effect, 1996). And yet I keep seeing stores that burn their whole budget on acquisition and neglect retention entirely.

Why is retention so profitable? Because a returning customer has already cleared the entire trust barrier — they've overcome the pain of paying (Knutson et al., Neuron, 2007), overcome the uncertainty about quality, and they know your brand. The second purchase is cognitively far easier for them than the first. On top of that comes cognitive dissonance (Festinger, 1957): a customer who has already bought from you has an inner motivation to reassure themselves it was a good decision — and the easiest way to confirm it is to buy again.

Loyalty programme

How to do it in Shoptet: the Shoptet Market (doplnky.shoptet.cz) offers several add-ons for loyalty programmes and reward systems (points for purchases, registrations, reviews, VIP tiers). Choose based on the current offering and reviews — the Shoptet Market changes over time.

Rules for setting it up:

The customer has to be able to see their points balance in their account. Otherwise they forget about it and the programme loses its point. And there's another reason to show progress: the goal-gradient effect (Hull, 1932) — the closer we are to a reward, the more effort we're willing to put in. When a customer sees they're 200 CZK short of the next tier, they're more likely to place another order just to cross the line.

Loyalty discounts by spend

Instead of points, you can set up automatic discounts based on total spend. Shoptet can assign these automatically. The customer sees their current discount in their account, and that triggers the endowment effect — a discount they "already have" psychologically becomes part of their identity as a customer, and giving it up (by switching to a competitor) would feel like a loss.

Win-back campaigns

A win-back email after 60–90 days without a purchase is one of the highest-ROI emails you'll ever send. Both the open rate and the conversion rate of a well-targeted win-back campaign tend to be several times higher than an ordinary newsletter, because you're targeting people who have already been through the whole buying process and cleared the barriers that cold traffic still has ahead of it.

A structure that works:

  1. Subject: short, personal, with the customer's name. Personalised subject lines consistently lift open rates in email-provider benchmarks — the reason is the cocktail party effect, the brain is wired to respond to its own name even in a noisy environment.
  2. Hook: "It's been a few weeks since we last saw you."
  3. Offer: a discount on the next purchase, free shipping, or a gift with the order.
  4. A reminder of favourite products (based on history). Here you're working with the endowment effect — products the customer has looked at before psychologically "belong" to them.
  5. CTA: "Come back to the store."

Send it automatically 60 days after the last purchase, a second wave after 90 days, a third after 120 days (with a different offer).

Birthday campaigns

Very underused. Collect birthday data in the customer account + an automatic birthday email with a discount or a gift. Both the open rate and the conversion rate of birthday emails are among the highest in the entire email programme, and the reason is psychologically clear: the email combines reciprocity (a gift) with personal relevance (name, date) and evokes positive emotions that get associated with the brand.

Review requests

7–14 days after delivery, an automatic email with a link to the review form. The incentive: a small discount coupon or points in the loyalty programme.

A practical rule: Reviews are doubly valuable — they help with conversion (social proof) and with SEO (UGC content). Ask for them systematically. And remember the counter-intuitive finding from social-proof research: a perfect five-star score converts worse than a realistic 4.0–4.7. People look for authenticity, not perfection. Minor negative reviews paradoxically raise your overall credibility.

Referral programme

"Refer a friend, you both get a discount." A self-amplifying mechanic that brings you quality customers at marginal cost. It works because a recommendation from a friend has an order of magnitude higher conversion value than a hundred reviews — the reason is the unity principle Cialdini describes in Pre-Suasion (2016). When I'm recommended by someone in my "in-group" (a friend, a colleague, family), I automatically inherit part of their trust.

Structure:

In the Shoptet Market you'll find add-ons for referral programmes, or you can connect independent global tools (Refersion, ReferralCandy and others) via API.

Pillar 3: CRO (Conversion Rate Optimisation)

Improving the conversion rate from 1% to 1.5% is +50% revenue without a single koruna more in advertising. This is the highest-ROI activity in all of e-commerce, and yet CRO is something most online stores don't do at all. Either because they don't know about it, or because it's "complicated." It isn't complicated. It requires discipline and measurement.

