We turn email, SMS and WhatsApp messages from announcements sent one at a time into flows that run themselves according to what the customer does. Abandoned basket, welcome, win-back and post-purchase flows are built; people who show interest are scored and handed over to the sales team. Consent and sending discipline are built into the flow itself.
In most companies marketing communication is done in one way only: a list is prepared, the same message goes to everyone, the result is discussed for a few days and then forgotten until the next announcement. This method creates work for every send and wears people down as it repeats; but the real problem is this: the message goes out on the company's calendar, not the customer's. Someone who abandons their basket is contacted not at that moment but in the bulk send two weeks later. A customer who has just placed a first order hears nothing. Someone who has not bought for six months and someone who bought yesterday read the same text. A message with the wrong timing usually creates irritation even when the content is right, and it pushes up unsubscribes.
Off-the-shelf email tools solve this up to a point. Preparing a template and sending a list is easy; but if the flow is to be tied to what the customer actually does, the job rests on data. For the tool to know about basket abandonment it has to receive events from the e-commerce site, to know order status it has to read data from the ERP or the order system, and at a manufacturer selling through dealers it has to look at service and warranty records. When the data does not arrive, what is left is hand-built lists again and the tool turns into an expensive outbox. That is why, when we build marketing automation, the first place we look is not the message itself but which event can be obtained from which system. What needs to be discussed on day one of the build is not which template will be used, but in what form and with how much delay the basket event will leave the site.
A marketing automation platform (MAP) ties the message to an event. A flow starts with a trigger: the basket was left half-finished, the first order was delivered, the catalogue page was viewed for the third time, there has been no purchase for sixty days. It then proceeds with conditions, waits and channel choices. Within the same flow, if the email is not opened it can drop to SMS, and if the customer buys in the meantime the flow can be stopped. Those who show interest are scored: which behaviour is worth how many points is decided together, and anyone crossing a given threshold is handed over automatically to the CRM, that is, to a sales representative. Two different subject lines, channels or timings are compared with A/B testing; the decision is made by measurement rather than guesswork. Every step of the flow is logged; why a particular customer received that message can be shown afterwards, one by one. Channel cost is part of the flow too: the free channel is tried first, and the paid channel only comes into play when it is needed.
Let us write the limits down at the outset as well. Sending commercial electronic messages is subject to consent in Türkiye, and recipient approvals have to be uploaded to the İleti Yönetim Sistemi (İYS), the country's central register of commercial message consents; the system we build runs that record and the opt-out requests, but legal responsibility and assessment rest with the sending company and its adviser. On the WhatsApp side there is no free bulk sending: unless the customer has written to you, only pre-approved template messages can be sent, and those sends are chargeable. On the email side, if a new sending domain is to be used, volume is increased gradually, otherwise the messages land in the spam folder. Finally: automation does not make a bad offer good, it does not rescue a weak product page, and we do not promise a number for your conversion rate. What we do promise is contact made at the right moment and with consent.
Basket abandonment, interest after a first order and a drop in purchase frequency are measurable moments and can be tied to a flow. In these high-volume structures the gain comes not from sending more messages but from sending to the right person at the right moment and leaving everyone else in peace. That is why the unsubscribe rate counts as a measure of the flow's success and is monitored regularly.
In work where months pass between quote and order, the real issue is keeping the prospect warm. Someone who downloads a catalogue, returns to the pricing page or asks for a sample is scored; when the threshold is crossed they are handed over to a sales representative. That way the sales team deals with prospects who are showing movement rather than with irrelevant lists. The scoring rules are written together with the sales team, because they are the ones who know which behaviour really means something.
Warranty registration, scheduled maintenance dates, consumable life and campaign participation are events well suited to reminders. In these companies automation manages the ownership period more than the sale; it establishes regular, useful contact with the end user without damaging the dealer relationship and creates the opportunity to increase returns to service. The timing of the reminder is set by the real service life of the product, not by whatever suits the calendar.