What to measure and where

1. Funnel analysis in GA4:

If your numbers drop below the benchmark somewhere, you know where to fix things. (These benchmarks vary by category — they're lower in luxury, higher in fast-moving goods. Focus on the trend, not the absolute number.)

2. Microsoft Clarity:

3. Hotjar or VWO — for advanced A/B testing.

Why session recordings work better than you'd think: GA4 tells you what is happening (where people leave). Recordings show you why. And that's a chasm of a difference. Eye-tracking studies from the Nielsen Norman Group have long documented that users behave on a website completely differently from how designers imagine — they scan in predictable patterns, click on things that look clickable (even when they aren't), and ignore perfectly visible CTAs if those aren't in the "right" part of the page according to the F-pattern or Z-pattern. Without watching real behaviour, you fix things nobody minds and overlook the ones that actually stop people.

The most common CRO opportunities

  1. An insufficiently visible CTA — colour, size, position. The Von Restorff effect helps here (Hedwig von Restorff, 1933): an element that stands out visually from its surroundings is automatically remembered by the human brain. A button in the same colour as the rest of the site simply vanishes.
  2. Too many fields in the checkout — every field = another reason to leave. Fogg's behaviour model (B=MAP) says behaviour occurs when motivation × ability × prompt crosses a threshold. Every field lowers ability (put simply: how much effort completing it takes).
  3. Slow loading — > 3 seconds = a lost customer (Google Search Central data).
  4. A hidden shipping cost — the customer only sees the price in the cart, gets a shock and leaves. This is the leading reason for cart abandonment according to the Baymard Institute (baymard.com).
  5. Undescribed availability — "is it in stock or not?"
  6. Missing trust signals — Heureka Ověřeno, reviews, certificates.
  7. Mobile layout problems — buttons too small, forms cluttered.
  8. An oversized cookie banner — it covers the content and frustrates.

A/B tests worth running

1. Test headlines. The homepage headline, category headlines, hero copy. Changing a single word can lift your conversion rate by 10%. One of the most famous small CTA tests: changing "Start your free trial" to "Start my free trial" — that is, switching to the first person — led to a recorded increase in click-throughs of tens of percent. The reason is psychological: the word "my" gives the user a sense of ownership and control.

2. Test CTA buttons. Colour, text, size. "Add to cart" vs. "Buy now" vs. "I'm interested". There's no universal winner — you have to measure it for your brand.

3. Test product photos. Studio vs. lifestyle. With a person vs. without. Close-up vs. wide shot. Lifestyle photos with people tend to have an edge thanks to mirror neurons (Rizzolatti et al., 1996) — when a viewer sees someone using a product, the brain partly simulates it and builds a desire to own it.

4. Test the product-page layout. The order of sections, the amount of text, the number of reviews.

5. Test the checkout flow. Single-page vs. multi-step. Mandatory registration vs. guest checkout. Fewer fields vs. more fields. Mandatory registration has long been one of the worst conversion killers, because it disrupts processing fluency and re-triggers a barrier the customer thought they'd already cleared.

Rule: Test one thing at a time. If you change three things and the conversion rate improves, you don't know which one did it.

Trust signals that work

Trust signals on the payment page are among the best-documented CRO interventions there are — they directly address the pain of paying (Knutson et al., 2007) and loss aversion (Kahneman & Tversky, 1979). When a customer sees a secure-payment logo and reviews right by the "Pay" button, the insula (the brain region that processes pain) sends a weaker signal and the reward centre a stronger one — and the purchase decision tips toward conversion.

Urgency and scarcity (carefully!)

⚠ Never lie with urgency. Fake countdown timers and bogus "last 5 left" claims are dark patterns, and the moment a customer realises they've been deceived, it destroys trust in the brand irreparably. On top of that, they trigger psychological reactance (Brehm, 1966) — the feeling of being manipulated makes us deliberately not do what's being asked of us. The Czech Trade Inspection (ČOI) also actively monitors and fines these practices.

Pillar 4: Scaling

Once you have retention and CRO under control, it's time to scale. Scaling isn't just about putting more money into advertising. It's about systematically increasing the capacity of the whole business.

Expanding the product portfolio

International expansion

Shoptet supports multi-language — you can run a store in Czech and other languages under a single admin.

What you need for expansion:

  1. Content translation — product descriptions, categories, legal pages.
  2. Local payment methods — a German customer wants SEPA, a Polish one BLIK.
  3. Local shipping — DHL, GLS, local pickup points.
  4. Local VAT (the OSS scheme).
  5. Local legal pages — German GDPR has its specifics.
  6. Local customer support — language, time zone.
  7. Local currency — EUR, PLN, etc.

In the Shoptet Market there are add-ons for AI product translation into foreign languages. It's not perfect, but for a start it speeds things up — you then fine-tune only the key pages by hand (homepage, About, FAQ, checkout copy).

A rule from experience: Start with one country, not three. Slovakia is the default first choice (language, culture, shipping). Poland and Germany come later. Expanding into 5 countries at once is a path to burnout and diluted resources.

Automating processes

What you should automate once you have more than 50 orders a month:

AI tools for scaling

In the Shoptet Market and beyond, there's a range of AI solutions gradually gaining ground in e-commerce:

Before deploying any AI, ask yourself two questions: (1) What will it save me? (hours per month), (2) What will it cost me in quality if the customer realises they're talking to a bot? If the second question outweighs the first, wait.

Hiring (and when to start)

Stage

Revenue/month

What to hire

Garage

0–100,000 CZK

You, everything

Getting going

100–500,000 CZK

Part-timer for packing / VA for support

Growth

500,000–2M CZK

Full-time customer support, marketing specialist

Scale

2–10M CZK

Marketing manager, accountant, in-house fulfilment

Pro

10M+

Strategic team, head of operations, head of marketing

Rule: Hire customer support first, not a marketer. You'll do more effective marketing yourself than a marketer who doesn't know your brand. But customer support eats up time you could be spending on strategy.

Increasing ad spend — when and how

Not before you have:

Once you have all of this, scale gradually:

Pillar 5: Reporting and the monthly review

What doesn't get measured doesn't get managed. If you don't track your KPIs regularly, you won't know whether you're growing or declining until an unpleasant surprise arrives.

Monthly KPI dashboard (the minimum you have to have)

Business KPIs:

Marketing KPIs:

Operational KPIs:

Monthly review (a mandatory discipline)

On the 1st of every month, give yourself an hour to review the previous month:

  1. What worked best? — find the top 3 things and do more of them.
  2. What worked worst? — find 3 things and either fix them or scrap them.
  3. What's the biggest growth opportunity for next month?
  4. What's the biggest threat for next month?
  5. What decision do you have to make this week?

That hour a month decides whether you grow by plan or by accident.

The full series: How to start an online store in 2026

This article is part of a seven-part series on starting an online store in 2026. The other parts:

FAQ

What is the conversion rate, and what counts as good?

Conversion rate = number of orders / number of visitors. The average across online stores tends to sit between 1–2%. Specialist stores with a well-built checkout and a strong brand run at 3–5%. Below 0.8% is a problem — typically a fault in the checkout, the pricing, or the quality of the traffic.

How much does a loyalty programme cost?

Technically it can start free (simple discount codes for returning customers) and run up to a few thousand CZK a month (platforms such as Smile, LoyaltyLion, ReferralCandy). For a smaller store, a solid choice is your own setup in Shoptet with coupons and email automation — no software cost, just the time to configure it.

How often should you run A/B tests?

Not everything at once. The rule: one A/B test at a time on the same page, a minimum of two weeks running, and at least 1,000 conversions per variant for statistical significance. Cadence: 1–2 tests a month for a store with 50+ orders a day. Fewer orders = longer tests = fewer tests per year.

When should you make your first hire?

When you personally work 60+ hours a week and 30% of that is repetitive tasks (packing, customer support, reports). The first hire is usually customer support/packing on a part-time basis (15–20 hours a week). Not a marketing specialist — that's better outsourced in the first year.

What is Customer Lifetime Value (LTV)?

LTV = the average value of everything a customer spends with you over their entire lifetime. Calculation: average order value × number of orders per year × length of loyalty in years. If your LTV is 5,000 CZK (≈ €200) and your CAC (customer acquisition cost) is 800 CZK (≈ €32), your LTV:CAC ratio is 6.25. A healthy benchmark is 3+, ideally 5+.