Where communication is run by one or two people, every send turns into a separate project and continuity is lost. Lifecycle flows, once built, keep running in the background; instead of preparing sends, the team spends its time monitoring and improving the flows. When a new campaign idea comes up, it is added on top of the existing flow rather than started from scratch.
Flows start with an event, branch according to conditions and wait for set periods. If the customer buys or opens a support request in the meantime, the flow can be stopped; the same person is prevented from colliding across more than one flow. Flows are also stored as written rules, so that a year later it remains readable who received what and why. The audiences that may and may not enter a flow are defined as rules. With a test record, the whole flow is tried step by step before it goes live.
Email sending is set up on your own domain with SPF, DKIM and DMARC records. SMS runs through operator services and WhatsApp through the WhatsApp Business Platform with approved templates. The choice of channel is tied to a condition inside the flow: the more expensive channel only comes into play when it is genuinely needed. Send results, errors and bounced addresses are reported broken down by source. Protecting the reputation of the domain is an inseparable part of setting up the channel.
Flows with adjustable timing are built for visitors who leave a basket or view a product again and again. The contents of the basket are carried into the message, stock is checked and the flow is stopped for a product that has sold out. Whether a discount is added is managed as a rule; giving a discount for every abandoned basket quietly erodes the margin. How many steps the flow has and at what intervals it runs are tuned to your product range.
Welcome, post-first-order, product usage support, reorder time and win-back flows are designed around the stage the customer is at. The stages are set by the real usage rhythm of your product; on a product that is consumed over six months, a thirty-day reminder only irritates. Each flow is measured on its own, so a step that is not working can be switched off. Transitions between stages are logged as well.
Which behaviour signals how much interest is defined together: a visit to the pricing page, a catalogue download, a sample request, a reply to an email. When the score threshold is crossed the record drops into the CRM and is assigned to a representative. The handover is not one-way; when sales closes the record, the related flows stop accordingly. Scoring rules change over time, which is why the thresholds are left adjustable from the dashboard. The behaviour history goes to the representative along with the handed-over record.
Subject lines, send times, channels and the waiting periods in the flow are tested comparatively. Measurement does not stop at opens and clicks; how far the flow converts into an order, a quote or an appointment is tracked too. No result is declared before enough of a sample has been gathered; testing on small lists may not always be meaningful. The change being tested is recorded in writing and results are not written over one another.
The source and date of consent are on record; an opt-out request takes effect immediately across all flows. Frequency caps, quiet hours and a per-channel upper limit can be defined, so the same person does not receive messages from five different flows in the same week. Approval and refusal statuses on the İYS side are kept in sync. Send records are stored in an auditable form and exported when needed. Complaints and feedback are tracked in the same record.
We map out which message goes to whom today, with which tool and from which list. We then look at which events can be obtained from which system: orders, baskets, service records, site behaviour. Where there is no data there is no flow; this step sets the realistic limit of the flows that can be built. The consent status of the existing lists is reviewed in this step too, because a dirty list causes trouble on the flow's first day.
The stages a customer passes through from first contact to repeat order are mapped out; which message goes at each stage, through which channel and after how long a wait, is designed in writing. How many flows to start with is decided here; starting with three or four flows almost always produces a better result than starting with ten.
Event collection is connected, consent records and the opt-out list are brought into service, SPF, DKIM and DMARC are configured for the email domain, and SMS and WhatsApp templates are prepared and submitted for approval. Because template approval and domain warm-up depend on outside parties, separate time is set aside for them in the schedule.
We start by running a small number of flows with a limited audience. Send volume is increased gradually, deliverability and unsubscribe rates are monitored, and the copy and the timing are corrected. This period takes three to six weeks in most projects, and its purpose is not to achieve high results but to confirm that the flow works correctly.
Verified flows are multiplied and the scoring thresholds are tuned together with the sales team. How flows are edited, how a template is updated and how the report is read are passed on to your team and left in writing. Sending errors and blocked flows are monitored in production; a flow that has gone quiet does not go unnoticed. When a new product or a new channel is added, the flows are extended accordingly. The measurement report is reviewed together at regular intervals.
The off-the-shelf tool does the sending; the real bottleneck is the data. If basket abandonment, order status, service records and site behaviour do not flow into the tool regularly, what is left is hand-built lists again. What we do is usually not to throw the tool away but to connect it to your systems and put the flows on a data footing. If the tool's limits are not enough for your business, we build the sending layer ourselves as well; that decision is taken during discovery, by looking at the numbers.
The CDP establishes who is who: it unifies records from different systems under a single identity, carries consent status and generates segments. Marketing automation (MAP) runs what is done with those segments: it executes the message, the flow, the timing and the channel. The two can also be built separately; when they are built together, the flow runs for the right person the moment the segment changes. Because their setup and their costs differ, we describe them on separate pages.
The software runs the process and the record: it holds the source and date of consent, applies opt-out requests immediately across all flows, works in sync with the approval and refusal statuses on İYS, and stores send records in an auditable form. However, which message counts as commercial, which consent is valid and where responsibility lies under the regulations are matters of legal assessment and belong to the sending company. We write this distinction down at the outset, because arguing about it later proves expensive.
We do not give a figure, and we would advise you to treat any proposal that does with caution. The result depends on your product, your price, the size of your list and the share of it that is consented, and on the quality of the purchase flow on your site; most of that lies outside automation. What we can promise is measurability: how many people each flow reached, how many engaged and how many turned into orders is tracked from day one. If the result is weak, you will see that on the same dashboard.
No. On the WhatsApp Business Platform, unless the customer has written to you, only pre-approved template messages can be sent; templates go through approval, are divided into categories, and sends are chargeable. When the customer replies, a limited-time window opens in which the exchange can be freer. Solutions that send in bulk from a personal number can get the account closed; we do not recommend that route and we build the setup on the official platform.
This page describes the software side: the flow engine, the data connection, channel setup and measurement. Advertising management, SEO and content production are separate service headings of ours and are described on their own pages. The two feed each other: advertising and content bring new people in, automation keeps the relationship going with the people who arrive. If you wish, both sides are run by one team; or we build only the software and you manage the communication.
In a 30-minute discovery call we listen to what you need and tell you honestly whether custom development or an off-the-shelf product is the better answer.
We build a learning management system (LMS) that brings together, in one place, the training catalogue, who is required to take which course, who has completed it and when a certificate expires. When an audit asks for the record of your health and safety and quality training, it comes out on a single screen rather than after hours of searching.
Details →We build a system that ties the company goal to team and individual goals and bases the end-of-period review on records rather than memory. Goal management runs on OKR and individual assessment on a performance management system (PMS); we keep the two side by side on the same platform without mixing them up.
Details →We do not replace Logo, Mikro or Netsis (the accounting packages in common use in Türkiye). We build a finance layer that gathers current accounts, banks, cash, cheques and promissory notes, expenses and cash flow alongside your accounting package, and puts collection, payment and approval processes on rules. The record is still kept in accounting; visibility and process run in this layer.
Details →We move the budget out of Excel files and into a system that is versioned, has a named owner and is compared with actuals automatically. When the month closes, the variance report is not put together by hand; the reason for the variance is recorded, the forecast is updated and the management pack is produced from the same numbers. The goal is not more reports, but seeing where you stand against target in a single figure nobody disputes.
Details →On legacy systems with no API, we build software robots that take over the repetitive steps a person performs on screen. The robot runs under its own user account and on a schedule; when something unexpected happens it does not guess, it stops and hands the job over to a person along with a screenshot.
Details →We build an inventory of your processes, map them as they really are using BPMN 2.0, and define an owner and a measure for each one. We then run them on a process engine (BPMS) and measure cycle time, waiting points and bottlenecks. If a single flow is all you need, we will say that plainly too.
Details →
